- U - t - H I
.-Az~ ~ ~ ~~~~
- F _ * _ r t t ,, , _ _ _ b _ _ IS
_ 4: >- ---- g '=5i
-- S -
~~~~ I : '
g 5-r ~ w1 - a x ~ -
Other EDI Development Studies
Does Privatization Deliver? Highlights fiom a
World Bank Conference
Edited by Ahmed Galal and Mary Shirley
ISBN 0-8213-2589-2
7The Adaptive Economy: Adjustment Policies in Small,
Low-Income Countries
Tony Killick ISBN 0-8213-2125-0
Financial Regulation: Changing the Rules of the Game
Edited by Dimitri Vittas ISBN 0-8213-21234
The Distribution of Income and Wealth in Korea
Danny Leipziger and others ISBN 0-8213-2124-2
Public Enterprise Reform: The Lessons of Experience
Mary Shirley and John Nellis ISBN 0-8213-1811-X
Also available in French and Spanish
Privatization and Control of State-Owned Enterprises
Edited by Ravi Ramamurti and Raymond Vernon
ISBN 0-8213-1863-2
Finance at the Fronder: Debt Capacity and the Role of Credit
in the Private Economy
J D. Von Pischke ISBN 0-8213-1818-7
EDI DEVELOPMENT STUDEES
Labor Markets
in an Era of Adjustme'nt
Volume 2
Case Studies
Edited by
Susan Horton
Ravi Kanbur
Dipak azumdar
The World Bardc
Washington, D. C.
Copyright 0 1994
The International Bank for Reconstruction
and Development / THE WORLD BANK
1818 H Street, N.W.
Washington, D.C. 20433, U.S.A.
AU rights reserved
Manufactured in the United States of America
First printing July 1994
TheEconomic Development Institute (EDI) was establishedby theWorldBank in 1955 to
train officials concemed with development planning policymaking, investment analysis,
and project implementation in member developing countries. At presentthe substance of the
EDI's work cmphasizes macroeconomic and sectoral economic policy analysis. Through a
variety ofcourses,seninars, andworkshops,mostofwhich aregiven overseasin cooperation
with local institutions, the EDI seeks to sharpen analytical skdlls used in policy analysis and
to broaden understanding of the experience of individual countries with economic develop-
menL Although the EDI's publications are designed to support its training activities, many
are of interest to a much broader audience. EDI materials, including any findings, interpre-
tations, and conclusions, are entirely those of the authors andshiould not be attnbuted in any
manner to the World Bank, to its affiliated organizations, or to members of its Board of
Executive Diectors or the countries they represenL
Because ofthe informality of thisseries and to makethe publicationavailable with the least
possible delay, the manuscripthas not been edited as fullv as would be the case with a more
formal document, and the World Bank accepts no responsibility for errs. Some sources
cited in this book may be informal documents that are not readily available.
The material in this publication is copyrightedL Requests for permission to reproduce
portionsofitshouldbesenttotheOfficeofthe Publisheratthe addressshown inthecopyright
notice above. The World Bank encourages dissemination of its work and will normally give
permission promptly and, when the reproduction is for noncommercial purposes, without
asling a fee. Permission to copy portions forclassroom use is granted thmugh the Copyright
Clearance Centerlnc, Suite 910, RosewoodDdve,Danvers,MassachusettsOl923, U.S. A.
The bacdist of publications by the World Bank is shown in the annual Inder of ublca-
dions, which is available from Distnbution Unit, Office of the Publisher, The World Bankl
1818 H Stret, N.W., Washington, D.C. 20433, US.A, or from Publications, Banque
mondiale, 66, avenue d'Idna, 75116 Paris, France-
Susan Horton is an associate professor of cconomics at the University of Toronto; Ravi
KanburistheWorldBank'sresidentrepresernative in Ghana;andDipakMazumdaris alabor
markets specialist in the World Bank's Office of the Chief Economist, Africa
Library of Congress Cataloging i-Publication Data
Horton, Susan.
Labor markets in an era of adjustment / Susan Horton, Ravi Kanbur,
Dipak Mazumdar.
p. cam-(EDI development studies)
Includes bibliographical references.
Contents: v. 1. Issues papers-v. 2. Case studies.
ISBN 0-8213-2680-5 (v-l).-ISBN 08213-2681-3 (v. 2)
1. Labor market-Developing countries-Congresses. 2. Structural
adjustment (Economic policy)-Dteloping countries-Congresses-
3 Labormarket-Developingcountries-Casestudies-Congresss
4. Strucural adjustrnent (Economic policy)-Developing countries-
Case studies-Congresses. I.Kanbur, Skl Ravi U. Mazumdar,
Dipak, 1932- .HIl ritle. IV. Series.
1D5852
331.12f09172'6-dc2O 93-34978
CIP
CONTENTS
VOLUME 2: CASE STUDIES
Foreword v
Acknowledgments vii
Preface ix
Abbreviations and Acronyms iii
1. Labor Markets in an Era of Adjustment An Overview 1
SusanHorton, RaviKanbur,an DipakMazumdar
2. Argentina 61
Luds A. Riveros and Carlos E. Sdnchez
3. Bolivia 99
Susan Hof-on
4. Brazil 143
MA Louise Fox, Edward Amadeo, and Jose Marcto Camargo
5. Chile 169
Luis A. Riveros
6. CostaRica 217
T. H. Gindling and Albert Berry
7. COte d'Ivoire 259
Richard BtundeL4 Christopher Heady, and Rohizton Medhora
8. Egypt 317
Ragui Assaad and Simon Commander
9. Ghana 357
P. Beaudry and N. K. Sowa
10. Kenya 405
William J. Mime and Monica Neiz,err
Hii
iv Contents
11. Malaysia 459
DipakMazumdar
12. The Republic of Korea 535
DipakMammdar
13. Thailand 585
Chalongphob Sussangkarn
Index 613
VOLUME 1. ISSUES PAPERS
Labor Markets in an Era of Adjustment An Overview
Susan Horton, Ri Kanbur, and DipakYMazumdar
1. Recent Developments in the Developed-Country Literature on
Labor Markets and the Implications for Developing Countries
.ea i-PazdAzam
2. Labor Market Distortions and Structural Adjustment
in Developing Countries
AI4andra CaoEdnds and Sebastan Edwards
3. The Poverty Effects of Adjustment with Labor Market
Imperfections
Tony Addison and Lionel Demery
4. Wage Indexation, Adjustment, and Inlaton
Micheelif. Devereux
5. The Long-Run Consequences of Short-Run Stabilization
Policy
Edwwrd F B<Rle
6. Gender Aspects of Labor Allocation during Structural
Adjustment
Paul Collier, A. C- Edwards, J. Robr, and Kalpana.Bardhan
7. Organized Labor, Politics, and Labor Market Flexibility in
Developing Countries
Joan M. Nelson
FOREWORD
This two-volume study is the result of a series of five conferences
organized by the Economic Development Institute of the World Bank
in collaboration with the University of Toronto and Warwick
University and supported by the Overseas Development Administra-
tion of the United Kinadom and the grovemments of Canada and
Ireland.
It comprises the research papers prtsented at the conferences and
revised in light of comments and suggestions by the participants as
well as by other experts in the field. Various chapters have been pre-
sented in seminars for World Bank staff and at the annual meetings of
the American Economic Association.
Armnon Golan, Director
Economic Development Institute
ACKNOWLEDGMENTS
This book is the result of a large research project that was initiated
by Ravi Kanbur and Dipak Mazumdar early in 1988. They were
subsequently joined by Susan Horton, who had at that time begun to
do work on a similar theme for Bolivia.
The research was facilitated by a series of five conferences: three at
Warwick University and two at the University of Toronto. The editors
would like to thank the universities for their support and the following
organizations for providing funding: the Overseas Development
Administration of the United Kingdom and the governments of
Canada and of Ireland for their support through trust funds
established at the World Bank.
Many people participated in the conferences. Some presented
papers, others provided useful comments, and all contributed to the-
progress of the research. In addition to the authors, these included
(with their affiliation at the time): A. Berry, D. Benjamin, M. Faig, and
Y. Kotowitz (University of Toronto); P. Brixen, L. Haddad, M.
Johnson, J. MacKinnon, A. McKay, S. Nath, G. Pyatt, and J. Round
(Warwick Unriversity); A. Chhibber, J. Daniel, L. Fox, C. Grootaert, T.
King, J. Newman, A- van Ada ms, and M. Walton (World Bank); A.
Atsain (University of Abidjan), C. Bean (London School of
Economics), I. Bradley (ESRI, Dublin), G. Fields and E. Thorbecke
(Cornell University), N. Gregory (ODA), T. Besley, P. Horsnell, S.
Kheng-Kok, J. Knight, C. Yves, and A. Zegeye (Oxford University), S.
Morley (Vanderbilt University), N. N'geno (University of Nairobi), A.
Plourde (University of Ottawa), B. Renison (USAID), G. Rodgers
(ILO), B. Salome and D. Tumbam (OECD), J. Svejnaar (University of
Pittsburgh), R van der Hoeven (UNICEF), K. Yao (CIRES, Abidjan),
and Z. A. Yusof (Malaysian Institute of Economic Research).
vii
PREFACE
Our interest in undertaking a project on structural adjustment and
labor markets in developing countries arose from our perception of a
gap in the existing literature. A good deal of work had been done on
structural adjustment and poverty, but without work on the labor mar-
ket little was known about how the effects of structural adjustment
were transmitted to the poor, most of whom depend heavily on labor
market earnings. At the same time some policymakers and interna-
tional institutions seem to believe that labor market rigidities are an
obstacle to structural adjustment, and several developing countries
have implemented rather draconian policies to regulate their labor
markets. However few empirical studies of developing countries exist
to justify such policies.
At the start of the project we invited researchers to tell us what the
existing theory and studies from developed countries suggested about
adjustment and labor markets. We then commissioned a series of the-
ory papers to extend the literature to developing countries, looking at
topics such as structural adjustment and poverty, the effects on women,
the political economy aspects, the long-run effects of adjustment, and
so on. At the same time we began twelve country studies to examine
the effects of adjustment on labor markets. The country studies took
longer and, as a result, were enriched by -insights from the theory pa-
pers, which were completed earlier.
In choosing countries to study, we wanted to have as wide a geo-
graphic coverage as possible. Data availability proved to be one limi-
tation. We felt that it was essential to have access to household labor-
force survey data over time (corresponding to the adjustment period)
for the countries concerned. Without these data, the effects on women,
on income distribution, on real wages, and on unemployment could
not easily be studied. Existing international compilations (primarily
the ILO Yearbook of Labour Statistics) have somewhat uneven cover-
ix
x Preface
age since not all countries report and the data they publish are for a
mix of household surveys and establishment surveys. Data constraints
were most serious for Africa, where very few countries had repeated
labor force surveys at different times. The final sample of countries
included five from Latin America, three from Asia, one from North
Africa and the Middle East, and three from Sub-Saharan Africa.
At the outset we were not aware of different "patterns" of adjust-
ment. In the course of the research, however, the countries fell into
four groups of three. One group (the Republic of Korea, Malaysia,
and Thailand) reflected what might be termed an "East Asian"
pattem of adjustment, with short, sharp recessions and a resumption of
fast growth of GDP (more than 5 percent per year). Another group
(Bolivia, Chile, and Ghana) undertook "severe" adjustment. These
economies had had more serious problems in the 1970s and 1980s,
experiencing either sharp f-alls in GDP in some years, or prolonged
stagnation or decline. Real wages fell more than 50 percent in the
course of adjustment, and in the case of Chile there was high
unemployment (more than 25 percent at the worst points). A third
group- of countries (Brazil, Costa Rica, and Kenya) might be described
as having undergone "partial" adjustment, where adjustment was less
painful than in the Asian case but the resumption of growth was also
less strong. The final group (Argentina, Cote d'lvoire, and Egypt)
represent "frustrated" adjustment, in that adjustment was delayed for
a number of reasons-for example, stop-go cycles in Argentina.
difficulties in devaluing because of membership in the franc zone for
COte d'Ivoire, and the later onset of problems for Egypt, as an oil
exporter.
The research suggested that labor markets m developing countries
were in fact working quite well to permit structural adjustment. Three
important conclusions were reached: real wages were more flexible
than generally supposed, which would support adjustment; labor real-
location across sectors has been more or less in the desired direction;
and labor market institutions such as unions and minimum wages, of-
ten argued to be an impediment to adjustment, have more subtle ef-
fects on the workings of labor market-a finding that is worthy of
further study.
Preface xi
As occurs in all research projects, we discovered other gaps in the
literature and topics that seemed worthy of attention but that did not
fit within the scope of the existing effort. Much more could be learned
about labor markets in developing countries by constructing time se-
ries from regular labor force surveys. In some countries, particularly
the richer countries in Asia and Latin America, these series have al-
ready been put together, but the same is not true for the poorer coun-
tries. Time series for key variables are very important since fluctua-
tions and cycles in the economy render one-year "snapshots" derived
from a single survey quite misleading. A great deal of work also
remains to be done in such areas as quantifying changes in income
distribution over time, analyzing the effect of structural adjustment on
women, and exploring the effects of labor market institutions in
developing countries.
ABBREVIATIONS AND ACRONYMS
BCEAC Banque Centrale des Etats de 1'Afrique de l'Ouest
CACM Central American Common Market
CAPMAS Central Agency for Public Mobilization and Statistics
(Egypt)
CBS Central Bureau of Statistics (Kenya)
COB Confederaci6n Obrera de Bolivia
CPI consumer price index
c.i.f. cost, insurance, freight (term for describing imports)
EPB Economic Planning Board (Kenya)
f.o.b. free on board
GDP gross domestic product
GLSS Ghana Living Standards Survey
GNP gross national product
IEERAL Institute of Economic Studies on Argentina
IMF International Monetary Fund
INE Instituto Nacional de Estadistica (National Bureau of
Statistics) (Bolivia, Chile)
LSMS Living Standards Measurement Survey
NBER National Bureau of Economic Research
NEP New Economic Policy (Bolivia, Malaysia)
NEER nominal effective exchange rate
OECD Organization for Economic Cooperation and
Development
RDER real domestic exchange rate
REER real effectite exchange rate
UMOA Western Africa Monetary Union
UNICEF United Nations Children's Fund
USAID United States Agency for International Development
WPT wholesale price index
xiii
LABOR MARKETS IN-AN ERA OF ADuSTMENT:
AN OVERVIEW
Susan Horton
Ravi Kanbur
Dipak Mazumdar
Issues and Country Studies
Labor markets play a central role in determining the
macroeconomic success of stabilization and adjustment policies and in
mediating the impact of these policies on the population's standards
of living, in particular the poor. The 7 issues papers and 12 country
studies in these volumes examine the different aspects of this
interaction between labor markets and adjustment. The object of this
chapter is to provide an overview and to draw out general conclusions,
policy lessons, and areas for further research.
Issues
To start with, let us define what we mean by adjustment and by
labor markets. Under adjustment we include both stabilization and
structural adjustment. Following convention, by stabilization we mean
the reduction of national expenditure to bring it in line with national
income or output, usually following external shocks. By structural
adjustment we mean attempts to increase national income or output
through more efficient use of resources. Of course, a myriad of macro
policy instruments, such as exchange rates, monetary policy, and fiscal
policy, are available to achieve these goals, which may sometimes be
stated in terms of inflation, balance of payments, and growth targets.
2 SusanHorton, RaviKanbur, andDipakMazumdar
The links between instruments and targets, however, almost always
touch on labor markets and their operation.
A labor market is a mechanism for matching the supply and
demand of the factor of prorduction labor, through the terms of the
contract between buyer and seller. As many different types of labor
exist, differentiated by skill, location, gender, and so on, many
different labor markets exist, but these markets are linked with each
other because the conditions in one can influence the workings of
another. The.system of interlinked individual labor markets in a
country can be called the labor market. The labor market is itself
linked to other markets in the economy: it influences their workings
and is in turn influenced by them.
The terms of the contract between buyers and sellers in a labor
market can vary, from wage payment in markets for unskilled labor to
complex packages of remuneration and benefits over time in markets
for skilled labor. The markets can vary in structure, from many buyers
and many sellers to small groups of buyers and sellers. Some analysts
talk of internal labor markets within large firms. National policy and
regulation affect the workings of the labor market, and the labor
market in turn produces institutions that become important in setting
national policy. While any individual labor market may be small, the
outcomes in the labor market as a whole can influence
macroeconomic conditions in the economy. Since the outcomes
determine the payment to labor, they also affect the distribution of
income in the economy.
Policymakers are often interested in knowing whether a country's
labor market is "working well," and what can be done to "improve"
its workings. But what does it mean to say that the labor market is
working well? As always, a general characterization is difficult. The
most general statement we can make is in the context of a competitive
general equilibrium model of the economy. In this stylized setting, we
know that if every other market operates in the manner of classical
competitive markets, then if the labor market also operates in this
manner the economy will achieve a Pareto optimal outcome- Thus, in
this framework and under these conditions what is meant by the labor
market working well is dear: it is that the labor market is working like
a classical competitive market where price adjusts to equate supply and
Labor Markets in an Era ofAdjustment:An Overview 3
demand. In most practical settings, this is indeed the test that is
applied, and discussion of policy and regulation is highly colored by
the use of this benchmark. However, the slightest reflection should
reveal how fragile this benchmark is, and how severe and unrealistic
are the conditions under which it is viable, as in reality there is no
guarantee that other markets are themselves working like classical
competitive markets, and a Pareto optimal outcome may not satisfy
distributional criteria for evaluating the economic system as a whole.
Since the search for any general characterization is likely to prove
futile, the best way to approach the analysis is with a more specific
notion of what is being asked of the labor market, given the structure
of other markets in the economic system and the particular economic
policy problem under consideration. The particular policy problems
we focus on here are those of stabilization and structural adjustment
through the use of macroeconomic policy instruments.
We discuss stabilization first. The role of the labor market here is to
ensure that the reduction in national expenditure takes place without
inducing a substantial reduction in national output. The basic
mechanisms are well known. As national expenditure falls there will be
downward pressure on output prices if output markets behave like
classical competitive markets. This downward pressure on output
prices will lead to cutbacks in production, and hence in the demand
for labor. If the price of labor falls in response to this reduced
demand, then this reduction in cost will help maintain the level of
production. If the price of labor falls sufficiently in relation to the
original fall in output prices, under certain conditions there need be
no fall in total output at all. To the extent that the real wage does not
fall this far, total output will be lower than it otherwise would be and,
because of unemployment, the wage bill will be distributed more
unequally than it otherwise would be.
In this framework, therefore, the test for whether the labor market
was working well would focus on whether the real wage - fell
sufficiently to maintain employment and output in the face of a
reduction in total national expenditures (vol. 1, chapters 1 and 3).
Clearly, labor market institutions are relevant here. If the labor market
is unionized, and the union cares more about the real wage of
employed members than about the number of the unemployed, then
LaborMarkets in an Era ofAdritstnmenLAn Overview 3
demand. In most practical settings, this is indeed the test that is
applied, and discussion of policy and regulation is highly colored by
the use of this benchmark. However, the slightest reflection should
reveal how fragile this benchmark is, and how severe and unrealistic
are the conditions under which it is viable, as in reality there is no
guarantee that other markets are themselves working like classical
competitive markets, and a Pareto optimal outcome may not satisfy
distributional criteria for evaluating the economic system as a whole.
Since the search for any general characterization is likely to prove
futile, the best way to approach the analysis is with a more specific
notion of what is being asked of the labor market, given the structure
of other markets in the economic svstem and the particular economic
policy problem under consideration. The particular policy problems
we focus on here are those of stabilization and structural adjustment
through the use of macroeconomic policy instruments.
We discuss stabilization frst The role of the labor market here is to
ensure that the reduction in national expenditure takes place without
inducing a substantial reduction in national output. Th-e basic
mechanisms are well known. As national expenditure falls there will be
downward pressure on output prices if output markets behave like
classical competitive markets. This downward pressure on output
prices will lead to cutbacks in production, and hence in the demand
for labor. If the price of labor falls in response to this reduced
demand, then this reduction in cost will help maintain the level of
production. If the price of labor falls sufficiently in relation to the
original fall in output prices, under certain conditions there need be
no fall in total output at all. To the extent that the real wage does not
fall this far, total output will be lower than it otherwise would be and,
because of unemployment, the wage bill will be distributed more
unequllay than it otherwise would be.
In this framework, therefore, the test for whether the labor market
was working well would focus on whether the real wage fell
sufficiently to maintain employment and output in the face of a
reduction in total national expenditures (vol. 1, chapters 1 and 3).
Clearly, labor market institutions are relevant here. If the labor market
is unionized, and the union cares more about the real wage of
employed members than about the number of the unemployed, then
LaborMarkets in an Era ofAdjusment:An Overview 5
but at its heart is a shift in the composition of national output towrd
the production of exportables and import-competing output
(tradables) through the use of relative price instruments such as the
exchange rate. Clearly, such a shif-t in the pattern of production
requires a corresponding shift in factors of production toward certain
sectors, and it is the labor market through which the sectoral
composition of labor use is altered. The general issue to which this
gives rise is the nature and extent of reallocations between different
labor markets. Essentially, what is required is for labor to flow to the
production of tradables, that is, to flow to those labor markets that
serve the production of tradables. This may require reallocation across
firms in the same area, across the formal/informal or coveredt
uncovered divide, or across regions.
In principle, this reallocation could tak-e place through a number of
mechanisms, but economic analysis focuses on the role of temporary
wage differentials in attracting labor to markets where demand is high.
Note, however, that the very reallocation to which the differentials give
rise wil tend to mitigate the differentials. If the wage differentials are
constrained between limits because of institutional factors, standard
results on the impact of changes on the relative output price in the
composition of employment and output will not occur (vol. 1, chapter
2). The same would happen if gender differences led to significant
misallocation of labor (voL 1, chapter 6). Movements in relative wages
can therefore be a deceptive test of whether the labor market is
working well. Concentrating directly on the nature and extent of
reallocation between output sectors is far better- The faster this
reallocation, the faster the desired adjustment in national output
However, labor is only one of the factors of production, and one must
take care before one pronounces that because labor reallocation has
not taken place, the labor market is not working well. If markets for
complementary inputs (for example, credit) are not playing their role,
the labor market may be hampered in achie.ving the desired
reallocation of labor, and therefore of output
Whatever the role of the labor market in achieving the
macroeconomic objectives of stabilization and structural adjustment,
how the labor market responds to macroeconomic instruments will
certainly determine the distribution of income in the economy. At the
6 Susan Horron, RaviKanbur, andDpakMaodar
simplest level, if stabilization necessitates a period of high
unemployment because of downwardly rigid wages, then inequality
will increase, and perhaps poverty wfll increase more than if real wages
had fallen sufficiently to maintain employment The extent and nture
of the reallocation of labor across sectois will also influence the
distnrbution of income. I, for example, in the initial situation the poor
are concentrated in sectors producing tradables, then the increase in
wages necessary to attract labor to those sectors will reduce poverty on
this count in the short to medium run, although what happens in the
long ran depends on how markets for other factors operate. One can
conduct a systematic analysis of the impact of adjustment on poverty
in the presence of a variety of labor market structures (vol. 1, chapter
3). To the extent that the labor market is segmented along gender
lines, the distribution of income will also be affected (vol 1, chapter
6)-
Economists now realize that stabilization and structural adjustment
policies, although designed to achieve macroeconomic baIance in the
short and medium term, -will have long-run consequences through
their impact on investment. There is a similar impact on human capital
investmenet To the extent that investment in hbuman capital is affected
by changes in relative wages, short-run policies (via their effects on
labor markets) will also have long-un consequences (voL 1, chapter
5).
Cozntry Studies
The interactions between labor markets and adjustment thus throw
up a number of interesting issues and questions. The answers to many
of these questions will be context and country specific. The issue
papers in this symposium take up specific conceptual matters and
develop or review the analysis on areas highlighted in this section. The
country studies, however, are at the heart of this symposium, since they
insert reality into the conceptual framework. Each country study
author was asked first to give a brief account of the adjustment
process: the nature of the shock, the policy responses, and the
macroeconomic outcomes. The authors were then asked to give an
account of the relevant characteristics of the labor market, for
example, labor force composition, wage differentials, and wage setting
Labor Markets in an Era ofAdjusiment An Ovewiew 7
mechanismsm Armed with these two accounts, the studies then assess
the labor market's role in the adjustment process, paying due attention
to institutional features. With these basics, the authors were also asked
to evaluate the impact of adjustment, as mediated by the labor market,
on poverty and on women. Fnally, they were invited to consider the
long-run consequences of labor market adjustment.
This symposium contains 12 country studies: four from Africa (of
which three -are from Sub-Saharan Africa), five from Latin America,
and three from Asia The countnies span a range of different income
levels (using the World Bank's classification), ranging from two low-
income countries (Ghana and Kenya), seven lower-middle-income
countries (Bolivia, Chile, Costa Rica, Cote d'Ivoire, Egypt, Malaysia,
and Thailand), and three upper-middle-income countries (Argentina,
Brazil, and the Republic of Korea). The countries also span a range of
adjustment experience.
Not all the country stUdies address all the issues, sometimes because
particular issues were iLmportant in particular cCuutries but not in
others, and sometimes due to data limitations. Almost all the Latin
countries, many of those in Southeast Asia, and some in North Africa
have periodic labor force surveys, although the data from these vary in
terms of accessibility and amount of previous analysis. However, few
labor force surveys cover the rural sector (exceptions are Kenya,
although the survey is infrequent, and Thailand for occasional years).
To the authors' knowledge, Sub-Saharan Africa has no regular labor
force surveys, and Kenya is the only country with comparable
household surveys for a year in the 1970s and a year in the 1980s.
The two other Sub-Saharan African studies rely on cross-sectional
data for the 1980s from the World Bank Living Standards Surveys.
Data from these surveys have the advantage of covering rural areas,
and with some ingenuity (for example, using information on length of
tenure in current job or length of residence in current location) can be
used to shed light on some changes that have occurred in the labor
market over time.
In addition to labor force surveys, many countries have other data
from employment and earnings surveys, collected usually at the
establishment level. These series tend to cover mainly the formal
sector and not a representative sample of households, and can
8 Susan Horton, RaviKanbur andDipakMamdar
sometimes be misleading, especially during a period of substantial
sectoral shifts and declines in formal sector employment (see, for
example, Lavy and Newman 1989 on the CMte d'Ivoire). The Bolivian
case study similarly poiuts out discrepancies between the real wage
and the unemployment series from household surveys as compared to
establishment surveys.
The 12 country studies pull* together a wealth of information. This
is particularly useful given the dearth of centralized international
reporting of labor data. The International Labour Organisation's
(ILO's) regional subdivisions do collate and report information within
the respective regions (Latin America, Africa, and Asia). However, the
ILO yearbook, for example, the basis for Johnson's 1986 work, is
spotty in terms of country coverage and seems to rely on
establishment survey results rather than the (arguably) more reliable
household survey data. The country studies here contain not only
whatever aggregate data are available, but in many cases aIso contain
ongmal econometric analyses (both micro and macro) of the data.
Although the workings of the labor market have been well studied
for some countries, there are relatively few comparative studies of the
effects of the crisis of the late 1970s and the 1980s. FalIon and
Riveros (1988), ILO (1987), and Johnson (1986) compare a range of
countries, Ghai (1987) and JASPA (1988) examine African countries,
and some work on Latin America is available, for example. by
Tokman (1984) and Riveros (1989), and by the Programa Regional
del Empleo para America Latina y -el Canbe (PREALC). Investigators
have also examined the public sector labor force (Lindauer and others
1988). The present set of studies tries to cover a broad range of
countries, including some that had not been studied much previously.
Although structural adjustment is by no means complete in these
countries, enough years of data have been accumulated since the onset
of crisis and adjustment, that it may be timely to assess experience so
far. As such, the country studies may provide a valuable basis for
generalization.
Varieties of Adjustment Experience
Before drawing conclusions on the role of labor markets in
adjustient on the basis of our case studies, it is useful to consider the
LaborMarkes in an Era ofAdjtmenr An Oven-iew 9
nature of the adjustment that has taken place in these countries.
Although quantifying the type or success of adjustment is hard, we
suggest that our 12 countries fall into four groups of three. One group
consists of the three Asian countries in the sample, which have by and
large had short and successful adjustments (based on previously
relatively outward-oriented economies). A second group consists of
three countries that had previously had strongly inward-oriented
economies that undertook severe and painful adjustment (Bolivia,
Chile, and Ghana). The remaining countries all undertook less severe
adjustments than the second group, but with less immediate success
than the Asian group. These six countries form somewhat of a
continuum, but three of them (13razil, Costa Rica, and Kenya) had
moderate success in adjusting without requiring major policy reversals,
and the last three (Argentina, Cote d'Ivoire, and Egypt) had somewhat
less success (in the case of Egypt as an oil exporter, efforts to adjust
began only very late in the time period under study).
Tables 11 and 1.2 summarize information on two key economic
variables: GDP growth rates (the most frequently used indicator of
economic performance) and real effective exchange rate (one possible
indicator of relative prices key to the adjustment process). As table 1.1
shows, the Asian countries have had occasional less successfil years,
but in general exlhbit growth rates of 5 percent per annum or greater,
and no years of negative growth. Of the "severe adjustment"
countries, Chile and Ghana exhibit economic problems dating back to
the 1970s, with large negative growth of GDP in some years, but since
1983 each country has grown at close to or more than 5 percent in
three of the following years. Bolivia (the other country in the group)
encountered economic problems later (bolstered by hydrocarbon
exports in the 1970s), and experienced the longest span without
positive growth of all the sample countries. Economic recovery there
remains weak. Brazil, Costa Rica, and Kenya (characterized here as
"moderate adjustment" countries) appear to resume reasonable
growth rates of GDP after the worst years (around 1980-82), although
their year-to-year growth rates following adjustment are more variable
than those of the Asian countries.The remaining three cases
(Argentina, Cote d'Ivoire, and Egypt, characterized here as "less
successful adjustment" countries) exhibit rather heterogeneous
Table 1.1 Growth Rates of GDP, 1970/71-1986/87
(constant prices)
Counry 1970-71 1971-72 1972.7.1 197.174 1974 75 1975.76 1976h77 i07*?N 1978.79 1979.80 198B8 198142 19824. 198344 19844$ 198586 18-87
Argentina 3,4 1.9 3.2 6.3 -0.7 -0.2 6.4 -3.2 7.0 1.5 -6.7 -4.9 3.0 2.6 -4.5 5.5 2.0
Bolivia 4.9 5,8 6.7 5.1 6.6 6.1 4,2 3.4 0.0 -0.6 0.9 -4.9 -6.5 -0.3 -0.2 -2.9 2.2
Brazil 12.3 10.9 13.5 9.7 9.9 9.7 2.9 4.9 6.8 9,3 -4.4 0.6 -3.5 5.1 8.3 7.6 3.6
Chile 9.0 -1.2 -5.6 0.1 -12.9 3.5 9.9 8.3 7.8 5.5 -14.1 -0.1 6.3 2.4 5.7 5.7 n.a.
Cosla Rica 6.8 8.2 7,7 5.5 2.1 5.5 8,9 6.3 4.9 0.8 -2.3 -7.3 2.9 8.0 0.7 5.5 5.4
C6te d'lvoire n.a . n n.a. vi.a. n.a. 12.0 4,7 9.9 5.2 6,3 1.4 3.0 0,0 -8.9 n.a. n.a. n.a.
Egypt n.a. n.a. 0.8 2.7 9.1 15.3 ,13.5 5.9 6.2 10.3 3.8 10.1 7.6 6.2 6.7 2.7 2.5
}haina 5.6 -2.5 15.3 3.4 -12.9 3,5 2.3 8.5 -3.2 0,0 -.1.8 -7.2 0.2 2.6 S.1 5.2 4.8
Rcp. of Kcnya 6.9 9.5 6.8 1.5 3,4 7.0 9.4 9.0 3.8 5,6 3.7 0.6 2.7 2,0 3.8 5.2 5.8
Korea 9.2 5.9 5.4 14.4 7.9 6.5 13.2 10.9 9.7 7.4 9.8 6.7 7.3 11.8 9.4 6.9 12.4
Malaysia 7.1 9.4 .11.7 8.3 0.8 11,6 7,8 6.7 9.3 7.4 6.9 5.9 6.3 7.8 9.9 1.2 5.2
Thailand n.a. 5.0 4.1 9.8 4.8 9,4 9.9 10.4 5!3 4.8 6.3 4.1 7.3 7.1 3.5 5.0 7.1
n.a. = not available
Soaurces: Caiculalcd fTom IMP, Inertinatlontat F'Iniancial Statistics (various ycars) except Bolivia data from country study,
Table 1.2 Real Effective Exchange Rate, 197686
(index, 1980 = 100)
Country 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986
Argentina ln., n.a. 54.5 76.7 100.0 91.1 50.6 42.7 49.7 44.0 44.1
Bolivia n.a. n.a. 87.3 91.6 100.0 125.9 136.6 125.4. 162.6 279.7 82.2
Brazil n.a. n.u. 122.8 112.5 100.0 121.5 128.4 104,2 104.2 100.1 94.4
Chile 93.7 102.1 85.2 86.1 100.0 118.0 106.7 86.8 85.3 68.8 58.2
Costa Rica 91.4 90.0 86.6 90.9 100.0 63.5 72.5 83.4 81.9 80.9 72.7
CCte d'lvoire n.a. n.a. 89.1 98.0 100.0 85.7 78.2 75.2 72.0 72.2 84.5
Egypt na.. n.o. 114.1 92.6 100.0 106.0 118.3 133.7 156.0 164,0 156.4
Ghana n.a. n.e. t 96.8 76.5 100.0 222.4 278.1 186,9 72.2 52.S 30.2
Kenya 94,2 97.3 104,2 101.0 100.0 96.7 100.3 95.0 101.7 100.3 87.0
Korea, Rep. of n.a. n.a. 97.6 107.4 100.0 104.4 106.9 102.7 101.3 95.5 80.6
Malaysia 106.5 105.9 101.4 105.8 100.0 100.4 106.7 111.8 116.1 110.3 92.6
Thailand n.a. n.a. 91.2 92.4 100.0 102.8 105.8 108.6 107.2 95.3 85.0
n.a, = not avallable
Note: Incrcasc implics apprcciation.
Sources; Calculated from IMF data.
12 Susan Horton, Ravi Kanbur, andDipak-Mazumdar
behavior. Argentina has continual stop-and-go cycles dating back to
at least 1974. C6te d'Ivoire encountered problems in the 1980s after
successful growth in the 1970s, but its ability to adjust has been
limited by its membership in the West African Monetary Union.
Finally, Egypt, as an oil exporter only, began to experience a growth
slowdown after 1985.
As the time pattern of growth rates in table 1.1 show, the timing of
adjustment was somewhat different in the various countries. Figure 1.1
shows the* sequence of events described in the country studies. The
years 1978 and 1982 were obviously watershed years (corresponding
to the second oil price shock and start of the rise in real interest rates
in one case, and to the onset of the debt crisis as signaled by inability
to pay in Brazil and Mexico in the other). Some of the variation
depends on price collapses in different commodity markets (coffee,
cocoa, and tin affected different countries in the sample), as well as
gooa or -bad harvests and weather, particularly for the African
countries. At least three countries had begun adjustment in the mid
1970s (Brazil, Chile, and Kenya), and Argentina had also made some
efforts in this direction. (Of course, countries such as Korea and Brazil
had made structural adjustments earlier still, changing their trade
regimes.) After 1982 all the countries in the sample undertook some
form of stabilization and/or structural adjustment.
Quantifying adjustment policy efforts or their success is somewhat
difficult. The real effective exchange rate (REER) may provide some
useful information, insofar as structural adjustment attempts to change
relative prices, and the exchange rate is a key price. However, one
encounters some problems in interpreting these data. First, some
countries may experience policy outcomes that differ from their
intentions (for example, C8te d'Ivoire recently tried to mimic a
devaluation, but due to changes in other currencies the value of C8te
d'Ivoire's currency actually appreciated). Second, there is no
benchmark as to what the equilibrium real exchange rate should be.
Some countries therefore appear to have succeeded in deep currency
depreciations, but from previously highly distorted rates, whereas
others appear to have been less successful, but because the previous
rate was less distorted.
Figure 1.1 Timing of Adjustment Efforts
Argentina deregulation of economy under pre-election series of failed SAs
milniry ease-up
Bolivia reasonable growth lending ends failed SA stabilizadon/ weak
SA growth
Bmzil SA under military stabillzadon stabilization crisis
Chile deregulation/ stabilization policies financial stabilization and
SA by military reversed, crisis export-led recovery
Costa Rica crisis onset cfforts to stabili- rnajorSA
spend way out of crisis zatdontSA
CBtc d'Ivoire coffee boom ends problemr stabilization furLber SA efforts hampered
begin lending dries up by exchange rate
Egypt good growth (oil windfall, aid remnittances) unsuccessful SA
Ghana long-run problems drought stabili- SA takes bold
zaiaon/SA
Kenya structural adjustment began devaluation drought
end of coffee boom trade policy shift
Korea, Rep, of fast growth focus on heavy industry brief rtccssion resumed fat growth
light industry and adjustment
Malaysia commodity boom spend way out of crisis short severe recovery
recession
Thailand recession, devaluations, resumed fast
change in trade regime growth
1970 1971 1972 1973 1974 1975 1976 1977 1978 i979 1980 1981 1982 1983 1984 1985 1986 1987 1988
Noe-: SA = siructural adjustment
Source: Country studies.
14 Susan Hiorto,i, Ravi Kanbuir, and Dipak Mazurndar
Table 1.2 provides some information on REERs. Of the three
countries with the least success in adjustment, two also failed to achieve
real currency depreciations after 1982 (Egypt's currency appreciated
quite sharply), and the third country (Argentina) did not sustain
depreciations. Most of the other countries for which data are available
achieved some depreciation of their currency: Kenya after the 1981
devaluation and reforms (although the policy may have begun to slip
in 1986), Bolivia after the 1985 policy change, Chile after the 1984
stabilization, Costa Rica after the major structural adjustment efforts in
1984, Malaysia after the onset of the 1984 recession, and Ghana after
the economic recovery program began in 1983. With this background
on the nature of the adjustment experience in the 12 countries, we turn
now to labor markets and their role in the adjustment process.
The' Role of Labor Markets during Adjustment
The country studies all address the issue of how well labor markets
worked during adjustment. The discussion here is organized as
follows: first, aggregate real wages and unemployment are examined,
then the effect of distributive conflicts and the ensuing macro-
tradeoffs are discussed, third, sectoral employment shifts and relative
wa,ge are examined, and finally, the role of labor market institutions is
dealt with.
Unemployment and Real Wages
Tables 1.3 and 1.4 summarize country experience as regards
unemployment and real wage trends for the 12 country studies.
Unemployment series are available for nine of the countries studied
(they are not availablc for the three countrics in Sub-Saharan Africa).
As the definition of unemployment varies across countries, cross-
country comparisons require .some caution. Most of the countries do
show cyclical or trend increases in unemployment related to periods
of recession and stabilization (see also figure 1.1). Chile exhibits the
most dramatic unemployment, with unemployment levels of over 10
percent in all the years from 1976 (when the series begins) until 1987,
reaching a peak of 26 percent in 1982. Understanding how the rate
could remain so high for so long in the absence of unemployment
benefits is difficult. At the other extreme, Korea's exceptionally low
Labor Mark-ets in an Era ofAdjusnnent An Overview I5
unemployment rate despite the large shocks it encountered as an oil-
importing open economy is notewortLiy.
Several of the studies discuss the composition of the unemployed
and generally confirm the "luxury unemployment" hypothesis,
whereby those openly unemployed are more frequently secondary
household workers (that is, not household heads) and are often the
more educated. E ,ypt represents an extreme case where a national
survey found that 76 percent of the unemployed were new entrants to
the labor force, and 74 percent had a high school education or above.
Educated female unemployment is a particular problem in Egypt, as
few opportunities are available ouitside the government sector. Of the
female unemployed, 97 percent were new entrants to the labor force
and 96 percent had a high school education or above. In Thailand,
unemployment is highest among those with a vocational education,
and unemployment of university graduates rose in the 1980s, when
government employment owth slowed dramatically. Likewise in
Malaysia, the educated unemployed phi.nomenon has changed over
time, from unemployed high school graduates to unemployed college
graduates, and as in Egypt, educated unemployment is coucentrated
among women. The Costa Rica and Bolivia studies both document
another feature of the composition of the unemployed, namely, an
increase in the share of heads of households and of job leavers among
the unemployed in crisis years.
In most of the countries, weak labor demand did not result only in
unemployment. Underemployment increased, although this is hard to
measure (Argentina, Bolivia, and Costa Rica studies provide data and
show that it has generally moved with the unemployment rate).
Participation rates also changed, and informalization increased. Only
Bolivia and Chile used formal emergency employmenc programs, but
many of the countries bolstered public employment at least as a
temporary measure during the crisis until fiscal stabilization measures
dictated cuts in public sector employment.
Changes in participation rates can affect the measurement of
unemployment. However, researchers do not agree as to whether the
added or the discouraged worker effect will predominate. (The added
worker effect is where the income effect of lower earnings during
recessions leads to the household supplying additional labor. The
Table 1.3 Unemployment Rates, 1979-89
(percentt)
Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1P79 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
Argentina n.a. n.a. n.a, n.a. 4,2 3.4 4.8 3.3 3.3 2.5 2.5 4.8 5.3 4,7 4.6 6.1 5.2 5.6 6.1 n.a.
Bolivia n,a. n.a. n.a. n.a. n.ra n.a. n.n. n,a. n.a. n11a. 7.5 6.2 7.5 8,2 6.6 537 4.2 5.9 11.5 10.7
Brazil n.a. n.a. n.a. n.a, n.a., n.a. na. n.a. -6.8 6.4 6.3 7.9 6.3 6.7 7.1 5.3 3.6 3.8 n.a. n.a.
Chile n.a. n.n. n.n. n.a. n.a. n,a. 17.4 16.9 17.9 17.0 15.0 25.0 26.2 21,4 19.0 13.6 10.9 7.2 n.a. n.a.
Costa Rica n.a. n.a. n.a. n.a. n,a, n.a, 6.2 4.6 4.5 4.9 5.9 8.8 9.4 9.0 n.n. 6.9 5.9 5.6 5.5 n11a.
COte d'lvoire n.a. n,a. n.a. n.a. n,a. n.a. n.a. n.a. n.a. n.a. 2.5 n.a. n,a. n.a. n.a. 2,8 2.5 n.a. n.a. n.a.
Egypt 2.4 1.8 1.5 4.7 2,3 2.5 n.a. 3,1 3.6 4.6 5.2 5.4 5.7 6,6 6.0 n.a. n.a. n.,a 6.8 n.a.
Ghana n.a. n.a. n.a. n.a. n,a. n.,. n.n. n.n. n.a. n.a. n.a. n.a. n,a. n,a. n.n. n.a. n,.a n.a. n, a. n,Aa.
Kenya n.n. n.a. n,a. n,a. n.a. n.a. n.a. 6.1 n.e. n.. n. a. n.a. n.a. n.a. n.a. n.n. 6.9 n.a. n.a. n,a.
Korea, Rep. of a 4,5 4.5 4,5 4.0 4.1 4.1 3.9 3.8 3.2 3,8 5.2 4.5 4.3 4.1 3.8 4.0 3.8 3.1 2.5 n.a.
Malaysia 7.6 n.a. n.a. n.a. 6,7 n.e, n.a. n.a, 6.3 5.7 5.7 5.0 4.7 5,5 6.3 7.6 8.5 8.2 n.a. n.a,
Thailand n.a. n.a. n.a. n.a. n,a, 0.4 0.8 0,8 0,7 0.9 0.9 0.9 3.6 1.9 2.3 3.7 3.5 5.8 n.a, n. a.
n.a. - not available
a. Manufacturing only.
Sources: Argentina, Bolivia, Chile, Costa Rica, Kcnya: country studies. Brazil: Riveros (1989), COte d'lvoire: 1980 census; Fields
(1990). Egypt: population census, Korea: Bank of Korea,. Priticipal Economic lindicators. Malaysia: Wong (1985) for the 1970s; Fifth
Malaysia Plan 1986-90 for 1980; World Bank economic reports for 1981-89, Thailand: Statistical Yearbook (various years).
Table 1.4 Real Wages Indices, 1970-88
(inidex, 1980 100)
country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 82R 1983 1984 1985 1986 1987 1988
Argentina 108 112 105 115 129 124 80 74 77 86 100 9 1 80 97 106 87 82 72 n.a.
Bolivia 100 1 14 1 13 118 94 86 9 8 107 108 110 100 80 56** 42 36 55 34 42 n.0.
Brazil 79 n.a. n.a. n.a, n.a. 84 8 5 89 9 4 9 5 100 109 122 113 105 113 122 106 n.a.
Chilo 110 n.n. n.a. n.u. n.n. 63 7 2 82 9 1 100 100 1 14 133 95 8 9 7 6 73 71 n.a,
Costa Ricaa n.a, n,a, n.a. n.a, n,a. n,a, 8 0 96 9 7 105 100 85 6 3 7 7 n.a. 9 2 95 n.a. n.a.
Costa Ricab n.n. n,a. n.a. n.,. n,a. n.a, n,a. n.n. n.n. n.a. 100 84 6 5 7 7 n,a. 93 97 n.a. n.a.
*C6te d'lvoire n.a. n,n, n,a, n,a, n,a, n.e. n.a, n,a, n,a, n.a. 100 n.a. n.a. n.a. n.a. 115 n.a. n.a. n.a.
*Egyptc n.a. n.a. n.a. 89 9 2 73 79 9 2 8 9 105 100 103 104 108 118 120 103 91 n,n,
*Egyptd n.a. n,a. n.a. 93 9 6 86 8 8 109 100 103 100 105 108 104 108 101 9 2 84 n.a.
*Fgypte n,1a. n1.0. n.a. 125 109 104 105 104 103 103 100 108 109 98 9 6 89 75 6 9 n.a.
*Ghana . n.a. n.a. n.a. n.a. n,a. n.a. n.a. n,a. 113 93 100 5 9 5 3 41 6 0 87 141 n.a, n.a.
*Kenyaf n,a. n,a. n.a. n.a, 8 8 86 8 9 87 9 0 9 1 100 90 81 79 8 0 7 8 7 9 8 2 84
*Kcnyag n.a. n.a. n.a. n.a. 106 103 114 109 110 106 100 96 85 8 4 8 1 7 8 83 8 0 8I
Korea, Rep. of 44 45 46 53 57 58 68 82 96 105 100 99 106 115 122 130 139 150 168
Malaysia 83 83 80 7 1 76 80 85 87 89 95 100 106 III 118 125 135 133 130 127
n.a. = not available
* Denotes data from employment and carnings surveys or household surveys,
* Denotes possible break In series.
a. Paid workers (employees). c. Private sector. c. Govcrnment. g. Public sector.
b. All workers. d. Public enterprises. f. Private sector.
Soutrces: Argentina, Bolivia, Chile, Costa Rica, Egypt, Ghana, Kenya: couintry studies, Brazil: Riveros (1989). Cote d'lvoire: Levy and
Newman (1989). Korea: Bank of Korea, Principal* Economic Indicators. Malaysia: Department of Statistics, Industrial Surveys ror 1968-
74; Department of Statistics, Monthly Industrial Statistics ror. 1975-87.
18 Susan Horon, Ravi Kanbwr, and DipakManundar
discouraged worker effect is where the substitution effect of lower
wages during recessions decreases household labor supply.) Three of
the studies discuss the issue and suggest that the added worker effect
predominated in Costa Rica during the crisis and the discouraged
worker effect in Bolivia. In Argentina the authors argue that the
discouraged worker effect prevailed in the 1970s and the added
worker effect in the 1980s, although they do not explain the change-
One complication in interpreting the data is that most countries have
also observed a trend increase in female labor force participation rates
(the Bolivia, Costa Rica, Kenya, Korea, and Malaysia studies all
mention this)- Thus, separating trends in female labor force
participation over time from temporary fluctuations in response to
economic crisis is not easy.
A final form of quantity adjustment involvee a shift from formal to
informal sector employment. As employees lost their jobs, one option
in the absence of unemployment compensation was to become self-
employed, and likewise output, and thus employment, often shifted
from large formal sector firms to smaUer informal sector ones. This is
again a difficult topic on which to obtain data, and studies often define
the informal sector differently. Tokman (1984) and others have
studied the phenomenon extensively for Latin America. The Bolivia
case study argues that informalization was an important method of
labor market adjustment In Argentina, although the authors state that
the informal sector was less important than elsewhere in Latin
America, nonwage employment grew faster than wage employment in
nontradables in all three periods considered (the 1960s, the 1970s, and
the 1980s), and in both manufacturing and agrculture nonwage
employment grew faster in two of the three periods. In Brazil the main
shift was into the unprotected employee sector (those without signed
contracts) rather than into self employment
Infonnalization has been less well documented in Africa. The C6te
d'Ivoire study, however, does mention a shift between formal and
informal establishments in manufacturing, and the authors of the
Kenya paper argue that a similar employment shift occurred in Kenya,
where employment in the informal sector grew 11 percent in 1988
when wage employment growth slowed. There is also similar evidence
for Asia. In Malaysia the trend rise in employees as a proportion of
Labor Markets n an Era ofAdjustmenr An Overview 19
the labor force was reversed during the short recession. In Korea the
trend toward an increased employment share in large firms in
manufacturing was arrested in the early 1980s, and the proportion of
women who were regular employees, as opposed to temporary or
casua, likewise reversed its upward trend.
Real wage behavior, perhaps more than unemployment rates,
differentiates the country groups discussed earlier (note that the wage
data available are for the formal sector except for the Latin countries).
The Asian countries show a fairly steady advance in real wages, with
brief interruptions during the recession (although the authors stress
the importance of these real wage slowdowns in allowing productivity
to stay ahead of real wage growth and ensuring declining unit costs).
The severe adjustment countries show dramatic wage fluctuations, with
wages at their lowest less than 50 percent of the peak, and with real
wage declines far greater than the fall in GDP- Real wages in Ghana
recovered by 1986, are still not back to peak levels in Chile, and are
continuing to fall in Bolivia. The other countries are somewhat
intermediate: in two of the three moderate adjustment countries (Brazil
and Costa Rica) there are gains between the start and end of the time
series, albeit less marked than for the Asian countries, and Kenya has a
wage decline in the public sector, but private wages are closer to
holding their own. The difference between Kenya and the other
countries in this group is probably due to population pressure. Finally,
the less successful adjustment countries show a less severe wage
decline than the severe adjustment ones, perhaps explaining a little of
the political opposition to such severe adjustments. However, one must
be somewhat cautious in interpreting the reaI wage data, particularly in
cases where it is not from household survey data, because of the
employment composition issue. Earnings functions can be helpful in
this regard.
How can we interpret the evidence discussed above, and what can
we infer about labor market workings? As discussed earlier, there are
three possible explanations as to why unemployment may persist
during stabilization. The first is that the labor market is not working
well because of real wage rigidity. The evidence presented by the case
studies certainly does not favor the view that real wages were rigid, and
therefore led to unemployment. Even for Chile, where unemployment
20 Susan Horeon, Rai Kanbr, and Dipak Maumdar
was highest and persisted the longest, real wages fell dramatically. Real
wages have also been flexdble in Brazil and Argentina, despite wage
indexation- Devereux (vol. 1, chapter 4) argues that the failure of
disinflation plans in these countries is due more to inappropriate and
noncredible fiscal policies. The effect of wage indexation is only to
magnify and lengthen the inflation response. Some crtics might stil
argue that real wages did not fall fast enough, but a good case for this
remains to be made. If the inflexible real wage explanation is
inappropriate because of the observed severe falls in real wages, this
leaves the other two explanations: aggregate demand feedback from
declining real wages and output market imperfections.
The wage level enters the system as a determinant of aggregate
demand through its effect on the distribution of private income. The
mechanism depends crucially on the assumption (generally valid
empirically) that the propensity to save is markedly lower for wage
eaners than for recipients of profits. If the share of wages in total
disposable income falls, for example, savings in the economy irncrease,
and aggregate demand will fall unless there is an offsetting increase in
investment or government spetzing. These ideas are implicit in the
works of Keynes and Kalecdki, and have recently been discussed in the
context of developing countries' stabilization and adjustment
problems by Taylor (1988) and others.
Taylor distinguishes between what he calls 'exhilaratienist' and
"stagnationist" economic scenarios. In the former, output is
constrained by bottenecks related to a short supply of capital. Real
wage cuts leading to a higher profit share will increase the supply of
savings, and may translate into higher investment In the stagnationist
economic scenario, however, the binding constraint on output growth
is the low level of consumer demand relative to capacity. A fall in the
share of wages in these conditions leads to stagnation. It is possible for
an economy to start from an exhilarationist position, but then slip into
a stagnationist position as wage share falls steeply.
The eistence of a dual labor market, with a distinctly lower wage
level in one sector compared to the other, reinforces the conclusions
drawn from the model with a homogeneous labor market. Consider
the case of a recession with a fall in labor demand in the formal sector.
Although the typical scenario as analyzed in the country studies is that
LaborMarkets in an Era ofAdjusnent An Overview 21
a fall in wages occurs, sometimes fairly drastic, this is not always the
case. In Malaysia the average earnings of workers in the formal sector
actually increased because those most recently hired and at the lower
spectrum of wages and skills were laid off first In all cases, however,
the workers displaced from the formal sector as well as those entering
the labor force entered the informal sector in growing numbers. Thne
share of total employment at the lower wage levels increased
significantly. For the economy as a whole, therefore, average wages
fell faster and to a greater extent than in the formal sector.
Does the stagnationist hypothesis still hold if we allow for the
possibility of exports? In the traditional model of the small open
economy (such as the one reviewed in vol. 1, chapter 3) it will not
hold unless wages are rigid downward in both the tradable and the
nontradable sectors. As demand contracts, with wage rigidity in the
nonttradable sector, unemployment will occur in this sector, but if
wages are flexible in the tradable sector, costs will fal in tradables.
However, the small open economy model assumes unlimited demand
for tradables at the going product price, thus the unemployed labor
will be absorbed in the more profitable tradable sector. Total demand
will be restored to its initial level with a larger share of GDP accounted
for by tradables. However, even if there are wage rigidities in both
sectors, the profitability of the tradable sector needed to attract
resources can still t achieved by a real devaluation that increases the
ratio of the prices of tradables to the prices of nontradables (iT/PN)-
This -is why in the textbooks devaluation is sometimes called an
alternative to wage flexbility.
The Asian and Latin American country studies provide sharp
contrasts as concerns the role of wages and devaluation in macro
demand contraction. Korea, for example, depended on continuous
nominal devaluation of its currency over a long penrod of time as well
as maxi devaluations during periods of severe external sbock. As an
export-oriented economy, Korea had to increase its competitiveness
by reducing its unit labor costs in dolar terms. Due to the rapid
growth of labor productivity, the response to external shocks was to
hold constant real wages rather than requiring a wage decline. The
combination of a maxi devaluation and temporarily preventing wages
from rising with productivity led to a very quick recovery of exports
22 SusanHornonRaviKanbur, andDipakMazundar
At the same time, since the slowdown in real wage growth was so short,
there was no significant deflationary impact in the- domestic market.
Another factor important in recovery was government policies to
counter the increase in nonwage costs following devaluation. Because
of its important role in the finance of large-scale industry, the
govcrnment was to some extent able to offset the increase in the cost
of borrowed foreign capital caused by devaluation by offering cheap,
subsidized credit to businesses. An important feature of the Korean
case of adjustment to the shocks was that exports increased rapidly in
Korea despite the rise in wages.
The last point touches on a generl point about the role of wage
flexibility in adjustment. In Southeast Asian economies the share of
wages in value added is typically one-third or a litte more (according
to Riveros 199, it is closer to 40 percent in Latin America). Thus,
changes in capital costs are often as important in determining
competitiveness as changes in wage levels. The course of events
leading up to the recession in Malaysia in the mid-1980s, and the
subsequent adjustments triggering recovery, illustrate the point vividly.
Unlike Korea, Malaysia is an oil exporter. In the early 1980s,
government spending in Malaysia increased enormously, partly to
bolster an attempt to prolong the boom associated with the oil boom.
The resultant pressure on extemal competitiveness came from three
sources: (a) wages increased, even after employment growth had
slowed down; (b) interest rates increased sharply as demand for private
capital funds competed with the public demand; and (c) the currency
appreciated in real terms because the capital account was uncontrolled,
and there was a massive inflow of capital to finance the budget deficit
The loss of competitiveness created an external imbalance that
could only be corrected through a sharp recession. Malaysia was
fortunate, however, in that all the relevant factor markets showed
remarkable flexibility. As wages fell from their early 1980s level, the
interest rate fell to a level that was nearly a third of its peak, and there
was a sharp depreciation of the currency. Clearly the "collapse" of all
the factor markets was instrumental in making the recession short-
lived. Of course, the improvement in the world economy was a factor
triggering the recovery, but it was the gain in competitiveness fed by
the downwardly flexible wages, interest rates, and exchange rates that
Labor Markets in an Era ofAdjustmentAn Overview 23
allowed Malaysia to seize the opportunity in the second half of the
1980s.
The Latin American studies illustrate almost the opposite ae in
terms of the effects of real wages on demand, with Bolivia providing
the clearest example. The fall in real wages in the 1980s was twice the
size of the fall in real GDP. Even by the end of the decade, real wages
and employment showed little sign of any recovery- The fall in the
share of wages must have depressed the domestic market considerably.
At the same time, despite the real value of the currency falling to less
than a third of the 1985 level, there was no sign of export-led
recovery- Evidently the market structure for favoring large shifts to
export did not exist in Bolivia.
By contrast, another IAtin American country, Costa Rica, hints at
the existence of a basic structure of links to the world market, and also
illustrates the advantages of an institutional mechanism that limits the
direction of wage deflation. Real wages fell between 1980-82 as
indexation tied to past inflation failed to protect workers as inflation
accelerated. In mid-1982, when stabilization was instituted, real wages
turned upwardagain as inflation deceleratedL By 1986, real wages had
regained their 1980 value. The short period to which the real wage
decline was confined might have helped to stabilize the domestic
aggregate demand. At the same time, the decline in the dollar price of
exportables, helped by the fall in real wages and the devaluation of
1980, was instrumental in improving the export situation. Thus, two
factors helped Costa Rica to stage a recovery in the post-1982 period.
Market links were important in ensuring that the fall in the real value
of the currency and in wage costs had the desirable impact on exports.
More surprisingly, indexation was significant in engineering the initial
fall in real wages and in limiting the period of wage stagnation-
Another instance of sharp deflation caused by a fall in the share of
wages in GDP comes from the case of Chile during the drastic policies
of stabilization pushed through by the military junta after the fall of
the Allende regime. This case illustrates the importance of proJuet
markets in the process of adjustment. Although extreme, it is worth
discussing because, as Ramos (1980, p. 468) points out: "other
countries may simply be experiencing in slow motion (stagflation)
what Chile experienced all at once (hyperstagflation)."
24 Susan Horton, Ravi Kanbur, andDipakMzunder
In October 1973, the junta freed prices that had been controlled
under the previous socialist regimes, but unlike Germany after World
War II, Chile did not have a monetary reform to put a cap on the freed
inflationary prices. Inflation immediately accelerated to 90 percent
during the month of October alone. Although prices moderated after
October, they continued to increase at rates higher than 300 percent in
1974 and 1975, clearly overshooting by a good deal the expected
equilibrium leveL
On the labor front, the junta's policy w-as to separate wage
readjustments fom the freezing of prices to prevent a wage-price
spiral. It postponed adjustment of wages by several months. When it
did take place, it was consistent with a much lower rate of inflation-
Thus, real wages dropped sharply, and by 1975 stood at nearly half
their pre-Allende level. Astonishingly, the real wage decline was
accompanied not only by high rates of inflation, but also by a rapidly
increasing rate of unemployment, which climbed from 3 percent in
the first half of 1973 to 10 percent in 1974 and 19 percent in the first
half of 1976. The rise in unemployment was, as one would expect,
associated with a sharp decline in industrial output, at least until the
end of 1975 (the index of industrial output halved between end-1973
and end-1975).
What explains the coexistence of a high inflation rate, faling real
wages, and declining output? The crux of the problem would seem to
be the inflationary expectations and noncompetitive behavior in the
product market There was clearly no demand pressure because
consumer demand fell very early with the fall in real wages, and
demand contraction intensified as real wages fell and unemployment
increased at a high rate. Nor was there any cost pressure, for 'whereas
the prices of imported inputs in the last quarter of 1973 rose to 30
times and wages rose 14 times their 1969 levels, product prices rose to
40 times their 1969 levels upon being freed in October 1973"
(Ramos 1980, p. 472). Prices seem to have increased in anticipation of
much higher demand and cost pressures than actually existed.
"Producers seem to have set prices to balance supply and demand not
as of the moment, but in three months' time so to speak" (Ramos
1980)- The anticipated increase could be on the side of money
LaborMarkets in an Era ofAdjusonenarAn Overview 25
demand, or in terms of unforeseen wage adjustments, devaluation, and
a rise in input costs.
The continuation of inflationary price increases in the face of
serious disequilibrium in the product market with producers unable to
sell their products is a difficult proposition to explain in terms of
textbook economics, and indeed came as a surprise to policymakers.
A major factor in the continuation of the process was that price setters
were not penalized soon enough for their erroneous expectations
because of the massive increase in the share of profits that the fall in
real wages entaileck
The inflationary expectation was finally broken when the currency
was revalued in 1976, when the balance of payments situation
reversed, showing a net surplus, and tariff reductions were undertaken
for reasons connected with the economy's long-run development. The
downward jolt these measures gave to the prices of both inputs and
final goods seems to have fueled the recovery after 1976. Prices
finally began growing less than the money supply, with output nising,
unemployment falling, and real wages rising much more than total
output.
Problems of Dtributive Conflicts
The availability of enough evidence to suggest that wages have
been flexible in many countries during the periods of adjustment does
not imply that distributive conflicts have not been major issues in
several countries. The country studies show that in Latin America, in
particular, the conflict between maint or increasing labor's share
of output and achieving external balance has been an important factor
in the limited success of stabilization policies. Countries, of course,
differ in the importance of distributive conflict in their economic
history. Apart from differences in labor market institutions, the
economy's structure seems to be critical in some cases.
One factor, that seems to be important is whether or not food is an
important tradable. The case of Argentina is a good example that
shows how the different objectives could be in conflict when basic
foods in the workers' consumption budget (cereals, meats, and so on)
are tradable goods, and- the government- does not interfere
significantly with the domestic prices of these commodities. In this
26 SusanHorton, Ravi Kanbur, andDipakMazmdar
case there is a close relationship between the exchange rate and the
product wage in the economy's nontradable sector. Currency
devaluations lead to increases in the domestic price of food, which in
tum leads to upward pressure on money wages. Such an increase will
not affect product wages in the tradable sector, since product prices of
traded goods would also have increased in the domestic market, but
other things being equal, the product wage in the nontradable sector
will increase. In this case a conflict of interests arises between the
producers of nontradables and the workers employed in this sector.
This exchange rate wage tradeoff, taken together with the
nontradable sector's more powerful political position because of its
.urban location, has given rise to the wage cycle documented in the
country study. When external markets for Argentina's food exports
are strong, the currency tends to slide into overvaluation, which helps
increase real wages without hurting profitability in the urban
nontradable sector and fiscal-balance in the public urban nontradable
sector. However, when the external terms of trade we2iken, devaluation
is imperative to ease the problem of external imbalance, and various
forces are set in motion that depress real wages to protect profitability.
In Argentina, as in much of Latin America, bunsts of inflation have
often been the mechanism for reducing real wages.
Note that not all countries have a large proportion of their wage
goods or food as tradables as Argentina does- In particular, in many
Asian economies (including the two in our sample, Korea and
Malaysia) rice, although an internationally traded good, is more like a
nontradable because of government price policies. In these countries,
the government plays a dual role in the rice market On the one hand,
it buys rice from the farmers at a high procurement price to help
support the level of earnings in this sector. On the other hand, it
distributes the rice through its retail outlets at a subsidized price for
the benefit of, for the most part, urban consumers. The financial
deficit caused by the difference between the buying and selling price
of rice is covered by the central government's overall budget. Thus,
although the govermnent imports rice to supplement the amount
procured from local farmers, the domestic price of rice is insulated
from the border price. This important wage good is; in effect, a
nontradable. The problem analyzed above, which stems from an
Labor Market in an Era of Adjusunent: An Overview 27
inverse relationship between the external value of the currency and the
price of the wage good, does not exist for such economies (although
the fiscal issue does).
The supply of capital may also lead to a tradeoff between wages
and the exchange rate. Let us assume that the growth of output is
constrained by the supply of capital (savings) rather than by demand
(in other words, the economic scenario is an exhilarationist one). The
share of wages in value added has a direct effect on total savings, and
hence on the growth rate of output. The exchange rate also affects
output growth from two angles. First, the higher the value of the
currency, the -greater the trade deficit that, if it can be sustained,
increases foreign savings (borrowing) in the economy. Second, a
higher value of the currency reduces the cost of intermediate inputs,
and effectively increases the marginal impact of savings on output
growth.
An exchange rate/wage tradeoff exists in the sense that a given rate
of savings (and growth rate) could be achieved with different pairs of
values of the exchange rate and the wage share; the higher the latter,
the higher must the degree of overvaluation be. Government policy
affects both the exchange rate and the wage share through its
determination of the rate of growth of the money supply, and hence
the rate of inflation. With indexation rules determining both exchange
rate and wage adjustments, lags in the system mean that a higher rate
of inflation achieves both a higher rate of overvaluation and a lower
share of wages. Thus, an equilibrium relationship exists that connects
the rate of inflation, the value of the exchange rate, the wage share,
and the real growth rate of the economy.
The case of Brazil illustrates the key problems and constraints in
this system. During 1967-83, Brazil followed a policy of stepping up
the growth rate by expanding the money supply. This led to a rise in
the raze of inflation and a fall in the share of wages. The associated
increase in the real exchange rate and the fall in the share of wages
both increased real output growth by increasing foreign and domestic
savings and reducing the domestic cost of imported inputs. The
mechanism for bringing about this change worked as long as changes
in the values of the relevant variables were sustainable- The feasibility
of a fall in the share of wages depended on the existence of an
28 Susan Horion, Ravi KanburD andDipakMazumdar
authoritarian political system. Similarly, the appreciation of the
currency meant an increase in the trade deficit that could only be
financed by foreign borrowing. The persistent increase in foreign debt
was one of the costs of this strategy of boosting the real rate of
growth.
The first oil shock of the mid-1970s meant, in effect, a change in
the parameters of the Brazilian production function, so that at the old
values of the variables, output growth was depressed. At the same time
the import bill increased sharply. The government's response to this
situation was to undertake a program of import substitution in capital
and intermediate goods, financed by stepped up foreign borrowing.
The second oil shock and the increase in interest rates finally made
this policy unsustainable. The debt burden had reached a level when
further foreign borrowing was no longer an option to maintain an
overvalued currency. A new element in the situation was the change in
the political system. It was no longer easy to reduce the share of wages
with a higher rate of inflation. The country study discusses the
distributional conflicts in more detail.
Thus, two barriers prevented achievement of a higher real savings
rate to counteract the effect of the deterioration of the extemal terms
of trade. The government could not continue to overvalue the
currency nor to depress wages. Nor could these be. changed with a
higher rate of inflation in such a way that a new equilibrium set of
values of the relevant variables could be achieved. This was at the heart
of the failure of stabilization efforts in the 1980s. One way out would
have been if total factor productivity growth could have been
increased to a sufficient degree, but evidently the Brazilian economy
was unable to achieve this goal. On the contrary, the country study
indicates that labor prod-uctivity actually fell as labor hoarding in the
formal tradable sector increased significantly in response to the
deteriorating employment situation. The contrast with Korea's
experience is striking. The country study documents the enormous
importance of total factor productivity growth in the Korean
economy's successful adjustment to the oil price shocks. Because of
the increase in total factor productivity, the required decline in the
share of wages could be achieved with a negligible decrease in the
Labor Markets in an Era ofAdjusimen: An Overview 29
absolute level of real w~ages, and the increase in the cost of imported
inputs due to devaluation could be largely offset
Sectoral Employment Shifts and Relative Wages
Sectoral employment shifts are a key part of structural adjustment,
and Edwards and Edwards (vol. 1, chi.. ter 2) discuss these in a basic
two-sector two-factor dependent economy model in the presence of
labor market distortions. They examine four different scenarios plus
the basic competitive case. In the basic model, standard results apply
and labor would tend to benefit from trade liberalization, which the
authors define as tariff cuts: the effects of devaluation, which usually
accompanies adjustment, are not considered. Even if economywide
wage rigidity is allowed for, the authors argue that trade liberalization
will result in unemployment in the short run wvhere capital is
immobile. However, in the longer run, if importables are the less-
labor-intensive sector, starting from an initial conditLion of
unemployment, trade liberalization will increase total employment in
the economy.
The authors then take the case where only the importable sector is
covered by a minimum wage. In the short run there will be lower
employment in importables and higher employment in exportables,
but employment in nontradables and total employment is ambiguous.
They conclude that: "In the presence of labor market distortions,
trade liberalization policies usually considered to be beneficial may
generate nontrivial (short run) unemployment problems." This
conclusion holds also in the third variant considered, that of capital
account liberalization.
The fourth and final case considered is where wage distortions in
importables are related to the degree of tariff protection. In the short
run, trade liberalizition increases unemplovment and depresses wages
in the economy's other sectors. Although some of this would
disappear in the long run, the scenario highlights possible political
economy conflicts. Labor, "the factor of production that is supposed
to gain from freer trade, is negatively affected in the short run, and the
long run gains are hard to perceive when compared to the initially
distorted situation of the economy."
30 Susan Horon, Ravi Kanbur, and Dipak- Mazundar
The country studies provide information on wages and
employment by various sectoral groupings: economic sectors
(agriculture, manufacturing, construction, and so on), formallinformal,
public/private, tradable/nontradable, and occasionally even finer
categories such as importablelexportable(nortradable. They also
provide some information on skill groups. How formal/informal or
skilled/unskilled categorizations correspond to the tradable(
nontradable distinction that is of key interest is not always clearly
specified in the country studies, and varies between countries (the
Brazil study provides the most complete breakdown). One important
problem in many of the studies is that the agriculture sector is an
important component of tradables, but no agricultural wage data over
time exist for the Latin American countries, and neither agricultural
wage nor employment data over time exist for the African countries..
Let us consider sectonrl employment first, and then sectoral relative
wages. Table 1.5 provides information on employment shifts by sector
of GDP for 8 of the 12 countries in the study (the ILO Yearbook does
not have data for the other four countries). The ILO Yearbook reports
sectoral employment data by 10 sectoral groups, which are here
further grouped into primary, manufacturing, utilities and
constructon, and tertiary. This classification is used on the assumption
that, roughly speaking, primary and manufactured goods are tradable,
whereas the output of the construction, utilities, and tertiary sectors are
not. Obviously this grouping is rather crude, and the Costa Rica and
Argentina studies provide more detailed information on the
tradablelnontradable shift, even to the extent of comparing
employment in the traditional and nontraditional export sectors (Costa
Rica).
The debt crisis years had clear effects on structural tnsformation
in the countries studied, in that the usual changes accompanying
development either halted or reversed in all cases. In Brazil, Costa
Rica, Korea, Malaysia, and Thailand (the, Asian and partial adjustment
countries), the manufacturing share declined somewhat during
stabilization, but then resumed growth. The recovery is strongest in
Korea. The data series for Malaysia and Thailand both end before
recovery sets in strongly. In Chile and Bolivia, declines in the
manufacturing share were more stiking. These were reversed under
Labor Markets in an Era of Adjustment: An Overview 31
structural adjustment in Chile, but not as yet in Bolivia. The data series
for Egypt stops before economic problems intensified, but a decline in
the share of manufacturing employment is already evident.
The LLO Tzarbook does not contain data on sectoral employment
trends for Sub-Saharan Africa, but the country studies contain some
information. The Kenya country study argues- that urban employment
figures suggest that the manufacturing share stagnated after 1978.
Data for tite formal sector for the C8te d'Ivoire suggest a large
decrease in modern manufacturing employment despite subsidies.
The debt crisis slowed the transition out of agriculture for most
countries, and for Bolivia, Cote d'Ivoire, and Ghana shifts back into
agriculture are evident. The ILO data for Bolivia show that although
the prirary share overall declined, the agriculture share increased in
the worst years (1982-83) and stagnated thereafter. In the Cote
d'Ivoire a shift back into agriculmure occurred (based on labor force
transition behavior). In Ghana, the capital city, Accra, changed from
being, the destination of 46.5 percent of migrants prior to 1970 to
being the source of 60.0 percent of recent migrants in 198247. Even
in the higher-income countries, agriculture played an important role
in absorbing labor market entrants: one-third of new jobs in Malaysia
during the 1986-87 recession were in agriculture, and agricultural
employment grew as fast as total employment in Chile in the
successful adjustment period after 1985.
The Argentina and Costa Rica studies both examine employment
shifts between the tradable and nontradable sectors. Argentina has
seen a secular trend toward increased employment in nontradables,
and Costa Rica has seen a sirilar trend out of exportables. As-
Argentina did not have a sustained adjustment program, this trend
continued in the 1980s, but in Costa Rica adjustment arrested, but did
not reverse, the trend. However, the study authors find some cause for
optimism in the growth of the small, nontraditional export sector.
A useful exercise is to examine sectoral wage data in conjunction
with sectoral ermployment figures. Simple theory suggests that the
effect of structural adjustment policies should lead to a relative
increase in wages in tradables to encourage labor movement (unless
markets are so frictionless that the reallocation does not require price
signals). However, employment shifts may also cause changes in
Table 1.5 Employment by Sector, 1971-89
(percenttage of total employmtient)
Coaunytseefor 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 )984 1985 1986 1987 1988 1989
Bolivia
Bolivi r 53.7 53.1 5 52,2 51.6 50.9 50,5 50.2 49.4 50.9 51.0 50.9 52.1 51.9 S1.S 50.0 49.9 50.0 50.0
=anuacturing 8,8 8.8 -8.9 9,0 9,0 9.1 9.2 9.3 9.4 10.3 10.0 9.1 8.9 8.8 8.7 7.0 1.1 7.1 7.1
Utilities and
construction 4.3 4.7 5.0 5.5 5,9 6.4 6.4 6.5 6.5 5.9 5.1 3.7 3.7 3.3 3.2 3.1 3.2 3.2 3.2
Teriary 33.2 33,4 33.5 33.3 33.4 33.6 34.2 34.1 34.7 32.9 33.9 36,3 35.2 36.0 36.5 39.9 39.8 39.8 39.8
Brazil
n.a. n,.e n.a. n.a. n.e. na. n.a. n.n. n e. n.a. 29.3 29.5 27.1 29.8 28.5 25.9 24.6 n.a. n.a.
Ma%acturdngb n.a. n. n. n.n, n,a. n. n.e. n.a. n.a. n.n. 24.7 23.4 14.0 14.2 1 4.7 16.2 15.7 n.a. n.a.
Utilides and
construction n,a. n,n. n.a. n.,. n.a. n.a, n.a. n.e. n.a. n.a. n.e. n.a. 11.4 7.5 7.4 8.0 8.1 n.a. n.a.
TestJry n.a. n.a. n.a. n,a. n,a. na. non n.a. n.a. n.a. 46.1 47.1 47.6 48,4 49.3 50.0 51.6 n.a. n,a,
Chile
n.e. n.a. n.a. n.a. 24.6 20,6 21.1 20.2 19.3 18.5 17.3' 18.1 17.7 18,0* 22.5 22.8 22.9 22,3 21.7
MJn2curing n,n, n.n. n.n. n.a. 16,8 16,8 10.7 16.3 16.5 16.1 15,5 12.7 12,6 13.8' 13,3 13.6 15.I 15.7 16.9
Utilities and
construction n. n.o. n,u. n,a. 5.3 4.8 4.7 4.8 5S1 5,4 6.0' 3.7 3.6 4.4' 4,6 5.4 5.8 7.1 7.3
Tertiary n.a. n.B. n.a, n.a. 52.6 57.4 57.2 58.5 58.9 59.8 60.7' 65.4 66.0 63.71' 59.6 58.2 56.0 54.8 54.1
Costa Rica
Prinmary na.a n.n. 38,2 n.e. n,n, 34,8 33.0 30.4 28.9 27.4 27.6 30.0 28.2 30.0d 27,3 26.90 28.1 28.1 26,2
Manufactuuing .na. una. 12.9 n.e, n.e. 14.6 15.8 15,2 16.5 16.3 15.4 15.2 16:6 15:2d 15.9 17.1' 17,5 16.7 18.8
Udildcs end
co istruction n.n, n,a. 6.9 n.a. n.a. 6.5 6.4 7.4 8.2 7.8 6.7 5.7 S,l 4,9d 5.1 5.8' 5.9 5.9 6.2
Terliars n.a. n.a. 42.0 n.e. n.a. 44.0 44.8 47.0 477 48.5 50.3 49.1 50.1 49:9d 51.7 50.2' 48.5 49.3 48,8
Prfua 54,2 53.8 51.5 47.6 49,1 n.e. 45.8 42.5 42,1 42.6 40.5 39.3 41,3 40.9 n.a. n.a. n.a. n.a. n.a.
nnuracturing 12.5 12.8 41 15.3 14,3 n.a. 14.7 15.1 16.0 14.7 15.9 15.3 14.7 13,9 n.a. n.a. n.a. n.a. n,n.
Ulilitles end
construction 2.7 2.83 3. 3.1 3,2 n.e. 4.2 4.8 5.4 5,2 5.9 6.3 6.2 5.9 n,. n.a. n.a. n.a. n,a.
Tertiary 30.6 30.6 31.0 33.5 33.3 n.e, 35.3 37.6 36.5 37.5 37,8 39.1 37.8 39.3 n.a. n.e. n.a. n,u. n.e,
Korea
EHlMary 49.4 51.1 50,4 48,6 46.4 45.1 42.6 39.2 36.6 34.9 35.1 32.8 30.5 28.1 26.0 24.8 23.0 21.5 20.1
M %tikiU uring 13,3 13,7 35.9 17.4 18.6 21.3 21.6 22.4 22.9 21.7 20,4 21.1 22.S 23.2 23.4 24,7 27.0 27.7 27.6
Utilities and
cort lruction 3.7 4,1 3,6 4.2 4.6 4.5 5.1 6.3 6.5 6.4 6.S 6.0 6.0 6.5 6.4 6.0 5,9 6.4 6.8
Tcrtmy 33.7 31,0. 30.0 29.8 30.4 29.0 30.7 32.1 34.1 37.0 38.0 4D.1 41,1 42,2 44.3 44.5 44.1 44.5 45.5
Pdfmary n.u. n.a. n.e. n.a. n.e. n.a. n.2. n.a. n.a. 38.2 36.8 32.1 32.9 31.3 31.1 31.3 31.4 n.. na.
Manufacluring n.a. n.a. n.a. n.a. n.n. n.a. n.a. n.a. n.a. 16.1 16.1 35.5 17.0 15.4 15.0 15.2 15.5 n.a. nn..
Utilities nd
construction n.a. n.a. n,a. n.n. n.n. n.n. n.n. n.e. n.n. 7.1 8.3 7,9 8,9 8.9 8.0 7,0 6.2 n.a. n.a.
Teriary n.a. n,a, n,a.. n.u n.e, n.a. n.n. n.a. n.s. 38.7 38.8 44.4 45.2 45.0 45.5 46.5 46.8 n.a. n.a.
Thallandd
Pdmn7 79.3 72,9 72.6 65,7 73,1 75,9 73,7 73.8 n,.. 70.9 64, 61.9' 63,4 64,9 63,9 63.9 n.n. ne. n.a.
= ctnutacluring 4.0 7.7 7.0 9.9 7.5 6.2 6.5 6.8 n.a. 7.9 9.2 10.2* 9.6 9.2 9.4 9.1 n.a. n.a. n.a.
Utilities and
c9struction 1.2 1.7 1.8 2.0 1.4 1.5 1.9 1.7 n.a. 2.2 3.3 3.2V 3.2 3.3 3.2 3.1 n.n. n.n. n.a.
Terllary 35.5 17.7 18.5 22,4 18,0 16.3 17.9 17.7 n.a. 18.9 22.9 24.7' 23.8 22.6 23.5 23.8 n.a. n.a. n.a.
n.a. = not available
* Change in sample or methodology. Sec ILO (1989).
a. Excludes mining.
b. Includes mining.
c. Includes utilities.
d. November not July (usual).
Nore: For Thailand, repair and installation services are included In manufacturing, san!tary services are included in utilities. Prior to 1983,
unpaid family workers working less than 20 hours were excluded. The primary sector throughout includes agriculture and mining, the
tertiary sector includes commerce, transport, banks, services, and other. Figures may not sum exactly to 100 percent due to rounding.
Source: ILO (various years), author's calculationb.
34 Susan Horton, RaviKanbur, andDipakMazumdar
relative wages. In practice, structural adjustment has been associated
with labor shedding from government and from formal sector
activities (either due to reduced tariff protection or the removal of job
security legislation). As workers cannot remain unemployed for long
in developing countries due to the lack of unemployment benefits,
labor has tended to move to sectors with flexible entry, frequently the
informal sector or agriculture. The crowding of labor in these sectors
may have also depressed relative wages in the shart rum Thus, relative
wages in nontradable sectors with easy entry (for example, commerce,
services) may have been depressed both directly due to exchange rate
changes and indirectly due to labor crowding, while wages in tradable
sectors with easy entry (for example, agriculture) could go in either
direction in the short run due to opposing effects. Furthermore,
changes in labor force composition within sectors can obscure trends.
Sectors losing labor may experience increases in aggregate wages due
to the loss of worker. with the lowest levels of human capital and
seniority. The latter effect can be dealt with by the use of earnings
functions as discussed later.
All the studies (except Cote d'Ivoire and Thailand) provide some
information on the changes in relative wages, whether between
economic sectors, formal/informal sector, tradable/nontradable, or skill
categories. Table 1.6 summarizes the results by broad GDP sectoral
categories for seven countries, and table 1.7 shows the results by
tradables/nontradables for two countries and for the public/private
sectors for five countries. The data in table 1.6 are for agriculture,
manufacturing, construction, and service sector wages, where available.
As construction is the largest component of the group utilities plus
construction, and services are similarly the largest component of the
tertiary group, the sectoral wage data in table 1.6 correspond
reasonably well to the sectoral employment data of table 1.5 In
general, relative wage changes did support structural adjustment
objectives, although this is not necessarily true for each country and
every sector.
In Ghana relative wages increased in agriculture and mining,
sectors featuring heavily in the Economic Recovery Program (see
table 1.6 and the Giana study). In Egypt relative wages increased in
agriculture (see table 1.6 and the Egypt study) largely because other
Table 1.6 Real Wage Indices by GNP Sectoral Classification, 1970-89
(index, 1980 = 100)
Country/scFtor 1970 1971 1P72 197 1974 197S 1076 1977 1978 1979 1Q80 1981 1982 1982 1984 1985 1986 1987 1988 1989
Manufacluring 98 III 112 121 97 96 112 109 109 112 ION * 81' 114' 136f 520 59a n.a. n.a. n.a.
Conslruclion 83 98 96 90 80 74 90 95 95 108 100 * 85a 47a 93a 52a 74a n,a. n.a. n.a.
Services 125 1228 125 130 109 8 8 95 115 III 105 Ill * 63a 57a 1041 46' 72a n.,. n.s. n.s.
Chile
Manufacturing 114 135 93 50 54 72 80 15 102 108 IOU 132 147 112 101 85 79 81 86 n.a.
Total 112 118 104 58 56 63 72 8 1 91 1(o 100 114 133 95 89 76 73 71 72 n.a.
Egypib
An rcullure n.a. n.m n.a. 48 53 63 75 84 87 97 100 IIS 129 139 157 IS 140 116 n.a. n.a.
manufacluringc n.. n.a. n.a. 74 82 79 85 100 99 Inn 100 107 113 118 132 124 110 99 n.a. n.a.
Construction n.a. n.a. n.a. 64 8 1 95 104 110 108 112 Inn 97 93 85 85 90 85 74 n.s. n.s.
ServiceaC n.a. n"a. n.a. 80 80 77 82 85 112 107 100 101 101 99 104 126 101 86 n.a. n.a.
Ghana
Manufacturing n a. n.a. n.a. n.m. n.a. n.a. n.a. n.a. 121 115 100 71 6f4 46 85 92 153 n.a. n.a. n.a.
Consiruction n,a. n,a. n.m. n.m. n,a, n.a, n.a. n.a. 115 91 100 6S 54 61 67 1818 175 n.m. n.a. n.a.
Services n,a. n.a, n.,. n",. n.a. n.a. n.a. n.a. 116 92 100 57 49 36 55 70 131 n.a. n.a. n.a.
Kenya'
Manufacturing n,a. n.m. n.a. n.n. 119 107 106 107 103 97 100 90 - 852 85 83 81 81 84 87 84
Construction n.m. n.a, n.a, n.a. 93 93 95 92 97 90 100 92 69 69 6f8 66 6f4 71 66 67
Services n.m. n.n. n.m. n P. 99 8YR 96 90 87 93 100 90 79 79 84 83 8f6 91 88 90
Korea
Agriculture 50 61 64 6f8 73 76 81 89 93 99 100 107 114 120 138 146 n.a. n.a. n.a. n.a.
Manufacluring 45 47 49 53 57 58 6 8 82 96 105 100 99 106 115 121 130 n.a. n.a. n.a. n.a.
Rubber 8 0 77 76 82 86 71 8 6 91 93 97 100 92 93 92 92 91 95 99 n.a. n.a.
Oil palm 64 61 63 66 73 73 76 80 88 96 100 107 103 101 106 110 111 108 n.e. n.s.
MlanufaclurIng 83 83 80 71 76 78 84 88 89 96 100 106 III 119 125 135 133 129 n.a. n.a.
Constructlon 86 f 87 89 92 82 90 94 96 97 10o n.a. 112 122 126 126 128 129 124 n.m. n.m.
n.a. = not available
4 Break in series
a. March 1982 = 100.
b. Private sector only (separate series for public sector available).
c. Entcrprises of 10 and more workers.
Note: DeTinitions of sectors may not be identical across countries.
Source: Country studies. Data for Bolivia and Chile arc household surveys, the rest are employment and earnings surveys,
lable 1.7 Real Wage Indices by Tradable/Nontradable and Public/Private Sectors, 1970-89
Cauntry/sectar 1970 1971 1972 1973 1974 1975 l1t6 1977 1978 1979 1980 1981 1982 1083 1984 1085 1986 1987 1988 1989
Argentina Traduble and nonirudrable (rain* of vavraga wages)
Tradables/
nonfradables 0.89 0.91 11.91 0,88 0.93 0.92 0.93 0.94 0.91 0.93 0.89 0,91 (.91 0.95 0.96 0.97 0.96 n.a. n.a. n.a.
Polenllally tradedl
nontradables 1.04 1.04 1.03 1,00 1.05 1.02 L.(4 1.04 1.01 1.04 0,99 1.03 1.03 1.05 1.10 1.10 1.12 n.a. n.a. n.a.
Costa Rica*
Exparl/nonlradables n.a. n.a. n.a. n.a. n.a, n.a. 0.92 0.93 0.87 0,68 0.,3 0.83 0.91 0.97 n,a. 0.90 n.a. 0.75 0,78 n.a.
lmponrs/nontradables n,a. n.a. n.a. n,a. n.a. n.a. 0.98 1.0 1,00 1.05 0.96 1.10 0.98 1.05 n.a. 1.03 n.a. 0.96 1.0o n.a.
Public/nontradebles n.,. n.a. n.a. n.a. n.a. n.a. 1.20 1.34 1.19 1.15 1.03 1.13 1.1 1.18 n.a. 1.08 n.S. 1.03 1.01 n.a.
Public ant private (Index, 1980 = 100)
B.razJI
Private" n,a. n.a. n.a. n,a. n,a. n,a. n.a. n.a. n.a. n.a. Ion 107 114 106 99 IOS 117 108 95 n.a.
Public n.a. n.a. n .a n,a, n.a, n.a. n.a. n,a. n.a. n,a. 100 97 103 86 78 99 116 n.a. n.a, n.a.
SpKiVale n.a. n.a. n,a. 89 92 73 79 92 89 104 100 103 104 108 1IR 120 103 91 n.a. n.a.
Publlc n,.e n.a. n.a. 125 109 104 105 109 104 103 100 108 109 98 96 89 75 69 n.a. n.a.
Public entities n.a. n.a. n,a. 93 96 86 88 95 100 103 100 lOS 108 103 1OR 101 92 84 n.a. n.a.
Ghana
P;ivatc n.a. n.a. n.. n,a, n,a, n.a. n.a. n.a. 116 95 100 53 56 38 69 q89 I IS n.m. n.a. n.a,
Public n.a. n.a. n.a. n.a. na. n.a. n.a. n,.e 113 92 100 61 52 41 58 86 147 n.a. n.a. n.a.
Kenya
PrIvate n.a. n.a, n.a. n.a. 88 86 89 87 90 91 10O 90 81 79 80 78 79 82 84 85
Public n.e. n.a. n.a. n.a. 106 103 114 109 110 106 100 96 85 84 81 78 83 Rn 81 H1
Costa Rkca
Privatc n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a, 100 82 63 79 n.a. 96 n.a. 108 105 n.a,.
Public n.R. n.a, n.a. n.a, n.a. n.a. n.a. n.a,. n.e. n.a. 100 94 71 91 n.a. 101 n.a, 115 106 n.a.
n.a. = not available
" Adjusted for human capital
' Sao Paulo
Source: Country studies.
LaborMarket in an Era ofAdjusnment:An Overview 37
sectors were unable to adjust employment. In Bolivia manufacturing
wages did relatively badly, which is consistent with falling employment
(table 15). In Chile manufacturing wages increased relative to average
wages (table 1.6), again consistent with an increasing share of
employment In Malaysia manufacturing and construction wages
tended to increase during the recession. The Argentina study found
that relative wages had tended to increase in nontradables during
1940-62 (vol. II, chapter 1), but that the failed structuraI adjustment
attempts since then had at least managed to arrest the trend. In Costa
Rica relative wages in importables and nontradables fell during the
1980-82 recession, but recovered faster during the ensuing
adjustment period, thus maintaining their relative position overall
during the period (table 1.7).
Government wages (table 1.7) seem to have fallen universally
during adjustment due to pressures on government expenditures
(although some country study authors suggest that the wends were
different in the central government and in the parastatal.s). This is
documented in the Brazil, Costa Rica, Egypt, Ghana, and Kenya
studies. The Bolivia study also provides evidence on falling relative
wages in government, and the Malaysia study states that government
wages rose less rapidly than in other sectors. The government sector
generally consists of nontradables. Thus changes in sectoral wages
seem to have generally supported structural adjintment aims, and also
corroborate the trends in employment
Six of the country studies also examine trends in the
formalfinforrnal wage differential. Here wage trends are likely to
reflect not only goods prices, but also the effects of crowding
discussed earlier. The country studies suggest that the patterns also
depend on institutions in place in individual countries. For example,
the fcrmal sector is generally better abIe to protect itself during
anticipated inflation, provided that institutional mechanisms provide
full compensation for inflation. The informal sector, however, is less
tightly bound by wage freezes, and in periods of unanticipated
inflation informal wages are more closely tied to the goods market. In
countries where the informal sector thrives because of distortions in
the formal economy, Ghana, for example, structural adjustment may
remove rents, and therefore benefit the formal sector.
38 Susan Hor:orz Ravi Kanbur and DipakMazumdar
Crowding seems to have been important in the early 1980s
recession in Brazil, Chile, and Costa Rica, when the informal sector did
relatively worse. In Korea also the formallinformal earnings gap
widened during recessions, probably because of a composition effect
(the formal sector shed the lower paid workers). However, in the
Bolivian hyperinflation and during the Brazilian heterodox
stabilization under the Cruzado Plan, where a price freeze was
combined with strong demand, informal sector earnings improved
relative to formal sector earnings. The Malaysian evidence is
somewhat mixed, as men's and women's wages performed oppositely..
The wages of self-employed men rose faster than employee wages
during the whole period (partly explained by the ircrease in education
of the self-employed). Eniployed women fared better during the
boom, but then their wages fell relative to those of the self-employed
during the recession (the Malaysian results are from earnings
functions, not aggregate wages, unlike the results for the other
countries). The author of the Malaysia study suggests that this
indicates the existence of pockets of women employed in the informal
sector that did not particpate in the boom affecting the rest of the
economy.
Finally, a couple of studies mention skill differentials. These
narrowed during inflation and the first structural adjustment period in
Chile (1970-76) and never recovered. In Egypt white collar/blue
collar differentials narrowed throughout the oil boom and continued
to narrow through the recession, perhaps due to slower growth in the
public sector.
Some further information on earnings can be obtained from
analyzing earnings functions (table 1.8). Altogether six of the case
studies present- eamings functions, of which four have separate
functions for years before and after the onset of structural adjustment
(Bolivia, Costa Rica, Kenya, and Malaysia). Two other African country
studies present earnings functions for a single year (Cote d'Ivoire and
Ghana), although the Ghana study divides the sample by length of job
tenure, which is an ingenious way to get some information on changes
in the labor market Thus, in five cases (that is, all but the Cote
d'Ivoire), one can get additional information on changes in sectoral,
malelfemale, and formallinformal differentials purged of the effect of
LaborMarkets in an Fra ofAdjustment:An Oveview 39
changes in 'human capital characteristics within sectors. Such a
correction is important during a period of large structural change (see,
for example, Lavy and Newman's 1989 work on the Cote d'Ivoire), or
when participation rates change greatly.
The two Latin American countries exhibit changes in the earnings
functions, both rather similar. The coefficient of determination (R2)
falls in both cases,, and the size of the coefficients of characteristics
associated with the formal sector declines, particularly in Bolivia
(namely, the coefficients on education and experience, and for Bolivia
being male, being married, and working in the formal sector). One
possible explanation is that labor market institutions, and hence
segmentation, were perhaps strongest in Latin America, and have
weakened somewhat during adjustment (this was an explicit aim of
Bolivia's adjustment program).
Earnings functions for the other countries also show changes
consistent with adjustment: in Ghana the returns to urban location,
working in the service sector, and being a union member declines, and
the mining coefficient increases. Kenya is an exception. The authors
argue that Kenyan labor markets did not adjust, and the coefficients
on formal sector characteristics (age, being male, working in the
formal sector, and working in Nairobi) increase.
For Ghana, earnings functions suggested a relatively well-working
labor market, which complemented the findings from the few trend
data available- Men's and women's hourly earnings were not too
dissimilar (although total eamings differed), first and second jobs had
similar hourly earigs (except in agriculture), and there was a
premium for seasonal labor.
For the CDote d'Ivoire data were available for two consecutive years,
including some repeated data on the same individuals. Participation
and employment transition equations were estimated rather than
earnings functions. The panel data showed that labor market
transitions generally were toward sectors favored by adjustment,
particularly agriculture, and that within manufacturing there was a
shift toward the infohnal sector. As regards the probability of leaving
employment, this was higher for women, lower for the services, higher
for construction, and lower for the more educated. Likewise higher
levels of education had a positive effect on the probability of entering
Table 1.8 Changes in Earnings Functions Over Time: Coefficients for Selected Independent Variables, Selected
Years
1Indepenident variable
Employment aomen Other variables
Coanfry Year chiaracteristic ScAtooling Experienice Experience2 (dwnmy) included R2
Bolivia 1981 .122 .053 -.000640 -.327 unmarried, informal, .478
1988 ,0951 .0322 -.000308 -.234 3 cities .253
Cdtc d'lvoirc 1985 .207 .053 -.082 -.002 nationality, yeOar .585
technical educational
Costa Rica 1980 paid workers .1348 .0505 -.00063 .3318 n.a. .472
all workers .1325 .0464 -.00055 -.3217 n.a. .402
1988 paid workers .123 - .03911 -.00045 -.1945 n,a. .356
all workers .113 .03911 -.00045 .1945 n,a. .251
ahana 198748 tenure > 5 ,1 10a .011 -.0001 -.268 region, sector, formal, .278
tenure c 5 .085a .027 -.0003 .272 urban, others .281
Kenyab 1977-78 0073 .0759c -.0O08c -.1188 occupation, city, .413
(.0033S2) age, education
19-6 .0222 .0784c *,0008c -.1587 .537
(.0057S2)
Malayslac 1970 Malay men ,142 .093 -.0012 n.a. n.a. .451
Malay women .147 .071 -.0011 n.a. n.a. A21
Chinesc mcn .139 .110 .001 n,a. n.a. .521
Chinese women .133 .680 -.0007 n.a. n.a. .437
1987 Malay men .171 I11 -.0014 n.a. n.a. .439
Malay wormien .196 .110 -.0016 n.a. n.a. .421
Chinese men .153 .098 -0102 n,a. n.a. .437
Chlnesc womnn .152 .076 -.0009 n.a. n.a. .326
n.a. = not available
n. Secondary school dummy.
b. Urban only,
c. Age.
Sources: Country studies, except Cdtc d'[voiro source is van der Gaiag anid Vijvorberg (1989).
Labor Markets in an Era ofAdjustment: An Overview 41
employment, in contrast to the results for Asia and Latin America,
where structural adjustment often adversely affected earnings and
unemployment for the educated. Unemployment could be relatively
persistent: of those seeking employment in 1985, 81 percent were still
unemployed in 1986, although 42 percent of the original group had
stopped looking. Finally, the study had somne interesting results on the
effects of crop price indexes on work behavior in rural areas.
Increases in these indexes had a positive effect on work supply both
for those who were working and in school in the first of the two survey
years, but a negative effect on work supply for those in fuill-time
education in the first year. In other words, crop price increases could
increase effort, but not at the expense of interrupting human capital
acquisition, an interesting finding.
The use of earnings functions is obviously a useful direction for
firther work on labor markets and adjustment, and in this respect the
technique of dividing the sample (as used in the Ghana study) seems a
promising wrt of teasing out trends from a single cross-section of
data, which zi"At be particularly useful for African countriea
Labor Market Institutions
Two issues papers deal with labor market institutions, Devereux's
on wage indexation (vol. 1, chapter 4) and Nelson's on political
economy issues (vol. 1, chapter 7). The latter paper focuses on the
effect of - unionism, both private and public, on labor market
flexibility. It also discusses economic and political factors that affect
how militant or cooperative labor movements are likely to be.
Nelson argues the existence of theoretical reasons that explain why
unions in developing countries might be more militant than in
developed ones. The relationship between union organization and
militancy is an inverted U-shape: weak unions exhibit a low level of
militancy, and very strong centralized unions are also less militant as
they can no longer consider only sectoral gains. Developing country
unions fall in the middle, with some strongly organiZed sectors, but no
strong central union body. Unions in most developing countries do
not fit the corporatist model, where wage gains are traded off for
better employment security and where labor may take account of the
macro impact of sectoral wage demands. Another feature of unions in
42 Susan Horton, RaviKanbur, andDipakMazumdar
developing countries is the greater role of public sector unions due to
the greater share of public sector employment in total formal
employment. A feature of the public sector is the greater difficulty
experienced in laying off workers and the large severance payments
offered.
Economic factors may affect labor's intransigence: they tend to
show more concem for wages during upswings and more concem for
employment protection during downswings, although unions
foreseeing times getting worse may try to grab what they can early in
the downswing. Political factors also matter. authoritarian regimes tend
to use coercion more than democratic regimes, with some exceptions
on both sides. The stage of the electoral cycle matters, as does labor's
role in the political and party process. Labor may be attached to one
party in a polarized system, or have access to more than one party in a
more open system, or be largely excluded from the political arena.
Likewise the regime's degree of stability matters, with new
democracies in particular being susceptible to the revolution of rising
expectations. Nelson makes the important point that successful
adjustment in the long run not only requires investor confidence in
the government's long-run ability to fulfill its promises, but also the
confidence of the labor movement. The degree of equity in a society
may be an important ingredient in sustaining such confidence.
The Latin American country studies dwell at length on labor
market institutions: unions, indexation, minimum wages, legislation on
benefits and job security, and segmentation. For the African countries
these institutions receive less coverage in the country studies, altho-gh
they do exist. As the Kenya study shows, however, it is one thing for
the institutions to exist, and another for them to be effective, and their
force tends to be weakened by the highly elastic labor supply to urban
areas in Africa. It is also likely that the much lower proportion of
urbanization and of for, nal sector employment makes a difference.
The Asian countries have some similar institutions (two-year wage
contracts in Malaysia and the same kind of long-term
contract/temporary labor division in Korea as in Brazil). However, the
role of unions in Asia is dearly very different from their role in Latin
America and Africa.
Labor Markets in on Era ofAdjunent: An Overview 43
The five Latin American country studies provide an interesting
contrast in terms of the alleged effect of labor market institutions in
causing rigidities in the labor market. In three of the countries
(Argentina, Brazil, and Costa Rica) the institutions remain strong
despite the economic crisis, whereas in the other two (Bolivia and
Chile) they have been suostantially weakened and/or dismantled. Some
of the country authors criticize these institutions. For Argentina and
Brazil they argue that they impeded adjustment and labor market
mobility, and in Chile they receive partial blame for the painful nature
of the recession and ensuing high unemployment. At the same time
the Bolivian and Costa Rican cases are interesting counterpoints. In
Costa Rica labor institutions survived relatively unscathed, for
example, over 500 minimum wages are legislated, and are generally
enforced, which did not prevent moderate adjustnent. In Bolivia much
labor legislation was dismantled and large-scale. labor shedding
occurred without as yet strong recovery. To some extent it sterns that
labor market institutions are often a symptom of underlying political
and economic difficulties, which make adjustment difficult, and the
institutions are unfairly blamed for causing problems.
The Brazil study describes labor market institutions in some detail.
Unions are very strong (in the form in which they reemerged during
the democratization period from the late 1970s onward), and are
linked to political parties along the lines of the corporatist state
discussed by Nelson. They combine strong plant-level organization
with a previously legislated strong centralized structure, which allows
them to transmit bargains struck at the best organized plants to
national level. Wage indexation is perhaps the most sophisticated in
Latin America, with monthly adjustments. Job security legislation used
to be an important hindrance to mobility, but the setting up in 1964 of
a fund (to which employers contribute) to provide severance pay has
eased the problem. Tradables predominantly hire formal sector (that
is, signed contract), unionized workers, whereas nontradables hire all
types of workers, formal and informal, unionized and nonunionizedL
Argentina has many of the same institutions. The author -links
union strength to inward-oriented economic policy, since the
oligopolistic nature of employers demands an equally centralized
representation for labor. The author also mentions a compulsory wage
44 Susan Horton, RaviKanbur, andDipakMazumdar
policy, whereby bargains struck by the unions are obligatory for all
firms, which he argues harmed small finns. One difference from the
Brazil case is that the main exportable in Argentina is food, and
unionization is therefore concentrated in nontradables or importables.
This arguably has been a major hindrance in changing the relative
price of tradables and nontradables.
One difference in Costa Rica is that although legislation is equally
strong, unions are relatively weak, having been broken in an
unsuccessful face-off with Standard Fruit in the 1970s. Wage
indexation in Costa Rica, far from being an impediment to desirable
relative price changes, is given much of the credit for allowing a real
wage decline at a critical point following devaluation. Since indexation
was imperfect, real wages fell, but by an apparently impartial
mechanism. This tactic, however, can only be used infrequently, and
Brazfl, for example, is no longer able to make such gains from
unanticipated inflation.
Two Latin American countries undertook major labor market
reforms. Chile between 1973 and 1975 eliminated unions and job
security and removed much of the force from minimum wages,
benefits, and wage indexation mechanisms (the government actually
cheated on the price index used for wage indexation). However, the
author argues that lack of labor legislation during 1973-79 was
detrimental to growth because employers feared that the law, once
reinstated, would be unduly favorable to labor. Bolivia, the other Latin
American severe adjustment case, likewise removed similar institutions,
with the exception that wage indexation had never been particularly
important and had not survived the hyperinflation as an institution.
Job tenure was ended and job security reduced, thus allowing labor
shedding- The government stepped out of previously centralized wage
bargaining. In both Bolivia and Chile the public sector shed a
substantial amount of labor, equal to 25 percent of Bolivia's public
sector labor force and 3 percent of Chile's total labor force (the
author does not specify as to whether total urban or total urban plus
rural is meant).
Comparisons between the Latin American countries in terms of the
success of adjustment are instructive. Contrasting, for example, the
relatively successful adjustment in Costa Rica and the problematic one
Labor Markets in an Era ofAdjument An Overview 45
in Bolivia, evidently dismantling labor institutions is neither necessary
(Costa Rica) nor sufficient (Bolivia) for successful adjustment.
Another interesting comparison is between Brazil and Costa Rica. In
Brazil large political-economic tensions exist, such that consensus over
the division of output is lacking, which causes continual inflationary
tendencies (tensions that similarly pushed Bolivia over the brink into
hyperinflation). Although wage indexation has sometimes been
blamed for perpetuating Brazil's inflation, it is more a symptom of the
defensive ability of one of the groups engaged in underlying conflict
In Costa Rica, by contrast, a higher degree of social consensus allowed
a union-backed president to undertake some of the painful initial steps
toward successful adjustment, in which wage indexation actually
helped the process.
The Asian countries also have institutional structures in the labor
market The Korean govemment has followed a highly interventionist
policy with respect to unions. The right to strike was banned in 1971
and only recently reinstated, and unions need government permission
to undertake collective bargaining. The author argues that wage and
productivity trends and their consequent effect on unit costs has been
crucial in Korea's export success. In this respect the govermnent was
heavily involved in ensuring that 'wages did not get ahead of
productivity, and at the same time that workers did share in the frits
of higher productivity. Increasing union autonomy and increasing
strikes in the late 1980s may herald a change in the so far virtuous
productivity and wage nexus in Korea.
In Malaysia union power is similarly limited. The level of
unionization is low, less than 25 percent in manufacturing, and unions
are banned in some sectors. Paradoxically unions are strongest in the
plantation sector, where wages stagnated in the 1980s. Malaysia has
relatively long (three-year) wage contracts, which may have hindered
adjustment. Unions in Thailand are also weak except in the public
sector. In both Malaysia and Korea the importance of bonuses in
earnings (around 30 percent of pay in Korea and 15 percent in
Malaysia) has been argued to cause flexibility, since earnings and
profits are related- Latin American countries also have bonuses, but
less related to productivity and profits than to Christmas, seniority, and
so on.
46 Susan Horton, Rav Kanbur andDipakMazrsndar
Although studies of Latin American countries frequently blame
labor market segmentation (formallinformal) as a problem, some
kinds of segmentation also exist in the Asian countries. In Korea labor
is divided into permanent, temporary, and casual, and much labor
market adjustment falls upon the casual and temporary workers,
particularly women- Another type of segmentation between large and
small firms is also quite marked in Korea, and small firms tend to pick
up the slack during recessionary periods. Segmentation also seems to
persist over time, although taking the form of a widening gap in the
human capital levels of large as compared to small firms, rather than a
widening of wage differentials.
Fimaly public sector employment and adjustment is a topic worthy
of separate study in its own right The growth of public sector
employment as an initial response to economic crisis is mentioned in
many of the studies (all of the Latin American studies, Egypt, and
Malaysia). The eventual need to shed public sector labor was a
difficult undertakdn& Bolivia, Chile, Costa Rica, and Ghana have bitten
the bullet, Argentina has been unable to; and in Egypt, Kenya, and
Malaysia adjustment took the form of a substantial slowdown in
government hiring. In the latter three countries one consequence
discussed was a rise in educated unemployment, particularly of women
in Egypt and Malaysia, where educated women have few private sector
alternatives. The relative decline in public sector wages observed in
almost all the countries reflects the greater difficulty of adjusting labor
quantity in the public than the private sector.
Consequences of Labor Market Adjustment
Labor market adjustment has consequences for income distribution
and poverty, and on long-mn growth. The country study authors were
asked to consider these, paying particular attention to the role of
women in labor markets.
Income Distribution
As Addison and Demery show (vol. 1, chapter 3), theoretical
discussion of the effects of adjustment on poverty yields ambiguous
predictions. Their paper begins Nwith the standard Salter-Swan account
of expenditure reduction and expenditure switching, and works out
LaborMarerr in an Era ofAdjusznnmeAn Overview 47
wage and employment effects, assuming competitive labor markets.
These wage and employment effects are then fed through a poverty
index, but yield ambiguous predictions.
The rest of the paper examines how these effects are modified by
the introduction of different labor market imperfections. The first case
is where there exists an economywide "quantity rationing"
framework, that is, unemployment can persist. In this case the
discussion of poverty becomes more complicated, since one must
consider poverty among those employed in tradables, those employed
in nontradables, and those unemployed. In this case although a
devaluation may increase poverty because it shifts workers to the
tradable sector, where greater poverty is assumed, and because it
lowers the real wage, it will decrease poverty because of the
unemployment reduction. Thus, ambiguity in predictions persists, but
of a different type thn before
The paper then moves on to discuss partial labor market
imperfections, dividing the labor market into a formal and an informal
sector. The analysis is similar to that by Edwards (1988) and Edwards
and Edwards (vol. 1, chapter 2). The authors consider different types
of wage inflexibility and trace out the consequences for sectoral
employment, wages, and unemployment These are again fed through
a poverty index Ambiguity is again the order of the day, although the
analysis does illuminate the different components.
A third variant is where barriers exist to entry into the formal labor
market. Here Addison and Demery (voL 1, chapter 3) argue that an
expenditure switching policy is quite likely to reduce poverty if
barriers to entry into nontradables or tradables exist
The fourth and final case is where labor market imperfections exist
in both sectors, and the authors distinguish between unemployment
and employment in informal tradables, formal tradables, informal
nontradables, and formal nontradables. They follow through the real
wage and labor allocation consequences of expenditure switching, and
again feed them through the poverty index. They conclude that the
effects of switching under these assumptions seem to be the most
promising as far as poverty reduction is concemed.
Tracing the effect of adjustment on poverty and income
distnrbution empirically is no easier than doing so theoretically.
48 SusanHorton, RaviYCanbur, and DipakLMazumdar
Asking the counterfactual question as to what happened during
adjustment as compared to what would have happened otherwise is
difficult, as many countries were on unsustainable courses. The data
available also affect the conclusions that one can reach. It is usually
more difficult to obtain information on overall economywide changes
in income distribution from nationwide income-expenditure surveys
than to obtain results on the urban distribution of earned income from
labor force surveys. However, if real wages fall by more than GDP and
urban-rural differentials change, then the latter data only tell part of
the story.
We focus here on relative earnings distnrbution. Several studies also
document increases in poverty, unsurpising as a consequence of
economic crisis. For Africa almost no time series data exist with which
to make comparisons. The Kenya study does cite 7NICEF's finding
that the share of the bottom 10 percent declined. For Egypt no
distnrbution data are available after 1981/82. Changes in urban-rural
income differentials are of great interest in the case of Africa and are
the focus of studies elsewhere (Jamal and Weeks 1987), but country-
studies here lacked the data to examine the issue.
In Latin America income distnbution is a key issue related to the
political economy of the economic growth process, and all the studies
provided data. Brazil's income distnbution has long been of interest
given that inequality increased during the long boom "economic
miracle" period between 1967 and 1974, when there was a type of
structural adjustment as the economy became more open. Some
improvement in income distribution is evident between 1974 and
1981, with a worsening duiing the recession and stabilization (1981-
85), and since then a slight recovery- One interesting finding is that
interregional equality increased during structural adjustment, which hit
harder at the more affluent urbanized south than the more rural
northeast
For Chile the pattern was somewhat similar, but more exaggerated,
with a sharp increase in 1974-76 accompanying the start of
adjustment, the Gini remaining constant during 1976-79, increasing
again in 1979-84, and since then decreasing slightly, but to a level
much higher than at any time during 1960-74. It is not surprising that
distribution worsened so much, given the massive cuts in real wages
LaborMarkets in an Era ofAdjustrment: An Overview 49
and the very high unemployment levels. The measured changes may
be offset somewhat by changes in social expenditures. In Argentina
income distribution also worsened during the stop-go cycles (although
the only data available are for income earners in Buenos Aires during
1974-88). The top two deciles gained at the expense of all others.
For Bolivia and Costa Rica data are more scanty and knowing
exactly what happened is harder. In Costa Rica inequality may have
increased between 1971 and 1983 (before adjustment), but after the
onset of adjustment different data sources give conflicting trends.. For
Bolivia the data are also not very good, but suggest a possible
improvement between 1982-85, when informal sector wages rose
relatively during the hyperinflation, but by 1988 distribution had
reverted back- to 1982 levels.
In Asia, income distribution may have improved in Malaysia and
worsened in both Thailand and Korea In Malaysia resources were put
into agriculture, including food agriculture, whereas in Korea policy
focused for at least some of the period on heavy industry, and in
Thailand little was done about. the problem of urban primacy
(concentration in Bangkok)-
Women and Labor Market Adjustment
Much of the literature on women and structural adjustment has
concentrated on the effects of structural adjustment on women. Collier
and others (vol. 1, chapter 6), using evidence from Africa, examine the
opposite issue, namely, how women's economic mobility may affect
the success of adjustment. They argue that women face constraints not
only in the labor market and in access to education, but also in credit
markets, which may affect adjustment. In particuIar, women in Africa
are frequently concentrated in food production. The authors present
three possible cases relevant to adjustment. Food may be a tradable, in
which case its output should expand with adjustment; it may be a
nontradable, in which case output should contract; or it might be
nontradable in rural areas but tradable in urban areas, in which case
food marketing (again frequently a female preserve, at least in West
Africa) would need to expand. If food crops are to contract, this
requires a reallocation of women's labor into other activities, and if
they are to expand, this requires women's access to credit. In either
50 Susan Horton, Ravd Kanbur, and DipakMazumdar
case, constraints on women's flexibility will hinder the success of
structural adjustment.
Collier and others therefore urge that government policies should
focus on relaxing constraints to women's economic activities. AMother
reason cited in favor of this strategy is that it also improves household
income security if higher women's incomes offset the loss of men's
jobs in the formal or government sectors during adjustment, although
they do not consider the potential costs involved, such as women's
responsibilities for children.
The paper by Collier and others also discusses women in South
Asia, again focusing on women as participants in, rather than victims
of, structural adjustment It deals with both rural and urban activities
of women, and draws somewhat on the earlier experiences of women
in export-oriented industries in East -Asia Bardhan sees structural
adjustment as potentially altering the existing U-shaped pattern of
female labor force participation with education: in South Asia women
tend to participate either with very low education in menial and low-
productivity activities, or in high-skilled, high-education activities. The
author argues that adjustment may increase the demand for labor-
intensive industry output, requiring women workers with medium
education, with resulting beneficial effects on reduced fertility and
increased incentives for female education. Adjustment may also
involve costs for women, such as those where the male family
members or the whole family migrate,. and the costs imposed
particularly on women's time when social infrastructure deteriorates-
Like Collier, she sees a role'for government in relaxing the constraints
on women's activity. Labor market legislation aimed at protecting
women has ended up tending to exclude them from the formal sector.
Bardhan foresees benefits to women in selective deregulation of some
sectors in India, such as electronics.
Another aspect of the paper by Collier and others focuses rather
more on the effects of structural adjustment on women. In Latin
America, studies on women seem to focus mainly on labor force
participation, and little information is available on trends in relative
earings. Women's labor.force participation has been increasing,
partly due to sectoral shifts, in particular, increased employment in the
service sector, but largely due to higher participation within sectors.
Labor Markets Ui an Era ofAdjusiment An Overview Sl
The participation increases vary somewhat across countries. The
authors undertake econometric analysis for Chile, which suggests that
unemployment that accompanies structural adjustment does not have
differential effects on discouraging female and male labor force
participation. One interesting avenue they suggest for future work is to
examine how increased female participation fits in with the trend in
much of Latin America toward increased informalization of the labor
force.
The country studies concentrate more on the effects of structural
adjustment on women. As Collier and others argue, the effects are
likely to depend on the preceding sectoral distribution of women
workers and on the effect on participation rates. However, the
likelihood exists that women workers' more tenuous attachment to the
labor force means that they are more likely to lose jobs during periods
of labor shedding. The countiy studies do not give a single story,
although there seems to be a lot of evidence of adverse impacts, but
the data are not very complete. Even for the United States, where data
are available, understanding how male/female wages, for example, had
changed over time due to changes in female labor force participation
was difficult. For the deve!oping countries female labor force
participation has exhibited trend changes plus cyclical responses due
to crisis. Tracing the effects on women's welfare is even harder if most
women live in households with men. Although the effects on female-
headed households are less ambiguous to interpret from the data, this
was a topic well beyond the scope of the country studies.
The Ghana study documents that women suffered rather more
from structural adjustment than men as they were concentrated in the
informal sector, which tended to absorb excess labor. Women are also
predominantly in food crop agriculture, whereas resources have gone
instead to cash crops. In C6te d'Ivoire, insofar as education had a
positive effect on the probability of remaining in employment or of
entering employment, and women tend to have less education, they are
likely to have faced disadvantages. The Egypt study documents an
adverse effect on women due to the lengthening queue for
government employment, and the more limited private sector
alternatives available to women.
52 Susan.fIorton, RaiKanbur, andDipakMazmdar
In Bolivia the male/female differential fell between 1981 and 1987
as measured from earnings functions, although aggregate data suggest
the opposite (the difference is perhaps explained by changes in
participation rates). Although anecdotal evidence suggested that labor
shedding from the, formal sector-was to the detriment of women, who
are more costly workers in terms of benefits, this may have been offset
by much of the employment loss being focused in mining, a male-
dominated sector. In Costa Rica the male/female earnings differential
increased during the crisis and decreased thereafter, which the authors
attnrbute to rising female participation during the crisis (added worker
effect), where the female entrants were less well quaified. In both
Bolivia and Chile the emergency employment schemes explicitly
targeted male workers, at least initially, and in Chile public sector
hring in the early part of the crisis also favored men.
In Malaysia some evidence suggests that women last ground during
the recession due to the firing of labor in a weaker position in the
labor market; however, a trend increase in female wages is evident over
the 1970s and 1980s. The relative earnings of Malay women in
particular increased between 1970 and 1984, and the returns to female
education and experience rose absolutely and relative to the same
returns for men. However, these gains were all reversed in the 1984-
87 recession. Nevertheless, Malaysia differs from some of the other
countries studied in that women are a higher proportion of wage
employment th;an of self-employment, and are concentrated in some
export industries, such as electronics. In Korea women are at a
disadvantage, crowded into low paying, white collar sectors, and
providing a disproportionately high share of family workers, the most
disadvantaged group in the labor force. Female participation rates are
also surprisingly low in Korea compared to other East and Southeast
Asian countries. Women also tended to lose out in the recession.
Whereas male employment shifted continuously toward the permanent
category, this proportion declined for women during the recession.
Effects on Long-Rum Growth
Most of the issues papers focus on demand side effects of
adjustment and the labor market. Buffie's (vol. 1, chapter 5), by
contrast, highlights the supply side consequences of fiscal contraction,
LaborMarkets in an Era of Adjustment An Overvew 53
and hence the impacts on long-run growth. Demand side
complications are abstracted from by assuming that the economy is
small and open. Two traded goods, agricultural exports and
manufactures, are produced using labor and capital. Manufacturing
also requires an intermediate input, which is supplied by the public
sector. Labor employed in the public sector and in manufacturing is
paid a higher than competitive wage, and the rest of the labor is
underemployed in agriculture. Buffle assumes a fixed wage
differential between the modem and the informal/agricultural sectors.
Capital accumulation dynamics are also mcdeled.
Human capital is modeled by distinguishing between skilled and
unskmilled labor. Skilled labor growth is determined by human capital
investment by the government. If factors are complementay, then the
productivity of unskldled labor declines when investment is cut, as does
the productivity of capital. Overall, Buffie shows that disinvestment in
human capital leads to capital decumulation. The paper suggests two
broad policy lessons. First, productive government investments in
human capital should be protected, which requires broadening the tax
base. Second, a more gradual approach to adjustment is likely to entail
fewer adverse impacts on productive investments vital for long-mn
growth.
To some extent the topic of adjustment and long-rn growth is a
difficult one to study empirically, since many countries are still
grappling with short- and medium-term issues, but some of the studies
provide information on investment, in particular, human capital
investment, as discussed by Buffie. The Kenya and COte d'Ivoire
studies discuss falling investment, but do not blame labor markets. The
Argentina study throws the blame for stop-go cycles onto the labor
market's inability to allow prices of tradables to rise relatively in a
srstained way, thereby harming long-run growth. Similarly, in Chile a
lack of labor legislation and fears of a return to previous laws that
favored labor are assigned the blame for lack of investment.
As regards human capital investments, the Costa Rica study
documents a sharp drop in school enrollment during the crisis,
especially at the secondary and technical levels, with likely adverse
effects on growth and distribution. By contrast, no such effect was
predicted from cross-section regressions for the CBte d'Ivoire. In Asia
54 SusanHorton,RaviKanbur, andDipakMazundar
where short-run problems of adjustment have been largely solved, the
studies had more room to focus on long-run issues. The Malaysia and
Korea studies examine changing returns to education, and the Thai
study examines potential labor market skill mismatch issues.
Conclusion
This overview has summarized theoretical predictions and country
study experience on two important topics related to labor markets and
adjustment. First, how well have labor markets functioned, and have
they assisted or impeded macro adjustment efforts? Second, what were
the effects of some of these adjustments on the labor market?
With respect to the issue of labor market functioning, labor markets
have at least three allocative functions: they match workers to
employment in such a way that overall unemployment levels and real
wages matter, they allocate workers between sectors, and match worker
skills to job requirements so that relative wages and employment
matter, both for economic sectors and for skill categories; and they
provide incentives for intertemporal allocation of resources,
specifically for human capital accumulation in edlucation and firm-
specific training. Applying these three criteria to the often descriptive
country studies to assess how well or how badly labor markets
performed is not easy. By and large individual country authors argue
that the labor markets performed well; although authors of the studies
for the big three Latin American countries, Argentina, Brazil, and
Chile, were more critical.
Theory suggests that labor market rigidities are only one of three
possible reasons for unemployment. With the exception of Chile, the
countries have not had prolonzged unemployment despite severe
recession, however, cyclical increases have occurred. This fits with the
presumption that in developing countries without unemployment
insurance schemes, unemployment is not an option for primary
household earners unless the household is unusually wealthy. The
evidence on real wages casts considerable doubt on theoretical
concerns about aggregate real wage rigidity and labor market
inflexibility as a hindrance to adjustment. Real wage declines have
been dramatic, and often far greater than the fall in GDP. For some
Labor Markets in an Era ofAdjuslment: An Overview 55
countries the real wage declines may have been excessively large and
led to a fall in domestic demand, which inhibited recovery.
With regard to the sectoral employment shifts, these have generally
been in the desired direction, that is, toward tradables, although this
has generally meant that agricultural employment has increased
relatively and manufacturing employment declined in all but the most
successful countries. Shifts of employment into services and
comnerce are, however, indicative of weak GDP growth, and hence
growth of labor demand. Sectoral wage changes have aiso been
largely in the appropriate direction, although little information is
available on agricultural wages. The decline in relative government
wages is one factor causing relative wages in nontradables to decline.
Finally, on the intertemporal aspect, the evidence is a little more
mixed. In Costa Rica the evidence showed that the recession had
induced decreases in school enrollment, whereas in C6te d'Ivoire
econometric results suggested that increases in crop prices, which
would help adjustment, would not lead to parents pulling their
children out of school. Earnings fimctions for Bolivia. Costa Rica, and
Malaysia showed that returns to all formal sector characteristics
including education and experience declined during adjustment, and
in that government relative wages declined universally, and
government tends predominantly to hire the more educated, this
would decrease the incentives to acquire schooling. The country
studies did not discuss another human capital issue, namely
intemational migration, although for at least three of the countries-
Cote d'Ivoire, Egypt, and Ghana-this was important.
The country. studies also explicitly discussed labor market
institutions, thought to be a source of rigidity. One possible
interpretation is that where these institutions lack binding force,
whether because of elastic labor supply (Africal or weak unions (Asia
and perhaps Costa Rica), they were not perceived as obstacles to
adjustment. Nevertheless, dismantling of the institutions and
weakening of the unions as in Bolivia does not seem to be sufficient to
ensure recovery, in that country imperfections in the functioning of
the capital market seem to bear at least part of the responsibility for
poor growth. The authors also argued here that labor market
institutions in Latin America often receive the blame, whereas they are
56 Susan Horton, Ravi Kanbur, and Dipak Mazumdar
only the symptoms of underlying political economy problems
detrimental to growth.
Turning now to the second broad topic, the outcomes of labor
market adjustment, the authors had some difficulties in separating how
far outcomes were due to structural adjustment, how far due to
recession, and how far due to pre-existing trends. Severe adjustment,
as in the case of Chile with high unemployment and sharp falls in real
wages in an economy where urban employment predominates, can be
very adverse to income distribution. Perhaps Brazil's worsening
during the 1964-79 structural change period has some parallels, as
does Korea's heavy industry phase. That is, unless countries make
explicit provision for poorer groups, for example, the emphasis on
food crop agriculture in Malaysia, structural change can worsen
income distribution, although some of the changes, such as improved
rural-urban relative income and possible improvement in
informal/formal relative income, night militate in the opposite
direction. Country-specific factors-success of indexation, wage and
price freezes-also affect distnbution. The effects on distribution also
depend on the level of the GDP. No data are available for Africa to
test this hypothesis, but it seems plausible that improving rural-urban
terms of trade and abolishing rents from prce distortions as part of
adjustment programs could improve income distnbution nationally.
The effects on women might be somewhat country specific,
depending whether women were in tradables or notL but women are
likely to face adverse effects of the employment shrinkages in some
sectors due to their weaker attachment to the labor market The
country studies generally confirmed this. Finally, the effects on long-
term growth were adverse, but not directly attributable to labor market
malfunctioning.
Where should one go from here? One issue is that the apparently.
benign conclusion that labor shifted into tradables masks that in
response to structural adjustment, labor has moved in the direction
opposite to that usually associated with economic development. Labor
has shifted back into agrculture, out of manufacturing, and out of the
public sector, although one might argue that this later sector was too
large given the level of development reached. Recession plus
adjustment has also resulted in an increase, in informalization,
Labor Markets in an Era ofAdjustment: An Overview 57
increased use of casual labor, decreased worker benefits, and declines
in skill and possibly education differentials. These trends are observed
even in the most successful adjustment cases in Asia. Developing
countries have long resisted being relegated to the role of primary
producers in the internai3onal economic order, and it is unlikely that
structural adjustment entailing further shifts of labor into agriculture
would be highly sustainable.
As regards possible further research, country study and some issues
paper authors pointed the finger of blame for adjustment problems
onto the capital market and possible. price rigidities in the output
market. Another possibly fruitful topic is that of the role of labor
market institutions, unions, and the political economy; something
worth examining before launching into a wholesale advocacy of
dismantling such institutions- Finally, as in all empirical research,
better data are needed. One useful step would be to improve
international collation of labor force statistics, clearly separating the
results from household surveys from those of establishment surveys.
Another would be to encourage further analysis of, and increased
accessibility to, labor force surveys, which tend to be more expensive
to analyze, but arguably yield more reliable results.
References
Dutt, A. K. 1984. "Stagnation, Income Distribution and Monopoly
Power." Cambridge Journal of Economics 8(1): 25-40.
Edwards, S. 1988. "Terms of Trade, Tariffs and the Labor Market
Adjustment in Developing Countries." World Bank Economic
Review 2(2): 165-185.
Fallon, P. R., and L. A- Riveros. 1988. 'Macroeconomc Adjustment
and Labor Market Response: A Review of the Recent
Experience in LDCs." Washington, D.C.: World Bank. Draft,
processed.
Fields, G. 1990. "Labor Market Policy and Structural Adjustment in
Cote d'Ivoire." Ithaca, New Yorlc Cornell University. Draft
CGhai, D. 1987. Economic Growth, Structural Change and Labor
Absorption in Africa: 1960-85. Discussion Paper No. 1.
58 Susan Horton RaviKanbur, andDipakMazwndar
Geneva: United Nations Research Institute for Social
Development.
Horton, S., R. Kanbur, and D. Mazumdar. 1988. "Labor Markets in
an Era of Adjustment: A Project Proposal." Washington, D.C.:
World Bank, Economic Development Institute. Processed.
ILO (International Labour Organisation). 1987. World Recession
and Global Interdependence: Effects on Employmen4 Poverty
and Policy Formation in Developing Countries. Geneva: 0LO
World Employment Program.
Various years. Yearbook of Labour Statistics.
Geneva: ILO.
. 1989. Yearbook of Labour Statistica Geneva: ILCO.
Jamal, V., and J. Weeks. 1987. Rural-Urban Income Trends in Sub-
Saharan Africa. World Employment Programme Labor
Market Analysis and Employment Plamning Working Paper
No. 18 (WEP 2-431WP.18). Geneva: ILO
JS'JPA (Jobs and Skills Program for Africa). 1988- Africa
Employment Report 1988. Addis Ababa: ILO.
Johnson, 0. E. G. 1986. "Labor Markets, Extemal Developmegts,
and Unemployment in Developing Countries." Washington,
D.C.: IM Staff Studies for the World Economic Outlook
Lavy, V., and J. Newman. 1989. "Wage Rigidity: Micro Evidence on
Labor Market Adjustment in the Modem Sector." World
Bank Economic Review 1(1): 97-111X.
Lindauer, D. L., 0. A. Meesook, and P. Suebsae.ng. 1988.
"Government Wage Policy in Africa: Some Findings and
Policy Issues." World Bank Research Observer 3(1): 1-26.
Ramos, J. R. 1980. "The Economics of Hyperstagflation:
Stabilization Policy in Post 1973 Chile.' Journal of
Development Economics 7(4): 467-88.
Riveros, L 1989. "Recession, Adjustment and the Performance of
Urban Labor Markets in Latin America." Washington, D.C.:
World Bank. Processed.
LaborMarkets in an Era ofAdjustment.-An Overview 59
Taylor, L. 1988. Varieties of Stabilization Experiences Towards
Sensible Macroeconomics in the Third World. Oxford, UK.:
Clarendon Press.
Tokman, V. E. 1984. "The Employment Crisis in Latin America?"
International Labor Review 123.
van der Gaag and Vijverberg. 1989. "Wage Determinants in C6te
d'Ivoire: Experience, Credentials, and Human Capital."
Economic Development and Cultural Change 37: 371-381.
Wong, P. K 1985. "Economic Development and Labor Market
Changes in Peninsular Malaysia." Working Paper No. 12.
Kuala Lumpur and Canberra. ASEAN-Australia Joint
Research Project
2
ARGENTINA
LuisA. Riveros
Carlos E. Sdnchez
Poor growth and macroeconomic imbalances have characterized
Argentina's recent history. In combination with long-mun stagnation,
the country has suffered chronic higb inflation and deep cyclical
fluctuations compounded by intractable balance of payment crises.
Attempts to stabilize the economy and achieve a structural adjustment
to restore sustained growth and basic balances have failed due to both
a fragile political climate and inconsistent policies. The lack of ad-
justment has also been partly due to anticipated negative short-run la-
bor market outcomes. In turn, the existence of persistent economic
imvalances over the long run has negatively affected wages, employ-
ment, and income distnbution.
This chapter analyzes the performance of Argentina's labor mar-
kets in recent years. It points out both how poor economic perfor-
mance has affected labor market outcomes and how anticipated short-
run costs have hindered reform efforts.
Long-Term Economic Trends and Short-Term Adjustment
Policies
Global economic trends in Argentina have produced contradictory
quantitative results and forced growing state intervention in the labor
market.
The authors gratefu2lly acknowledge comments on earlier drafts by M. Faig, R.
Newfarmer, S. Horton, R. Paredes and seminar participants at the World Bank, the
University of Chile, and the University of Warwick, as well as the efficient research
assistance of 0. Giordano.
6C
62 LUis A. Riveros and Carlos E. Sdnchez
Macroeconomic Policies, Economic Organization, and Growth
An understanding of the trade and macroeconomic policies
Argentina foUlowed after the Great Depression helps explain the
country's poor economic performance. During 1860-1929, the gov-
ernment pursued an export-led growth strategy, which included almost
free trade and appropriation of the benefits of trade according to the
country's comparative advantages. After the Great Depression, the
government adopted an import substitution strategy. After 1945, gov-
ernment policies aimed at expanding domestic markets through over-
valued exchange rates and high import tariffs, which distorted re-
source allocation and thwarted exports over the long-run. The eco-
nomic results of the inward-ornented policies were poorer than those
under the export-led strategy (table 2.1).
Macroeconomic policies that affect variables such as the share of
government consumption, public debt, and the money stock in total
income, as well as commercial (export taxes and import controls) and
exchange rate policies, were mostly responsible during the 1950s for
the increase in the effective exchange rate for imports relative to that
for exports. The resulting price increase of import goods made import
substitution activities relatively more attractive for investment d6ci-
sions, thereby prompting an inefficient specialization of production.
Macroeconomic policies also affected the relative prices of productive
Table 2.1 Strategies and Growth
(real growth in per capita G1DP)
Strategy Years Growth (percent)
Export-led strategy 1900-29 1.5
Inward-oriented strategy 1929-58 0.9
1958-87 0-7
(Entire period) 1900-87 1h0
Source: IERAL data base.
Argentina 63
factors, resulting in a distorted capital/labor mix in production that
affected resource allocation at the sectoral and regional levels (see
Cavallo -1986; Cavallo and Cottani 1986; Cavallo and Domenech
1988; Cavalo and Mundlak 1982; Nogues 1981; Sanchez 1987).
The Argentinian economy can be divided into three sectors: a rural
sector and two urban sectors (Llach and Sanchez 1984). The rural
sector is a net exporter of wage-goods, mainly agriculturaL Therefore,
real wages are inversely related to the incentives to produce exportable
goods for a given exchange rate. The two urban sectors are net im-
porters. They consist of an import substituting sector (which under the
prevailing inward-oriented strategy is both a marginal exporter and a
net importer of inputs and capital goods), and a sector producing
nontradables.
This economic structure led to a tradeoff between the trade balance
situation and the prospects for domestic growth- Long-term economic
growth requires a stable, high real exchange rate, but a high exchange
rate implies higher food prices, and thus lower urban real wages. Since
the level of -urban real wages have traditionally been a key political
variable, the government has tended to hold down incentives to export.
Thus, due to a deliberate policy of overvalued exchange rates (that is,
a policy mix aimed at yielding high real wages in combination with
low real exchange rates), relative prices attained two simultaneous
roles: a mechanism for resource allocation and a distnrbutive device.
This led to contradictory economic targets in a long-term context: mn-
efficient organization of production or improved well-being of urban
workers; a conflict generally resolved in favor of the latter
- A deIiberate policy of overvalued exchange rates was the usual
mechanism the government used to enlarge domestic markets, and
therefore to increase real GDP and real wages in the short run. As
soon as the growth in foreign terms of trade began to decline, a bal-
ance of payments crisis arose, which made a devaluation and a decline
in wages unavoidable. Thus, in the context of a deliberate policy of
overvaluation, FTT can be seen as a determinant of the real exchange
rate (figure 2.1).1 As a result of this policy, Argentina's economic
1. Cavallo and Domenech (1988) have modeled the behavior of the real exchange
rate depending on the foreign terms of trade, taxes on imports and exports, the
income leveL and macroeconomic policy.
64 Luis A. Riveros and Carlos E. Sdnchez
Figure 2.1 Real Exchange Rate, Foreign Terms of Trade, 1962-87
(index, 1970 = 100)
160
140
820 .
. \ ~~~~~~~~~~~~/
100 -~ ~ ~ ~ ~~
V
40
20 _ m ,_ __ arrr-r
1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987
Year
-Rcal echangemate Foreign tmsoftade
Source: Institute of Eccrnomic Studies of Argentina (IEERAL).
history in recent decades can be described as a series of redistrbutive
periods (domestic currency overvaluation as well as increased activity
and real wages) followed by periods of stabilization (devaluation and
reductions in the levels of activity and real wages).
Poltical and Economic Developments of the 1970s
Argentina experienced crucial political and economic changes
during the second half of the 1970s. The years 1963-73 had seen un-
usually high growth rates-of per capita GDP, which averaged 3-9 per-
cent per annum. However, this growth was not a result of specific do-
mestic policies, but of favorable foreign terms of trade, especially
during 1964-66 and 1971-73. Nonetheless, this economic growth
allowed for policies aimed at overvaluing the real exchange rate and at
increasing both the domestic absorption of goods and real wages
(table 2.2, figures. 22 and 2.3). The new (Peronist) administration that
took over in 1973 inherited the combination of a satisfactory eco-
Argentina 65
nomic performance with persistent imbalances associated with the
economy's structural organization.
Table 2.2 Economic Indicators, 1963-87
(index, 1970 = 100)
Foreign terms Real Per capita Real
Year of trade exchange rate GDP wages
1962 94 125 78 83
1963 101 120 75 81
1964 115 100 81 92
1965 114 95 87 98
1966 112 90 87 98
1967 107 112 88 96
1968 106 106 90 91
1969 101 103 96 96
1970 100 100 100 100
1971 114 94 103 104
1972 125 109 105 98
1973 142 S8 109 107
1974 114 82 114 120
1975 95 82 111 115
1976 90 117 109 74
1977 94 105 114 69
1978 92 80 107 72
1979 100 58 113 80
1980 117 46 113 93
1981 125 57 104 85
1982 105 100 97 74
1983 102 122 98 90
1984 111 107 99 99
1985 95 128 93 81
1986 81 102 96 76
1987 70 105 97 66
1988 82 107 93 63
1989 88 123 87 50
Source IEERAL data base.
66 LuisA. Riveros and Carlos E. Sdnchez
Figure 2.2 Real Wages and Per Capita GDP, 196247
(index, 1970 = 100)
120
110
90
60
50.
1963 196S 1967 1969 1971 1973 197519'77 19U79 1981 1983 1985 1987
Year
---- - RcalwSgs R per otpia GDF
Source IEERAL
Figure 2.3 Foreign Terms of Trade and Real Wages, 1962-87
(index, 1970 = 100)
160
140-
120
60.
1963 196S 1967 1969 1971 1973 1975 1977 1979 1981 193 1985 1967
Rcal wages Foxeiptaus oftru
Source: IEERAL.
Argentina 67
The foreign terms of trade were still changing favorably in 1973-
74, and the Peronist administration continued to use redistributionist
policies. Achievement of increasing real per capita GDP and real
wages at the cost of deteriorating real exchange rates was still feasible.
However, the increase in oil prices and the ensuing global recession
revealed the fragility of this policy. Imported inflation and a sharp
decline in foreign terms of trade in 1975 made it virually impossible
to maintain an overvalued domestic currency without creating a sharp
external deficit When the military overthrew the Peronist government
in 1976, it inherited an acute balance of payments crisis and a huge
fiscal deficit of more than 13 percent of GDP.
The military government's objective was to reduce inflation drasti-
cally and to initiate a longer-term strategy that would encourage sus-
tained growth and full employment A central part of that strategy was
a two-stage trade reform in 1976-81. During the first stage (1976-
78), the govermment introduced an unannounced tariff cut, eliminated
export taxes, and replaced quotas with tariffs for all commodities ex-
cept steel and aluminum. During the second stage (1979-81), the
government implemented a preannounced schedule of quarterly tariff
reductions. Even though the program was supposed to have continued
until January 1984, trade reforms were reversed in 1981 due to a
sharp balance of payments crisis.
Completion of the trade liberalization was jeopardized not only by
the excessive gradualism used- in the second stage, but also because a
uniform tariff was not the final objective. Stikingly, the final result on
overall nominal protection-including tariff and nontariff barriers-
was contradictory: in 1979-80 the unweighted average nominal tariff
rate decreased slightly from 51.9 to 49.2 percent, but its dispersion in-
creased greatly (Cavallo and Cottani 1986).
Another problem with the trade liberalization program was the use
of an accompanying inconsistent mix of fiscalfmonetary and ex-
change rate policies. In 1979 and 1980, the government's use of an
active crawl reduction scheme based on a preannounced schedule of
future devaluations resulted in a severe overvaluation. Although tax
collection increased, the fiscal policy produced a large budget deficit
because of the increase in both current spending and public invest-
ment. Thus, the overall deficit rose from 13 percent of ODP in 1976
08 LuisA. Riveros and CarlosE. Sdchez
to 16 percent in 1981 and 17 percent in 1932. Since a basic aim of
the macro policy was to curb inflation, the government reduced the
monetary financing of the deficit and began to rely heavily on do-
mestic and external borrowing, thereby pushing up market interest
rates and significantly increasing the public external debt. The persis-
tence of a large fiscal imbalance in combination with a severe overval-
uation produced disequilibrium in the balance of payments accom-
panied by high real interest rates and low employment, production,
and investment.
The economic policy of the late 1970s did not improve the prevail-
ing domestic imbalances. Although stabilization reduced annual infla-
tion of the consumer price index from a peak of 441 percent in 1976
to 101 percent in 1980, inflation remained high: in 1981 it was still
104 percent per annum. Likewise, the active crawl reduction scheme
implemented after 1978 and the external financing of the governmen-
t's excess demand caused the real exchange rate to appreciate: taldng
1976 as a base year, its level was 39 in 1980 and 49 in 1981.
The financial policies used after 1978 also resulted in volatile real
interest rates. In January 1979, the government implemented its previ-
ously announced nominal devaluations and eliminated most restric-
tions on capital mobility. During the first eight months of 1979, when
the policy still had some credibility, real interest rates were low, but
reached negative values at times. Later on, uncertainty increased and
risk premiums became high, wnich raised real interest rates from 2 to
6 percent per month. The increased uncertainty observed in 1980-81
was closely related to both the existing gap between inflation and the
rate of devaluation (in 1980 inflation was 17 percent while the rate of
devaluation was only 6 percent) and to observed changes in external
accounts (table 23).
The long-term structural adjustment program was barely imple-
mented, and its final result was a shift in incentives in favor of non-
tradable activities. The short-term stabilization program failed: in the
presence of a persistent budget deficit, financing via capital markets
(which replaced simple money creation) produced a crowding out ef-
fect and was strongly deflationary (Mann and S6nchez 1984, 1985).
Table 2.2 also shows the significant variability in real output between
Argentina 69
Table 2.3 Selected External Accounts, 1980-84
(US$ millions)
Account 1980 1981 1982 1983 1984
Exports 8,021 9,143 7,623 7,838 8,100
Imports 10,540 9,430 5,336 4,505 4,600
Trade balance -2,519 -287 2,286 3,331 3,500
Interest payments 956 2,925 4,400 4,983 5,273
Current account
balance -4,769 -4,714 -2,357 -2,461 -2,492
Source: Cavallo (1986, table 1).
1975 and 1980, before the economy moved into another recession in
1981.
The Crisis of the 1980s
In 1981, new economic authorities had to address the external and
internal imbalances that had resulted from overvaluation and the fis-
cal/monetary mismagement. The authorities instituted a drastic pro-
gram of exchange rate devaluations to deal with the most urgent pol-
icy problem. From 1981 to 1983, the real exchange rate depreciated
by 115 percent In 1981 and 1982, years of macroeconomic adjust-
ment, the real exchange rate depreciated sharply and both the real per
capita GDP and real wages experienced large reductions (table 2.2)_
The current account deficit in 1982 was substantially smaller than that
ob:trved in 1980 and 1981 (table 2.3), thus many policymakers
probably believed that further adjustment was not necessary, and
opted for a new shift in policies during 1983 and 1984.
During 1983, the last year of the military government, policies were
aimed at recovering real wages, thereby reinstating the deliberate pol-
70 LuisA. Riveros and Carlos E. Sdndzez
icy of overvalued exchange rates. As a result, real wages rose 22 per-
cent in 1983, while the wage/exchange rate ratio increased over 40
percent. The overvaluation "'as accompanied by active fiscal policies
and a resurgence of inflation. Long-term adjustment was abandoned
and traditional populist policies returned to guide policymaking.
A civilian administration (the Radical Party) took office in
December 1983. This government inherited a very weak economic
situation and vast public expectations of improved social welfare re-
sulting from the restoration of democratic institutions. Activity levels
and wages continued to grow in 1984 accompanied by a high fiscal
deficit, growing inflation, low public utility rates, and exchange rate
overvaluation (table 2.2 shows that the real exchange rate declined 13
percent between 1983 and 1984, while real wages increased 10 per-
Cent).
At the end of 1984, the government signed an agreement with the
Inernational Monetary Fund (IMF) that initiated an external sector
adjustment based on demand reduction. During the last quarter of
1984 and the first half of 1985, the economy suffered a drop in real
wages and activity levels, a depreciation of the exchange rate, and ris-
ing inflation. In June 1985, after sharp increases in public utility rates
and a drastic devaluation, the government introduced the Austral Plan,
which contained both heterodox and orthodox measures io curb in-
flation. The former included a wage and price freeze and a deindexa-
lion of debt. The latter inc-luded long-term measures, such as a high
exchange rate and fiscal restraint.
Inflation declined rapidly in 1985 due to price and wage controls.
HoNvever, since the fiscal problem remained unsolved, monetary policy
continued to play an active role. At first, the demand for money in-
creased substantially, but then inflation returned because of percep-
tions that the program was -unsustainable. In August 1986 and
February 1987, the government made two other attempts to reduce
infiation by means of a tight monetary policy and control on wages
and prices. However, inflation remained high because its primary
source-lack of fiscal discipline-was not eliminated. In addition, in
August 1986 a period of overvaluation began: The real exchange rate
averaged 113 (index, December 1976 = 100) during the first seven
Argentina 71
months of 1986 and then dropped to 106 during August 1986 to
August 1987.
A Frustrated Process of Adjustment (1987-88)
During September and October 1987 the government began to
implement some new policies that were much more in line with a pro-
gram of structural adjustment. However, no positive result has been yet
observed and no structural adjustment has taken place. The relevant
question is why a government politically committed to structural ad-
justment ended up with quite different results. The period 1987-88
can be divided into two phases. During the first one, from September
1987 to July 1988, the government implemented a devaluation fol-
lowed by a crawling peg adjustment of the exchange rate. During the
second one, after July 1988, the feal exchange rate again appreciated
significantly.
During the first phase, the wholesale prices of nonagricultural
(essentially tradable) goods experienced significant increases with re-
spect to private services and construction (49 and 15 percent, respec-
tively). Cereal and oilseed prices increased 84 percent with respect to
private services with the help of the increase in intemational grain
prices. These figures give some idea of the improvement in the relative.
price of tradables versus nontradables. During this phase, the ex-
change rate policy provided substantial incentives to the tradable sec-
tor (table 2.4). As a result, export activities expanded and a realloca-
tion of resources towards export-oriented activities began. The effects
of this policy on the volume of exports and the trade surplus were
significant: in 1988, exports increased 43.6 percent and the trade sur-
plus increased 607.0 percent. In the specific case of manufacturing-a
potentially exportable sector in Argentina-the change in relative
prices (wages, exchange rate, and domestic terms of trade) led to a
pattern of increasing profits and remarkable export growth.
Inflation accelerated, mainly due to the failure to reduce the fiscaI
deficit. After falling from 25 percent per month. in October 1987 to 3
percent per month in December, the inflation rate climbed to over 20
percent per month in July 1988. At the time it imposed price and ex-
change controls, the government announced long-term measures
aimed at shifting resources to tradable activities, improving the x-efEi-
72 LuttA. River,s and CarlosE. Sdnchez
Table 2.4 Changes in Relative Prices, June 1985-October 1988
(percent)
June 1985- Aug 1986- SepL 1987- July 1988-
Ratios Aug. 2987 Aug. 1987 July 1988 Oct 1988
Versus private services:
Exchange rate -40.1 -5.2 25.0 -12.7
Nonagricultural WPI -44.7 -8.9 49.0 -0.5
Cereals-oilseeds -37.6 i 1.0 84.0 -21.8
Versus construction cost:
Nonagricultural WPI -7.2 1.0 15.1 -4.2
Note: WPI = wholesale price index.
Source: Institute of Economic Studies on Argentina (IEERAL).
ciency of the public sector, and freeing up rigidities in factor and oAt-
put markets.
During the second phase-after July 1988-real wages in
manufacturing increased and profits began to decline (table 25 and
figure 2.4). Tabh!? 2.4 also indicates that nonagricultural wholesale
prices (that is, mainly manufacturing prices) deteriorated compared to
private services (-0.5 percent), and constuction (-4.2 percent). In
other words, the domestic terms of trade turned in favor of
nontradables. Since the increase in wages and in the price of
nontradables was accompanied by overvaluation and elimination of
import restrictions, the manufacturing sector faced faling profit
margins and lower domestic market shares. The poor timing and lack
of coordination between short-team and long-term economic policies
hindered the achievement of structural adjustment.
Argentina 73
Table 2.5 Manufacturing Wage, Exchange Rate, and Product Wage,
1987 and 1988
(index; 1988 3rd quarter = 100) _
Wa gel Product- Wagel Product-
exchange wagel exchange wagel
Year Ouarter ratea " insb Year Ouarner rarea rainsb
1987 1 S * 1989 1 116 115
2 0 2 43 78
3 121 125 3 47 56
4 111 116 4 74 86
1988 1 114 125 1990 1 70 -81
2 109 109 2 111 101
3 100 100 3 177 111
4 102 116 4 207 133
* not calculated. for this study
a. Manufacturing nominal hourly wage divided into the exchange rate at which
imports are traded.
b. Ratio of wage cost (ratio between the nominal hourly wage paid in manufacturing
and the corresponding wholesale product price) to productivity (output per manhour).
Source: IEERAL data base.
Figure 2.4 Real Wage and Product Wage in Manufacturing
(index, 1988 3rd quarter = 100)
250
* 200
150 -
100 Ns X -
50 -
0
1 2' 3 4 1 2 3 4 1 2 3 4 1 2 3 4
1987 l9S8 1989 1990
Ycurs
Wage/changc mae --- -Prod wagewPtucdan gns
Source: IEERAL
74 Lui A. Riveros and Carios E. Sdnclhez
The Structure and Trends of Labor Markets
The performance of labor markets has reflected overall economic
tiends, which have produced low wage growth over the long run, as
well as reduced employment growth in the private sector.
Observed Trends in Real Wages
Observed labor market results are paramount in analyzing
Argentina's import substitution policy. The driving force behind
labor markets was a wage setting mechanism based on a deliberate
policy of overvalued exchange rates, restricted by the trade
balance/domestic growth tradeoff. The basic policy tool was
government -intervention supported by urban-based unions and
political groups. The main observed outcomes were a slight growth in
real wages over the long run, significant short-run economic
fluctuations, and distorted relative wages among productive sectors-
Between 1940 and 1985, real wages neither rose nor fell for more
then three consecutive years. In all but one case (1969-71), periods of
growth in real wages were followed by periods of sustained decline.
By 1985, real wages were only 61 percent higher than in 1940,
implying an average yearly growth rate slightly higher than 1 percent
(Riveros 1989; Sanchez 1987). If the shorter period 1962-87 is
considered, the evidence more than confirms the wage deterioration
over time (see table 2.2 and figure 2.2). After growth during 1962-
74, real wages declined much more than real per capita GDP. In
addition, the magnitude of short-term. fluctuations increased. For
instance, in 1962-74 real wages rose at a yearly rate of 3.1 percent
only to drop subsequently at an even higher rate (-4.4 percent).
During the whole period 1962-87, real wages fell by an average of
0.8 percent per year.
The observed trend in relative wages between tradable and
nontradable sectors is an outcome of the inward-oriented growth
strategy. Domestic market-oriented growth required relative prices
favorable to urban activities and high purchasing power for wage
earners. Most of Argentina's population is concentrated in a few
urban centers, while services, construction, and import substituting
industries produce and sell most of their output in these markets.
Argentina 75
Thus, the evolution of relative wages from the 1940s to the 1980s has
clearly favored labor in nontradable activities (Riveros 1989; Sanchez
1987). Moreover, observed wage changes did not reflect changes in
labor productivity in nontradable activities (Sdnchez 1987), and did so
only mildly in manufacturing.
During 1962-87, however, relative wages in tradables and
nontradables remained relatively stable (see table 2.6 and figure 2.5).
However, this stability probably reflects government intervention more
than relatively stable relative labor productivity. To avoid problems of
interpretation associated with the peculiar behavior of public sector
wages after 1985, namely, sharp wage cuts due to stabilization policies,
they are not included in the group of nontradables.
Table 2.6 The Evolution of Relative Wages, 1962-86
Wages in Wages in
Wv'ages in potenialy tradable Wages in potentaly tradabk
tradable industes tradable industieui
indnriesAvagesin wages in undunriesivagesin wages in
nonradable nontradable -intbi nontradabek
Year industries industries Year industries indurui&s
1962 0.83 0.99 1976 0.93 1.04
1963 0.79 0.95 1977 0.94 1.04
1964 0.82 0.96 1978 0.91 1.01
1965 0.(8 1.01 1979 0.93 1.04
1966 0.90 1.03 1980 0.89 0.99
1967 0.89 1.05 1981 0.91 1.03
1968 0.88 1.03 1982 0.91 1.03
1969 0.88 1.04 1983 095 1.05
1970 0.89 1.04 1984 0.96 1.09
1971 0.91 1.04 1985 0.96 1.08
1972 0.91 1.03 1986 0.95 1.10
1973 0.88 1.00 1987 1.02 1.16
1974 0.93 1.05 1988 1.05 1.21
1975 0.92 1.02 1989 0.94 1O10
D __________________________ I990 0.95 1.06
Source. ISERAL data base.
76 Luis A. Riveros and Carlos E. Sdnchez
Figure 2.5 The Evolution of Relative Wages, 1962-89
1.3.
1.1 . ,-<o \N
Os
0.8
0.7
0.6
1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990
-- - --Ratioofwagesin potelly dabte Ratioofwages in tmadable industries
indumtris to wages in nantudablc lo wages in nontradable industies
industries
Source: LEAL
Regional and Sectorat Labor Allocation
The concentration of population in a few urban centers is a result
of trends in public investment and social expenditures, and of
distorted relative output prices and prevailing wages in the regions
that, in turn, resulted from the centralized nature of the wage setting
process. Although population growth has traditionally been low in
Argentina, the urban population has grown much more rapidly, while
rural population growth rates have been negative for many years
(table 2.7).
According to 1980 figures, the urban population represented 83
percent of the total population. In 1988, of the country's 33 million
inhabitants, 42 percent lived in Greater Buenos Aires, C6rdoba, and
Rosario. This pattem of demographic concentration is probably a
consequence of the deliberate policy of overvalued exchange rates
(and of the accompanying commercial and macroeconomic policies)
that affected relative prices and wages and biased labor allocation
Argentina 77
Table 2.7 Average Annual Rate of Population Growth
(percent)
Census Greater
period Total Urban Rural Buenos Aires Cdrdoba Rosario
1914-47 2.1 2.5 1.40 n-a. n.a. n.a.
1947-60 1.8 2.8 -0.50i 6.1 3.6 2.4
1960-70 1.6 2.5 -1.40 3.8 3.0 2.0
1970-80 1.8 2.6 -0.03 2.5 2.2 1.6
n.a. = not available
Sourcec Sinchez (1986, tables 1 and 2).
against tradables. Any increase in the average wage of urban residents
achieved through government mandated increases and/or union
demands, increased wages in the nontradable sectors by more than
labor productivity. Since nontradables are prmairily urban goods
while agriculture is an important component of tradables, a policy of
favoring nontradables inevitably favored the urban sector. Thus, prices
in this sector had to grow at a quicker rate relative to that of tradables
to maintain relative wages at their targeted levels. The fmal outcome
was that the change in relative prices necessary to obtain a given
increase in the average wage of the urban population increased over
time.
In addition to the bias in favor of nontradable sectors derived from
aggregate economic policies, government investment in education,
health, housing, transport, communications, and culture was also
concentrated in urban areas. This allocative pattern was.reinforced by
social policies that were characterized by subsidies that instead of
supplementing the income of the poor regardless of where they lived,
provided free public services to people tiving in the largest cities,
regardless of their income (Sanchez 1986, 1987).
Legal regulations on nominal wage setting also contributed to the
concentration of resources in urban areas. Nominal wages, whether
determined by legal procedures for bargaining or established by the
government during periods when union activity was prohibited, were
78 LuisA. Riveros and CarIos E. Sdnchez
compulsory for all workers and firms. Therefore, differences in the
-relative availability of labor in different regions did not affect, the level
of wages. At the same time, this wage policy discriminated agaiLst.
small firms and marginal investors, which also restrained the creation
of modem activities in rural areas and produced a one-way flow of
migrants from the provinces to urban centers.
This latter result is consistent with economic theory. A
disaggregation of labor into two groups, skilled and unsldlled, reveals
that the provinces have excess demand for the former and excess
supply of the latter, while the opposite occurs in the wealthiest urban
areas. Wage differentials among provinces should move accordingly
and instigate a flow of skilled workers toward the regional markets,
and a flow of unskilled workers in the opposite direction. For this to.
occur, wage differentials would have to be large enough to
compensate for the cost of moving in either direction. Thus, labor
flows did not occur as expected for two reasons. First, even though
wage differentials -existed, they were not large enough to compensate
fully for the rate of institutional wage intervention in equalizing wages
among provinces. Therefore, wage differentials were lower than those
that would have prevailed in a decentralized and more competitive
wage setting process. Second, welfare policies and related social
expenditures increased the cost of moving for the skilled and reduced
it for the unskilled. For both groups, any move away from urban areas
entailed ilie loss of cheap or free social services.
The effects of the government interventions discussed above on
sectoral labor allocation were as expected. The share of construction,
commerce, and services in total employment was almost 50 percent in
1940 and more than 64 percent in 1980 (table 2.8). Likewise, the
share of construction tripled during the same period. The nontradable
sector accounted for 69 percent of all new employment created
between 1947 and 1980 (85 percent in 1960-70 and 87 percent in
1970-80). Finally, the share of the nontradable sector in total
employment was 64.1 percent in 1980 compared with 49.6 percent in
1940. By contrast, the share of agriculture in :otal employment
declined by more than half, from 27.1 in 1940 tU' [2.9 percent in
1980, and the share of manufacturing has declined steadily since
1947.
Argenitna 79
Labor Market Segmentation and Increasing Informality
As well as the notable expansion of employment in services and
construction relative to that in the tradable sectors, economic growth
caused a higher increase in nonwage than in wage employment (table
2.8). Following a standard approach to analyzing labor markets in
developing countries, urban employment can be classified into the
broad categories of formal and informal. This classification depends
on the existence or lack of protection of certain sectors with regard to
coverage by and enforcement of labor regulations. Thus, formal
employment corresponds to the public sector and to relatively large
private sector firms, while informal employment corresponds to
private sector firms that employ fewer than five persons. Since not all
the workers in the second category earn a low income, informal
employment may be further classified on the basis of income ievels:
those earning high incomes due to advanced skiUs, high capital
intensity, or an oligopolistic market environment are classified in the
quasi-formal sector, while those earning low incomes are classified in
the informal sector (Sanchez and others 1981).
Table 2.8 Employment by Sector, Selected Years
(percent)
Avcrage annual employment growth
Secwrol distribuion Agriculture Manufacturing Nontradables
Wage NVonwage- Wage Nonwage Wage Nonwage
Agr- Mana- employ- emyyv- employ- employ- employ- employ-
Year culture facuring Ret ment mnent ment mer ment mert
1940 27.1 23.3 49.6 n.a. n.a. n.a. n.a. n.a. n.a.
1947 26.1 27.2 46.7 0.1 n.a. 7.2 n.a. 3.8 n.a.
1960 20.8 27.1 52.1 -1.6 1-3 1.2 2.2 1.9 3.6
1970 16.2 23.3 605 -0.6 -2.1 0.4 -3.6 2.6 2.8
1980 129 23.0 64.1 -0.9 -0.6 1.1 2.6 1.7 3.2
n.a. = not available
Notes: Data on sectoral distnbution are expressed as a proportion of. total
employment.
Source. Sfnchez (1984, mable 8).
80 LuisA. Riveros and CarlosE. Sdnchez
The proportion of quasi-formal and informal workers in
Argentina-which is equivalent to the concept of the informal sector
used elsewhere-is about 30 percent of the urban work force.
Therefore, compared to other Latin American countries, where the
informal sector accounts for 40 percent or more of the labor force,
Argentina's informal sector is less important in urban labor markets.
Nonetheless, given that formal wages are protected by labor
regulations and government and union intervention, adjustment
policies have increased the formal-informal wage gap (Lopez aud
Riveros 1989). This increase probably has a negative influence on the
political sustainability of adjustment programs.
The lack of macroeconomic adjustnent has probably been at the
root of the relatively high income levels observed in the informal
sector, which produces mostly nontradables. For example, statistics for
1984 and 1985 show that self-employed workers with less than eight
years of schooling (a reasonable proxy for the unskilled) earned
approximately 18 percent more than wage earners with the same level
of education. Indeed, the average self-employed income was only 7 or
8 percent less than the national average wage. However, if one uses the
concept of total labor costs to account for all incomes received by
wage labor in the formal sector, a different conclusion may arise: in
1985, for example, nonwage labor costs (fringe benefits, social
security, and regular bonuses) were about 51 percent of total wage
costs (Riveros 1989).
Labor Market Institutions
The role of labor market institutions is important in leading to the
formal/informal distinction, which is, in turn, a key factor with rcgard
to the labor market response to macro policies. The Argentinian
economy is characterized by politically strong oligopolistic firms
producing for domestic markets that have created powerful
protectionist lobbies. Their counterpart is an equally powerful labor
union, organized as a corporation, whose influence on government
policies has been paramount. Although some changes in the labor
movement's power structure have occurred during the last decade,
unions are still extensive, disciplined, and politicized.
* Argentina 81
The law allows workers to organize unions freely, but to obtain the
legal right to bargain with employers, unions must have "trade union
representative" credentials. The government usually grants these to
the most important union (mainly in terms of membership) in each
area of activity, aIthough exceptionally they have been granted to
more than one union.
Even though the law protects workers' rights to join or not Join a
union, the terms of the labor contract agreed to by the representative
union and the employer are compulsory for all workers in the activity.
In turn, unions may join federations and confederations that, once
they have obtained legal representative credentials, can negotiate
wages at a very aggregate level. In this manner, wages and working
conditions are centrally determined by negotiation between national
unions and entrepreneurs; a process also characterized by strong
government intervention. During periods when collective bargaining
was practiced, wages were centrally determined by government decrees
and resolutions.
Job security regulations hinder labor mobility- The law maintains
workers' rights to keep their jobs regardless of the circumstances,
although it does not guarantee income maintenance. The law also
states that in case of dismissal, employees must receive severance
payment equal to the highest monthly wage or salary earned in the
current job (up to a ceiling of three times -the minimum wage),
multiplied by the number of years worked.
Mandated minim wages also exert important effects on labor
market outcomes. First, they set a limit for severance payments, thus
affecting the normal rate of job turnover. Second, they affect
prevailing equilibnrum wages due to the effect of minimum wages in
shifting the entire wage structure upward (Paldam and Riveros 1988;
Safnchez and Giordano 1988). Third, given the positive effect of
minimum wage changes on average wage changes, the former are also
related to existing inflationary pressures (Paidam and Riveros 1988).
Fourth, evidence suggests that relatively high minimum wages affect
the formal-informal ivage gap positively and the employment level in
formal activities Degatively (Lopez and Riveros 1989; Sanchez and
Giordano 1988).
82 Luis A. Riveros and Carlos E Sdnaw-z
The Role of Labor Markets in the Adjustment Process
Successive attempts to introduce macroeconomic equilibrium in
Argentina have failed, partly due to the absence of accompanying
labor market policies. The populist aim of maintaining relatively high
real wages has determined the existence of an overvalued exchange
rate.
The PerformWanc of Labor Markets in Adjustment
As discussed above, the two most recent attempts to introduce
structural adjustments in Argentina failed. Tables 2.2, 2.6, and 2.9
illustrate the interactions among policies and relative prices affecting
labor markets. Despite a series of nominal devaluations, the
government followed populist policies of overvalued exchange rates
between 1962 and 1983, which resulted in wage rigidity and
increasing nontradable prices relative to the price of exportables.-
As discussed earlier, Argentina has seen a sequence of devaluation-
recession periods followed by periods of exchange rate appreciation.
During the latter, wage indexation normally hindered the intended
adjustment process initiated through an usually severe devaluation.
Devaluation, via its effect on food prices in particular, led to union
attempts to restore real wages. These were usuaily attained first in the
manufacturing sector (where unions acted as the leaders of the entire
urban labor movement), and then spread to services and construction.
(Llach 1987 estimated sectoral productivity figures and concluded
that urban sector development depends much more on agricultural
productivity than its own productivity.)
The strong relationship between the exchange rate policy and
relative wages discussed earlier is illustrated by table 2.10, which is
derived from tables 2-2 and 2.6. The figures show the important role
of an appropriate exchange rate in the success of a stabilization effort
Despite the structural adjustment program announced in 1976, the
relatively low real exchange rate that prevailed from 1973 to 1980
held the ratio of wages in tradables and nontradables practically
constant. This suggests that the effects of the policy of overvalued
rates were more important than tbose of the trade liberalization and
other adjustment measures announced in the rnid-1970s- By
Argentina 83
Table 2.9 Relative Prices and GDP, 196247
Year PlPt PM/Px pipm W/ie whAIW, GDP
1962 91 98 78 57 0.99 69
1963 89 92 78 58 0.95 67
1964 84 82 73 79 0.96 74
1965 87 92 72 90 1.01 81
1966 97 106 87 98 1.03 82
1967 98 106 90 78 1.05 84
1968 102 109 97 80 1.03 87
1969 103 110 100 90 1.03 95
1970 100 100 100 100 1.04 100
1971 96 69 99 114 1.04 105
.1972 87 79 186 97 1.03 108
1973 95 88 99 149 t.Oo 115
1974 102 104 102 211 1.05 122
1975 108 138 95 219 1.02 121
1976 99 123 83 103 1.04 121
1977 101 Ill 94 114 1.04 128
1978 110 118 112 166 1.01 122
1979 111 114 116 289 1.04 131
1980 125 134 135 487 0.99 133
1981 127 150 131 391 1.03 125
1982 103 129 89 196 1-03 I18
1983 101 127 85 199 1.05 121
1984 106 134 92 254 1.09 124
1985 113 164 95 175 1.08 118
1986 110 138 110 199 1.10 125
1987 112 137 115 174 1.16 127
1988 111 143 100 168 1.21 124
1989 107 147 89 120 1.10 118
Pu = urban price, or price of goods and services produced by the import
substitution sector (manufacturing) and the service (including commerce)
and construction sectors
Pt = price of tradables, ie, price of goods produced by the export and import
sectors
Px = price of exportables, i.e, goods produced by the rural sector and by the
food and beverage sectors
n= price of nontradables, iLe., the price of goods and services produced by the
- service and construction sectors
Pm = price of importables, i.e., the price of goods produced by the import
substitution sector and the price of imported goods
W = wages and salaries
c = nominal exchange rate
Wh = wages and salaries paid in the import substitution or home goods sector
Wn = wages and salaries paid in the nontradable sector
Source: JEERAL-
84 LuisA. Riveros and CarlosE. S4nchez
Table 2.10 Relationship Between Exchange Rate Policy and Wages
(index, 1962-72 = 100)
Real exchange
Reform period WWWn Wh/Wn rate
1973-76 105 101 R3
1976-80 106 101 80
1981-83 106 103 92
1984.47 113 111 112
Wt = wages in agriculure and manufacturing
Wn = wages in nontradables
Wh = wages in the import substitution or home goods sector
Source: Authors' calculations.
comparison, the 1987 adjustment program resulted in an immediate
increase in relative prices of urban tradables. Policymakers had
expected that in response to this change in relative pnrces, wages in
nontradables would- have increased relative to wages in manufacturing,
thereby creating incentives for labor reallocation. As discussed earlier,
this did not occur due to a series of policy reversals.
The outcomes of these two adjustment attempts suggest that labor
market variables played a negative role in the effort to achieve struc-
tural changes.' While the main policy target was a change in re�ative
prices of tradables to nontradables to permit the economy to shift to-
ward production for external markets, a wage indexation mechanism
supported by government and union intervention created substantial
rigidities. This prevented sufficiently large falls in real wages in re-
sponse to the devaluation, and the inevitable result was a loss in inter-
national competitiveness. In addition, each adjustment attempt wors-
ened inflation, the resource allocation across industries, and poverty.
Level and Composition of Aggregate Employment
Traditionally, Argentina did not have employment problems result-
ing from an excess labor supply. The situation tended to be one of la-
bor scarcity, especially in the least skilled segment of the labor force
Argentina 85
(Llach 1978). Nevertheless, urban labor markets, especially in the
largest cities, experienced successive periods of scarcity and of relative
abundance of labor due to fluctuations in domestic and foreign immi-
gration to large urban centers and in labor force participation.
(Researchers have studied this adjustment problem in urban labor
mark-ets extensively since 1979, when Sanchez and others published
their first paper on the subject. See, for example, Beccaria 1980;
Beccanra and Orsatti 1985; Dieguez.and Gerchunoff 1984; Llach
1980; Mann and Sanchez 1984; Riveros 1989; Sinchez 1982, 1987.)
During the second half of the 1970s, the large urban centers expe-
rienced a situation of relative labor scarcity. Migratory flows changed
after 1970 when the largest urban markets were no longer the recipi-
ents of large number of workers in search of better job opportunities.
Rural-urban migration continued in the second half of the 1970s, but
from rural areas of each province to its capital city (Sdnchez 1984,
1986). In addition, a series of institutional and economic develop-
ments, particularly the 1976-81 adjustment program, reinforced the
falling trend observed in labor force participat;jn: the real wage was
probably below the reservation wage of many labor market partici-
pants, and a considerable number of women, young adults, and even
males aged 20-59 abandoned the labor force.
The period from the early 1970s to the early 1980s witnessed not
only declining labor force participation rates and employment (table
2.11), but the sectoral allocation of labor tended to diminish the sup-
ply of wage labor to the goods producing sectors. Workers were trans-
ferning to construction and service activities and to nonwage occupa-
tions, thereby effecting the amount of labor available to the industrial
sector (table 2.8). Although relative sectoral wages did not show a
well-defmed trend (table 2.6), employment creation in manufacturing
and agriculture was poor.
Due to these economic trends, open unemployment was not a
significant problem in the urban labor markets until the early 1980s,
and an open employment problem has now arisen.2 Although labor
force participation rates have moved ii. the same direction as
2. Open unemployment is defined as the ratio of those identified as unemployed in
the Permanent Household Survey to the economically active population.
86 LuisA. Riveros and CarlosE Sdnchez
Table 2.11 Indicators for Main Urban Labor Markets, 1980-85
(percent)
Year LFP L U Year LFP L U
1950 46 44 4.3 1970 44 42 5.0
1951 46 44 3.5 1971 43 41 5.9
1952 45 43 4.0 1972 42 40 6.7
1953 45 43 4.B 1973 41 39 5.5
1954 46 43 6.2 1974 40 39 3.9
1955 46 43 5.8 1975 40 .39 3.2
-1956 46 43 6.8 1976 39 38 4.7
1957 46 43 6.3 1977 39 38 3.2
1958 46 43 7.8 1978 39 38 3.0
1959 44 42 5.5 1979 39 38 2.2
1960 45 42 5.6 1980 39 38 2.5
1961 45 42 7.2 1981 39 -37 4.7
1962 45 42 7.3 1982 39 37 4.9
1963 45 41 8.9 1983 38 36 4.4
1964 45 42 6.3 1984 38 37 4.2
1965 45 43 5.3 1985 39 37 5.6
1966 45 42 5.8 1986 39 37 5.2
1967 46 41 6.4 1987 40 38 5.7
1968 44 42 5.3 1988 40 37 6.0
1969 44 42 4.4 1989 41 38 7.5
1990 40. 37 7.2
LEP = labor force participation as a percentage of total population
L = employment as a percentage of the economically active population
U = unemployment as a percentage of the labor force
Source: Sinchez (1987). 1950-62 computed from population and census data; em-
ployment data from Llach and Sanchez (1984); 1963-85 data from the Permanent
Household Survey.
economic activity in the short-term, long-run trends have also had an
effect As table 2.12 and figures 2.6 and 2.7 show,, participation rates
decreased from 1974 through 1983, while open unemployment and
underemployment remained low compared to their historic levels
(table 2.11).3 Labor force participation has increased during the last
five years despite high variability in output growth and the recent
slowdown in activity levels (table 2.12 and figures 2.6 and 2.7). The
reason is that the labor force response to the existing labor market
3. The underemployed are those who work less than 35 hours per week and are
actively seeking a job.
Argentina 87
situation is now different than that which prevailed until the late
1970s. The "discouraged worker" effect was predominant in those
years as declines in employment and real wages were accompanied by
withdrawal of people from the labor force (including migrants from
nearby countries). Currently, however, the dominant effect derives
from the loss of family income and shows up in the increase in the net
inflow of secondary workers into the labor market. Of course, if the
probabilities of finding a job are low in the case of experienced
Table 2.12 Labor Participation and Unemployment, 1974-88
(percent)
Laborforce Open Unaer-
participation unemployment employment Eqivalent
Year raerazea nite awi
yesr . rSe rate ra rtea
1974 40.4 4.2 5.0 6.7
1975 39.9 3.7 5.3 6.4
1976 39.3 4.8 5.3 7.5
1977 38.7 3.3 4.0 5.3
1978 38.9 3.3 4.7 5.7
1979 3823 2-5 3.8 4 4
1980 38.4 2.5 5.2 5.1
1981 38.4 4.8 5.5 7.6
1982 38.3 5.3 6.6 8.6
1983 37.3 4.7 5.9 7.7
1984 37.9 4.6 5.7 7.4
1985 38.1 6.1 7.3 9.8
1986 38.9 5.6 7.8 9.5
1987 39.2 5.9 8.2 9.9
1988 39.3 6.3 7.9 10.2
1989 40.6 7.6 8.7 12.0
1990 38.9 7.4 9.0 11.9
Notes: The data are from Greater Buenos Aires, 20 provincial capitals, and 5 other
cities. An underemployed person is one who is working less than 35 hours a week and
is seeking more work The equivalent unemployment rate is calculated using the con-
vention that two underemployed people are equivalent to one unemployed person, and
these "converted" unemployed are added to open unemployment.
a. Given as percentage of the economically active population.
Source: Permanent Household Survey.
88 LausA. Riveros and Carlos E. Sdnchez
Figure 2.6 Labor Force Participation Rate, 1974-90
(percentage of total population)
41
40.5-
40 -
39.5
39
38.5
38-
37
365 -
3-6 i
1974 1976 1978 1980 1982 1984 1986 1988 1990
Yers
Source. Permanent Household Survey.
Figure 2.7 Equivalent Unemployment Rate, 1974-90
(percentage of economically active population)
12
10 -
9
7-
5
4 -
1974 1976 1978 1980 1982 1984 1986 1988 190
Years
Source: Permanent Household Survey.
Argentina 89
workers (for example, males 20M59 years of age), they are even lower
in the case of secondary workers. The result has been an increase in
open unemployment and underemployment in association with the
"added" worker effect.
Unemployment and the Public Sector
The foregoing analysis suggests that construction and services of
low productivity are characterized by substantial hidden unemploy-
ment. Another sign of hidden unemployment is observed in public
sector employment trends: estimates indicate that in 1985, total public
sector employment was over two million people, representing as much
as 17.5 percent of the labor force and 25.0 percent of all wage earners
(table 2.13).
Argendna has exhibited rapid growth of public sector employment
in the provinces. Table 2.14 shows that during 1960-87, piublic em-
ployment grew by more than 40 percent. Most of this growth was as-
sociated with a huge increase in employment in local governments
(more than 200 percent), which took place during the pernods of un-
successful adjustment programs. Thus, at the time that overall adjust-
ment failed, partly as a consequence of labor market intervention
policies and the overvalued exchange rate, the resulting low employ-
ment growth demanded active job creation in the public sector.
Table 2.13 Public Sector Employment, 1985
Relative share (percent)
Public
Total sector Labor Wage
Subsector (thousands) employment force earners
National administration 605 30.1 5.3 7.5
Public enterprises 398 19.8 3.3 4.9
Local government 1,006 50.1 - 8.9 12.6
Total 2,009 100.0 17.5 25.0
Source: Sinchez and Giordano (1988).
90 LuisA. Riverns and Carlos E. Sdnchez
Table 2.14 Growth in Public Sector Employment, Selected Years
(index, 1960 = 100)
Subsector 1960 1965 1970 1975 1980 1985 1987
National administration 100.0 93.7 100.1 110.3 97.6 106.3 104.5
Public enterprises 100.0 81.1 76.6 101.4 73.9 80.5 79.1
Provincial governments 100.0 105.7 131.7 178.1 205.7 239.0 259.7
City govemments 100.0 116.0 134.4 181.6 179.2 202.4 220.0
Total 100.0 94.4 103.2 129.2 122.0 136.9 141.7
Source: Estimated from figures published in El Cronista Cornercia4 April 27, 1988,
p. 15.
In addition to the political commitment of keeping open unem-
ployment 16w, high employment growth in local governments was as-
sociated with changing migratory pattems. Domestic migratory flovws
changed after 1970 as economic incentives to move to the large cities
declined. Instead, people moved from rural areas to provincial capitals,
where their 'chances of finding employment were higher. Local
govermnents startcd active job creation mostly due to -the resulting
demands for more employment (see Lindenboim 1985; Sanchez
1986).
Another problem with public employment is its internal composi-
tion. In 1985, about 20 percent' of pubolic employment was in public
firms, in many cases with very low levels of labor productivity, as in
the case of railroads. By contrast, less than 30 percent of public em-
ployment was in services like health and educat;on. Thus, the relative
amount of workers in purely bureaucratic activities is high, and a
thorough public sector reform program is probably needed.
Women, Labor Markets, and Adjustment
The 1980s saw a sustained increase in the rate of female labor force
participation, especially by women aged 35-49. The drop in family
income due to the economic crisis and the need to compensate for the
associated loss in welfare might explain the higher participation of the
secondaiy labor force. As table 2.15 shows, the main sources of.em-
Argentina 91
Table 2.15 Sectoral Composition of Female Employment, Selected
Years
(percent)
Sector 1960 1970 1980
Tradable 32 25 21
Agriculture 5 4 3
Industry 27 21 18
Nontradable 68 75 79
Total 100 100 100
Source: Estimated from census dat
ployment for the rising female labor supply were nontradable activi-
ties, possibly in occupations of low productivity. In 1970, approxi-
mately 58 percent of the total female employment was in clerical po-
sitions, sales, and other services (mostly domestic services). in 1980
this figure was 63 percent. As expected, the female average wage was
approximately 56 percent of the male wage.
Income Distibution and Poverty.
The absence of structurl adjustment in Argentina has been associ-
ated with an increasing deterioration in income distribution. This de-
terioration derives from poor economic growth, which has, in turn, re-
sulted from the disarray of policies that have produced higher relative
prices.of nontradables and protective regulations that favor urban
workers in the formal sector.
Examination of the distribution of poverty by economic sectors is
less satisfactcry than analyses based on geographic or functional char-
acteristics or the formal/informal sector distinction. Unfortunately,
however, data are not available to study the latter disaggregation. In
addition, data are limited to the cities of Cordoba and Buenos Aires.
Finance, a nontradable sector, and industry, a tradable sector, enjoy
the highest income levels per worker, the lowest unemployment rates,
and the lowest levels of poverty (table 216). Commerce, services, and
construction, all nontradable, not only have relative incomes lower
92 LuisA. Riveros and Carlo: E. Sdnchez
Table 2.16 Unemployment, Sectoral Income Distribution, and Poverty
Equivalent Percentage ofpoor in
Relative income unemployment rate (%lo) total employment
Greater Greater
Sector BuenosAires Cdrdoba Buenos Aires Cdrdoba Poor Indigenz
Finance 1.4 1.6 7.1 5.1 9.6 2.7
Industry 1.1 1.1 7.1 5.5 11.7 1.8
Commerce 1.0 1.0 11.7 11.1 13.0 4.3
Services 0.9 0.9 11.2 10.6 15.5 4.5
Construction 0.8 0.8 15.3 14.5 25.2 4.6
Tradables 1.1 1.1 7.1 5.5 11.6 1.6
Nontradables 0.9 0.9 10.2 10.0 15.6 4.3
Total 1.0 1.0 9.4 8.9 14.6 3.3
Notes: Relative income includes wages and the income of the self-employed for
C6rdoba, and only wages for Buenos Aires. For percentage of poor, data refer only to
C6rdoba The poor were identified based on (a) size and composition of household; (b)
calorie intake; (c) average per. capita income. The poverty line was defined as the in-
come level necessary to buy a basic basket of food (IPA-INDEC 1988) and to cover
other basic needs. Household members with per capita incomes below the poverty
line were classified as poor, while the indigent were those with incomes below the
amount required to buy the basic basket of food only.
Source: Own estimates based on Permanent Household Survey data: October 1986 for
C6rdoba, and April 1987 for Greater Buenos Aires.
than the other two sectors, but also a significantly higher level of un-
employment and underemployment The result is a large incidence of
poverty in the latter sectors. These characteristics-lower than average
incomes, high unemployment rates, and an above average mcidence of
poverty-are typical of the nontradable sectors.
The increase in poverty in Argentina during the last 15 years is
mostly a result of failed adjustment; Ia addition to rigid wages and
distorted labor allocation, open unemployment and underemployment
have been growing for the last five years. Young males and females in
the age group 35-49 are the most significant group of secondary
workers swelling the labor supply in the informal sector. Most of this
surplus labor probably engages in construction, commerce, and ser-
vices; activities that have very low productivity rates and higher ina-
dences of unemployment and poverty.
Argentina 93
At least in the short and medium term, any adjustment process will
inevitably increase poverty due to higher transitional unemployment
and the drop in real wages. Adjustment will also imply declining pub-
lic sector employment and wages and reduced expenditure on social
welfare programs. Until now, however, Argentina has been suffering
the social cost of not adjusting and trying to maintain an inappropriate
productive structure, combined with intervention in the labor market
and use of a deliberate policy of overvalued exchange rates.
Political Economy, Labor Markets, and Adjustment
Hyperinflation is now forcing the government to attempt a thor-
ough adjustment program that includes profound changes in the
functioning and institutional organization of the labor and goods
markets. Deregulation, privatization, reduction in public expenditures,
trade liberalization, a shift of factors of production against the non-
tradable sector, and other similar actions are not without cost.
Therefore, a set of labor market policies are required to facilitate effi-
cient change at minimum cost.
The government will face severe difficulties in shifting labor
among sectors. Reforms in the labor market should be designed to
facilitate this shift at minimum cost, which requires deregulation poli-
cies. Therefore, a- set of policies should aim at decentralizing collective
bargaining and the labor unions. Another set of policies should aim at
making labor contracts more fIexible to attain higher labor mobility
and to implement feasible programs to assist the unemployed. Finally,
macroeconomic policies should be applied in combination with a dis-
mantling of labor market intervention, which will probably diminish
labor market segmentation and increase the trickle down effect of
growth.
The program of economic reforms needs to be sustainable and
credible. In Argentina, this implies strong political support and clear
commitment on the part of the govermment. This in turn requires in-
volving the unions in the reform process, or a substantial change in
their traditional political attitudes. This is not going to be an easy task,
not only because of the enormous lags that may be involved in
changing institutions, but also because the unions are aware that
structural adjustment will mean a loss of power.
94 Luis A. Riveros and Carlos E. Sdnchez
Conclusion
Although implementing and sustaining a program of structural
adjustment will involve severe difficulties, change is inevitable in
Argentina. The current economic crisis is not only the result of inap-
propriate domestic policies in response to recent external shocks, it is a
result of poor policies applied for years that have damaged the econ-
omy. Even if no external shocks had occurred, the country would still
have to shift the structure of production in appropriate directions.
Frustrated adjustment experiences provide some lessons that can
help policymakers design a sustainable program to achieve price sta-
bility and to change the system of incentives as outlined below.
* Policymakers must ensure that macroeconomic and trade poli-
cies are consistent. In particular, they must combine trade pol-
icy with fiscal reforms, since a reduction in public spending
will exert downward pressures on the real exchange rate.
* Policymakers must also emphasize changes in labor relations
and labor market institutions. Institutional changes, such as de-
centralizing wage bargaining and eliminating both wage in-
dexation and govemment intervention, are necessary to shift
resources among sectors and regions.
* A shift in resources among regions and industries requires
eliminating labor mobility, rigidities and restrictions.
Restrictions have resulted from the regional allocation of pub-
lic investment and social welfare expenditures and regulations
concerning public services and subsidized urban utilities. Thus,
more mobility will require less pervasive intervention.
* Prospective losers in an adjustment program may have suffi-
cient power to stop its implementation. Thus, incorporating
these losers into the program is vital. Public expenditures must
be reallocated to reduce the social cost of the transition period.
This will require external financing and debt alleviation.
APPENDIX
Table 2.A1 City of Buenos Aires, Income Distribution in Total
Population
(income earners only)
Decile
October (year I II III IV V VI Vl vrII IX -X
1974 4.4 4A 4.5 6.2 7.5 9.2 10.7 13.6 15.9 23.6
1975 3.1 4.1 5.5 6.6 7.3 9.0 10.9 12.5 16.4 24.6
1976 239 4.2 4.8 5.8 6.9 &2 10.7 14.5 145 27.5
1977 2.9 4.3 4.4 6.1 6.3 10.8 10.8 108 173 26.3
1978 2.7 4.0 5.4 S.6 7.1 9.9 9.9 123 18A 24.7
1979 28 4.7 4.7 4.8 7.8 7.8 9.6 13.4 16.6 27.8
1980 3.2 4.6 4.6 5.3 7.7 7.7 9.8 14.0 16.4 26.7
19t1l 1.9 5.1 5.1 5.1 5.7 8.6 9.4 23A 17.1 286
*1982 2.5 3.7 4.4 5.4 6.2 7.6 9.2 11.1 14.8 35.1
1983 2.2 3.7 4.6 5.5 6.7 8.1 9.6 11.9 ;59 31.8
1984 2.4 3.5 4.2 5.4 6.6 7.7 9.5 11.3 14.8 34.6
1985 26 3A 4.0 5.3 6.4 7.7 9A4 11.8 15.9 33.5
1986 2.5 33 3.8 5.0 6.2 7.3 9.2 11.5 15A 35.8
1987 2.0 3.0 4.1. 5.0 . 6.1 7.4 9.0 1.1A 16.0 36.0
1988 20 33 4.0 4.8 5.9 72 9.0 115 16A 36.0
Bottom 40% Middle 40% Top 20%
1974 19.5 41.0 39.5
1975 19.3 39.7 4L0
1976 17.7 40.3 42.0
1977 1-7.7 38.7 43.6
1978 17.7 39.2 43.1
1979 17.0 38.6 44.4
1980 17.7 392 43.1
1981 172 37A1 45.7
1982 16.0 34.1 49.9
1983 16.0 363 47.7
1984 155 35.1. 49.4
1985 15.3 35.3 49.4
1986 14.6 342 51.2
1987 14.1 33.9 52.0
1988 14.1 33.6 52.4
95
96 Luis A. Riveros and CarlosE B. Sdnclzez
References
Beccaria, L. 1980. "Los Movimientos de Corto Plazo ern el Mercado
de Trabajo Urbano y la Coyuntura 1975-78 en la
Argentina." Desarrollo Econdrnico 20:78.
Beccarfa, L., and A. Orsatti. 1985. "La Evoluci6n del Empleo y los
Salarios en el Corto Plazo. El Caso Argentino 1970-1983."
Documento de Trabajo No. 14. Buenos Aires: United Nations,
Economic Comniission for Latin America.
Cavallo, D. 1986. "Long-Term Growth in the Light of External
Balance Policies: The Case of Argentina." Washington, D.C.:
World Bank.
Cavallo, D., and J. Cottani. 1986. "The Timing and Sequencing of
Trade Liberalization Policies: The Case of Argentina."
Washington, D.C.: World Bank.
Cavallo, D., and R. Domenech. 1988. "Politicas Macroecon6micas y
Tipo de Cambio Real, Argentina,. 1913-84." International
Food Policy Research Institute, IEERAL, and International
Center for Economic Growth. Unpublished manuscript.
Cavallo, D., and Y. Mundlak. 1982. Agriculture and Economic
Growth in an Open Economy: The Case of Argentina.
Research Report 36. Buenos Aires: International Food Policy
Research Institute.
Dieguez, H., and P. Gerchunoff. 1984. "La DinAnica del Mercado
Laborat Urbano en la Argentina, 1976-1981." Desarrollo
Economico 24:93.
IPA-INDEC. 1988. Canasta Bdsica de Alimentos, Gran Buenos
Aires. Documento de Trabajo no. 3. Buenos Aires.
Lindenboim, J. 1985. La Terciarizacidn del Empleo en la Argentina.
Una Perspectiva RegionaL Proyecto PNUDIOIT, ARG184/029,
Ministerio de Trabajo y Seguridad Social Secretarfa de
Planificaci6n. Buenos Aires.
Llach, J. 1978. "Estructura Ocupacional y Din&nica del Empleo en
la Argentina. Sus Peculiaridades 1947-1970." Desarrollo
Economico 17:68.
Argentina 97
. 1980. "Poblaci6n Econ6micamente Activa, Tasas de
Desempleo y Demanda Agregada: La Experiencia Argentina
Reciente en Busca de una Teorfa." Anales de la Asociacidn
Argentina de Econornfa Polftica, XV. Annual meeting. Mar
del Plata, Argentina.
. 1987. Mercados de Trabajo, Tirminos del
Intercamblo Externo y Salarios Reales en el Largo Pkazo. La
Argentina y Amdrica Latina, 1940-1985. Serie Documentos
de Trabajo. Buenos Aires: Instituto Torcuato Di Tella.
Llach, J., and C. S6nchez. 1984. "Los Determinantes del Salario en
la Argentina. Un Diagn6stico de Largo Plazo y Propuestas de
Politica." Estudios 7:29.
Lopez, R., and L. Riveros. 1989. "Macroeconomic Adjustment and
the La'bor Market in Four Latin American Countries." PRE
Working Paper 335. Washington, D.C.: World Bank.
Mann, A., and C. Sdnchez. 1984. "Monetarism, Economic Reform
and Socioeconomic Consequences: Argentina 1976-1982."
International Journal of Social Economics 11:3Z4.
1 1985. "Labor Market Responses to Southern Cone
Stabilization Policies: The Cases of Argentina, Chile,
Uruguay." Inter-American Economic Affairs 38:4.
Nogues, J. 1981. "Sustituci6n de Importaciones Versus Promoci6n
de Exportaciones: Lmpactos Diferenciales sobre el Empleo en.
el Sector Manufacturero Argeutino." Desarrollo Economico
22:86.
Paldam, M., and L. Riveros. 1988. "El Rol Causal de los Salarios
Minimos en Seis Mercados Laborales de Amdrica Latina."'
Estudios 11:46.
Riveros, L. 1989. "Labor Market Reforms for Structural Adjustment
in Argentina." Washington, D.C.: World Bank, Country
Economics Department. Processed.
SAnchez, C. 1982. "Salario Real y Oferta de Trabajo en la
Argentina." Estudios 5:21.
. 1984. "Empleo y Crecimiento en la Economia
Argentina." Estudios 7:31.
98 Latds A. Riveros and CarIow E. Sdnclhez
, 1986. "La Distribuci6n Regional de la Poblaci6n
Argentina y los Efectos Distorsionantes de los Subsidios
Sociales e Inversiones en Infraestructura Social." In
Economras Regionales y Crecimniento. C6rdoba, Argentina:
IEERAL.
. 1987. "Characteristics and Operation of Labor
Markets in Argentina." Report No. DRD 272. Washington,
D.C.: World Bank, Development Research Department.
Sanchez, C., and 0. Giordano. 1988. "El Salario Minimo en la
Argentina: Alcances y Evolucicn (1964-1988)." Documento
de Trabajo/13 Proyecto Gobierno Argentino-PNUD-OIT,
ARG/87003. Buenos Aires: PNUD-OIT
Sdnchez, C., F. Ferrero, and W. Schulthess, 1979. "Empleo,
Desempleo y Tamafio de la Fuerza Laboral en el Mercado de
Trabajo Urbano de la Argentina." Desarrollo Econdmico
16:15-24.
Sfinchez, C.. H. Palmieri, and F. Ferrero. 1981. "'The Informal and
the Quasi-Formal Sectors in C6rdoba." In S. V. Sethuraman,
-ed., The Urban Informal Sector in Developing Countries,
Employment Poverty and Environment. Geneva: International
Labor Office.
BOLIVIA-
Susan Horton
Bolivia is a particularly interesting country to study from the
viewpoint of structural adjustment. In the late 1970s and 1980s it
suffered from a number of dramatic events: the debt crisis, falling
world commodity prices, the collapse in the world market for its main
export (tin),, domestic political instability, hyperinflation, currency
reorganization, a stringent stabilization that succeeded in controlling
inflation, and a structural adjustment with as yet only very modest
effects on long-run growth. As Bolivia is the poorest country in South
America (and the second poorest in the Western hemisphere), it is a
particularly useful case study for comparison with low-income
countries in Sub-Saharan Africa and Asia.
The Origin and Nature of the Adjustment Problem
Like all the developing countries, Bolivia suffered external shocks
in the 1970s and 1980s. As a not very diversified, mineral exporting
economy, in the 1980s it suffered particularly from the fall in
commodity prices and the collapse in the market for tin, its principal
export. However, the magnitude of its economic collapse has
suggested to observers (such as Morales and Sachs 1988) that both
The author would like to thank the Instituto Nacional de Estadistica (INE) in
Bolivia for assistance in obtaining data and for performing some of the cross-
tabulations. Thanks also to a large number of people who assisted with providing
information and references, including Alejandro Mercado, Marcel Mercado, Miguel
Fernandez, and Teresa Reinaga (INE); Rodney Pereira and Arthur Mann, Unidad de
Anhlisis de Politicas Econ6micas (UDAPE); Juan Autonio Morales, Universidad
Cat6lica Boliviana (UCB); Roberto Casanovas, Centro dc Estudios para el Desarrollo
Laboral y Agrario (CEDLA); John Newman, Peter Miovic, Steen Jorgenson, and
Beatriz Balcazar (World Bank); to Luis Riveros and Dipak Mazumdar for helpful
comments on an earlier draft; and to Rodney Schmidt for capable computer assistance.
99
100 Susan Horton
internal and external factors were to blame for its poor performance.
Bolivia's macroeconomic performance has been well documented; in
part due to the notoriety of the hyperinflation. The discussion here is
therefore fairly brief. The events of the 1970s and 1980s draw heavily
on Morales and Sachs (1988), Morales (1987), and unpublished
World Bauik documents. Table 3.1 sets down some of the salient
macroeconomic indicators for the 1970s and 1980s.
Table 3.1 Selected Macroeconomic Indicators, 1970-89
(Z) (2) (3) (4) (S) (6) (7) (8)
Debuil Government
Percentage axport Current Trade deficit Effecstive
growth in Inflation ratido balance balance (percent exchange REER
Year CDP (percent) (percent) (USSm) (USSm) of GDP) rate i(ddb indie
1970 n.a. n.a. 231.0 1.8 13.8 n.a. 95.9 na.
1971 4.90 n3.a n.a -33 16.8 n.a. 95.6 n.a.
1972 5.80 as. n.a. -3.7 24.9 u.&. 104.7 na.
1973 6.68 n a. n.a. -0.1 30.3 m.a. 135.5 n.a.
1974 5.15 LA. n.a 146.1 190.3 A.&. 98.8 n.a.
1975 6.60 A.R. 166.6 -171.5 -130.6 Dn.. 100.0 n.a.
1976 6.10 n.a. n.a. -70.8 -25.5 u.a. 101.0 na.
1977 4.21 na. V.a. -86.8 15.4 u.a. 98.3 n.a.
1978 3.35 13.5 n.a. -288.7 -139.9 D.a 96.0 31.3
1979 -0.02 45'5 n.a. -345.5 -134.5 n.a. 92.1 32.8
1980 -0.56 23.9 214.5 -54.1 263.8 -9,1 85.7 35.8
1981 0.92 25.1 272.1 -492.0 -63.0 -7.6 72.7 45.1
1982 -4.92 296.5 313.4 -219.1 250.2 -15.9 83.2 48.9
1983 -651 328.5 367.0 -204.1 166.0 -17.7 82.3 44,9
1984 -0.30 2,177.2 401.8 -194.5 232.9 -24.0P 82.9 58.2
1985 -0.15 8,1705 478.7 .-429.8 -69.4 -9.9w 97.9 100.0
1986 -2.93 66.0 596.1 -405.0 -1175 -4.0 109.9 29.4
1987 2.16 10.7 743.4 -531.2 -234.1 -9.8w ua. 28.4
1988 2.78 21.5 n.a. -396.1 -158.8 n.s. n.a 26.9
1989. n.a. aa. . 1n.a La. nfa. n.a. na. 25.4
n.a. = not available
Notes: P = preliminary, e = estimate, minus sign shows a deficit.
a. Includes public and publicly guaranteed debt only.
b. 1975 = 100, less than 100 implies undervalued, more than 100 implies overvalued
relative to base year.
c. 1985 = 100, more than 100 implies overvalued, 1989 value is for September.
Sources: Column (1) Muller y Asociados (1988), fbr 1988 World Bank data; column
(2) World Bank data; column (3) World Bank (1988); columns (4), (5), (6) World Bank
and Central Bank of Bolivia data; column (7) Cottani (1988); column (8) IMP (various
years).
Bolivia 101
The Bolivian economy suffers from some unusual problems. The
country is landlocked and relatively sparsely populated (6.6. million
people spread over 1.099 million square kilometers), and there are
extreme variations in altitude between the main cities, making
transport problematic. The country has three main climatic zones
(figure 3.1): the mountains and altiplano (high plains), location of
mining and traditional agriculture and site of the capital La Paz; the
adjacent valleys; and the eastern tropical lowlands, the most promising
area for future agricultural expansion and site of the most prosperous
lowland city Santa Cruz. The poorly. developed transport
infrastructure means that the internal market is fragmented. It also
means that the best export potential is in items with high value per unit
volume (at present tin, other minerals, and coca).
Bolivia's modem economic growth dates from the 1952 revolution,
which set the course of economic policy until 1985. The economy was
then and is now highly dependent on the primary sector, particularly
agriculture and mining. Two important cornerstones of the 1952
policy changes were agrarian reform (land was transferred from the
large landowners to the peasants), and nationalization of the mines.
Another important step was the heavy emphasis placed on the state's
role in subsequent industrialization. As in many Latin American
countries, the industrialization strategy involved import substitution,
with high tariffs and an overvalued exchange rate.
The problems that ensued from this strategy in the difficult years of
the 1970s and 1980s can be illustrated by some World Bank (1989)
figures. In 1965, Bolivia was an average to good performer among the
lower-middle-income countries in terms of the structure of
production, savings, export performance, and so on. By 1987 it was a-
poor performer. The share of agriculture in GDP had actually risen to
24 percent of GDP (the primary sector overall contributed 34 percent
of GDP by sector of production). Savings performance was dismal, 2
percent of GDP compared to an average of 21 percent for lower-
middle-income countries, and so was investment at 9 percent of GDP
for Bolivia versus an average of 21 percent for lower-middle-income.
countries. Exports had sunk to 14 percent of GDP compared to 21
percent in 1965 and to an average of 22 percent for lower-middle-
income countries in 1987. Of legal merchandise exports, 93 percent
102 Susan Horton
Figure 3.1 Bolivia IBRD 23507R
70s 66 62@
BOLIVIA
B R A Z I L - -CONWOUR LNES IN MErERS
.t DEPARrMENT CAPITLS
DEPARTMENT BOUNDARIES
INTERNATIONAL BOUNDARIES
KLOMETERS
. _ 4 . 0 100 200 300
Y7r' PA NOC' ~~~~ ~~~~100 260
CjoPANDO : MILES
j -'- BRAZIL
BENI
ua/5 sTrinidad
16' ~ oPc ~1
&%tII 1 VSANTA CRUZ
� ORURM\
-20- tt POTOSI /0%
2' I --S - l
N\ cN - * PARAGUAY
CHILE~� f \
ARGENTINAA R G E N T I N A 24 A i2
( aly jud gwi a.n *o Jgl
s efu of m flrflo.
or a _r
or occuman. fof such
JUNE 1994
Bolivia 103
were from fuels, minerals, and metals. In 1970, the long-tem debt to
GNP ratio of 49.3 percent was the highest of all lower-middle-income
countries, and in 1987 it was the sixth highest at 185.6 percent, higher
than that for all other Latin American countries except Nicaragua
The political situation has been volatile, even by Latin standards,
with numerous changes of government since independence in 1825.
In the period of modem economic growth, a civilian govemment was
in power from 1951 to 1964, followed by military ruIle until 1978.
The period 1978-82 was exceptionally turbulent politically, with seven
different presidents, and the military prevented a left wing coalition
headed by Siles Zuazo from assuming power. The latter took office in
1982 and presided over the years of economic collapse- A more right
wing government headed by Paz Estenssoro took over in July 1985
and implemented a radical shift in overall economic policy (Morales
1987). In 1989 Paz Estenssoro's party achieved a plurality in the
elections, but power was assumed by a rather unusual coalition
between the left of center party and a party of the further right under
Paz Zamor
Morales (1987) highlights the fairly severe in:ernal conflicts that
contributed to the poitical instability and, in turn, to the difficulties of
economic management There have been long-standing conflicts
between the government and the miners: the government has wanted
to use implicit taxes on the nationalized mining sector to pay for
government expenditure, while the miners have demanded real wage
increase. During the military years, the middle classes became used to
consumption raises obtaned at the expense of the unions, and were
unwflling to give these up on return to civilian rule. The campesinos'
(peasants') poIitical power increased following land reform, but the
government has since neglected the traditional agricultural sector. The
campesinos' power has also been eroded by the shift toward
commercial agriculture in the eastern lowlands. Finally, regional
conflicts have arisen, with the prospering eastem regions unwilling to
subsidize declining traditional areas in the highlands.
Some of the seeds of the economic problems of the 1980s were
sown in the 1970s. The boom in oil, gas, and commodities, along with
the discovery of petroleum and gas in Bolivia, led to large public
capital inflows equal to 50 percent of export receipts between 1975
104 Susan Horton
and 1978 according to World Bank data. Public investment dominated
private investment during this period: 11 percent of GDP versus 6 to 7
percent between 1976 and 1978. The state's large role in the
economy, along with political instability, tended to inhibit private
.sector investment, and private capital outflows took place. According
to the World Bank, private capital flight and unrecorded imports
amounted to over 60 percent of the value of debt accumulated during
the decade 197181. However, economic growth was respectable at
55 percent per annum during 1971-78.
Bolivia's macroeconomic problems began in 1978/79 when
commercial banks became concerned about their exposure and slowed
down lending, which led to problems in debt servicing- Thus, the debt
crisis hit Bolivia a little sooner than some of the big Latin American
debtors. The year 1979 saw a large devaluation, an IMF stand-by
agreement, a World Bank structural adjustment loan, and the initiation
c? negotiations that subsequently led to a commercial debt
i.scheduling in 1981- However, the political instability prevented the
country from following a coherent economic policy. The Siles Zuazo
government attempted sLx stabilization packages between 1981 and
1985. However, these were ad hoc and not well coordinated, and the
Confederacion Obrero de Bolivia (COB), the union movement,
opposed them. Many strikes took place and the COB declined the
presidents overtues to join the government Some of the measures in
the stabilization packages actually fueled the accelerating inflation: the
fiscal component was weak, and "dedollarization" (insistence that
transactions be denominated in local currency rather than in U.S.
dollars) arguably reduced the amount of inertia in the system- The
economic difficulties were compounded by agricultural problems in
1983, when the west suffered from drought and the east from floods.
Morales (1987) descnrbes this period in more detail.
The economic collapse finally. resulted in hyperinfltion in
1984185, the only hyperinflation of the 20th century not caused by
war or revolution (Morales and Sachs 1988). According to Morales
and Sachs (1988), the government deficit contributed heavily to the
hyperinflation. Bolivia's tax base has traditionally been weak;
however, the deficit rose because the government was politically
unable to cut spending when one of its main sources of funds (foreign
BolWa 105
capital inflows) dried up rather than because of new spending.
Resorting to inflationary fmance initially helped the government's
position, but eventually eroded the tax base. Economic agents
(individuals, firms, and even the parastatals) delayed payments to such
an extent that the amounts the government ultimately received became
valueless. Some taxes (excise and property) were in nominaI terms,
hence their value dwindled to almost nothing, as did the value of
money received for government services. The overvalued exchange
rate, with the overvaluation itself exacerbated by rapid inflation,
encouraged smuggling, thereby reducing import tax receipts- Finaly,
the downturn in economic activity also reduced imports and tax
receipts. Although the government ceased debt service to the
commercial banks in early 1984, this was insufficient to prevent the
final onset of hyperinflation-
Stabilization and Structural Adjustment after 1985
In September 1985, the newly elected centre-right govermment of
Paz Estenssoro implemented a policy package combining elements of
both stabilization and structural adjustment This new economic policy
(NPE) occurred at an inauspicious time, only a month before the
world tin market collapsed Nevertheless, the stabilization measurcs
were immediately successful: inflation plummeted within a week or so
of the decree- The longer-run structural adjustment measures have had
less immediate success. Having been negative every year but one since
1978, economic growth was positive in 1987 and 1988, although this
is hard to attribute purely to the reforms.
Since 1985 the government has issued a number of other decrees
aimed at strengthening its structural adjustment policies. These include
the creation of the Emergency Social Fund in December 1986 and the
economic reactivation decree of July 1987.
The NPE included reforms in a number of important areas,
including the foreign exchange market, fiscal policy, tariff structure,
external financing, public enterprises, and liberalization in three
important markets, namely, those for goods, finance, and labor.
The reforms in the foreign exchange market were critical to
controlling inflation. The years of very rapid inflation had led to
domestic prices being set in U.S. dollar terms and then converted to
106 Susan Horton
domestic currency using the spot exchange rate. The exchange rate
was sharply devalued to a more realistic rate, and has since been
determined by weekly auctions. There is almost no difference from
the parallel market rate. The foreign exchange market has also been
liberalized. Morales (1987) argues that the latter development played
an important role in -destrangulation" of the external sector: it
encouraged the repatriation of at least some of the private capital
flight from the preceding years, and the government also whitew2sh5d
the dollars from the clandestine economy.
Morales and Sachs (1988) maintain that control of fiscal and
monetary policy is the other crucial component of successful
stabilization. As Bolivia does not have a bond market, the fiscal deficit
automatically leads to monetary expansion in the absence of foreign
financing. Following the inauguration of the NPE, the government
moved to control public expenditure and to raise taxes. It cut public
expenditure primarily by shedding employment. The mining.
parastatal (COMIBOL) suffered the largest cuts-the labor force
dwindled from 30,000 in 1985 to 7,000 in 1987-and there were also
cuts in the hydrocarbon parastatal (YFPB). Total public employment
(including the two parastatals previously mentioned) fell by 58,815
between 1985 and 1987 (CET 1988), a fall of about 25 percent (using
figures on total public sector employment from Klinov 1987). On the
income side, the government achieved immediate revenue from raising
the price of hydrocarbons to world levels-a tenfold price increase-
and by collecting back taxes from the mining and hydrocarbon
parastatals, which had had an incentive to delay payment during the
hyperinflation. More long-run reforms included a national value
added tax of 10 percent implemented in 1986, reforms of personal
taxes, and a land tax implemented (despite opposition) in 1988 that
for the first time requires campesinos to pay direct taxes.
Restructuring incentives for trade is a frequent component of
structual adjustment plans. The NPE abolished import prohibitions
and licensing requirements and set tariffs initially at 10 percent, plus
10 percent of the previous tariff, which was replaced in 1986 by a
uniform tariff of 20 percent on almost all items (World Bank
unpublished document).
Bolivia 107
External finance is usually important to ease the transition to
stabilization and structural adjustment. in 1985, Bolivia was in arrears
to all its creditois and in default to some. Since then the country has
become current on payments to multilateral creditors, rescheduled
payments to bilateral government creditors, and has tried to negotiate
relief on commercial debt. As a small country and the second poorest
in the region, Bolivia has tended to receive fairly high levels of aid per
capita. Although the NPE received only fairly modest extemal
support, the Emergency Social Fund was a channel- for sizeable
additional amounts of aid (Newman and others 1990).
The reform of the public sector and the dismantling of much of the
system put in place since the 1952 revolution was a far-reaching
change. In addition to the cuts in employment, the government also
stepped back from its involvement in most parastatals (except
COMIBOL and YPFB). It dissolved the Bolivia Development
Corporation's holdings and transferred them to regional development
corporations, to municipalities, and in some cases privatized them or
made them into cooperatives (Morales 1987; World Bank unpublished
document). The aim was to promote the private sector, and the
ensuing policy debate has emphasized the promotion of small-scale
enterprise (Delons and Bour 1988; Sinchez 1988).
Finally, the NPE included provisions for liberalization in various
important markets. In the goods market, the government removed
price controls, which had been particularly important in the
agricultural sector. The government also raised the prices of public
services using - neighboring country levels as guidelines and
deregulated trucking (World Bank unpublished document). In the
finance market, it freed interest rates and reduced controls on financial
contracts. However, the banks remain weak due to the aftermath of the
hyperinflation and real interzst rates are still high. This tends to
conflict with aims to encourage the private and small-scale. sector-
The government made a number of far-reaching changes in the
labor market (see Donoso 1988). In addition to public sector labor
shedding, the private sector was allowed to freely rescind work
contracts. Previously these were of unlimited duration, and any worker
hired for longer than a certain period of time became a permanent
employee. In an attempt to mitigate some of the effects of the labor
108 Susan Horton
shakeout, the government introduced a temporary relocation benefit
for fired workers.
Another important change was the consolidation of the wage
structure. Since 1971, governments had granted so many bonuses (a
bonus is usually equivalent to a month's salary) that the wage actually
paid and the base wage bore very little relation to each other. During
November 1982 to March 1983, the number of bonuses. paid in
annual income (in addition to base salary) ranged from 4 in
agriculture to 44 in manufacturing (Mercado and others 1988). The
previous government had already tried to consolidate wages, but
unions persisted in renegotiating the bonuses. After 1985, however,
only overtime and Christmas bonuses remained. The government also
stepped out of private sector wage negotiation, leaving it up to the
employer and employee (or union), whereas previously national
negotiations between the government and the unions had taken place.
Public sector wage scales were regulated so as not to exceed a certain
multiple of the minimum wage. Unlike some of the other stabilization
plans in Latin America, wages did not need to be de-indexed as formal
indexation had not been especially important in Bolivia. Although the
minimum wage had been indexed between November 1982 and
February 1985, it had mainly affected some social benefits such as
pensions, and direct bargaining between the unions and government
had been more important in the wage setting process.
Following the NPE, a number of other decrees were aimed at
further structural adjustment. The Emergency Social Fund was set up
as a temporary measure to create employment, rn )tivated pardy by the
nse in unemployment, the increase in the share of the unemployed
with previous job experience and the corresponding decline in the
share of new job market entrants, and the larger share of heads of
household in the unemployed (see PREALC 1985). The decree aimed
to set up a two-year program for immediately implementable projects
mainly in the large cities and in mining areas, and was divided into
three subprograms: infrastructure, housing, and small enterprises. It
was subseq-uently extended, but-is scheduled to wind down by the end
of 1990. The government hoped that the program would halve the
unemployment rate.
Boivia 109
In 1987, the government issued the economic reactivation decree,
which laid out policies for reactivating the economy (see Doria
Medina 1987 and Villegas Quiroga 1987 for a none too sympathetic
assessment). The decree was a three-year plan that aimed to increase
employment, increase the growth and diversification of exports, raise
domestic production behind low and uniform tariffs, enhance the
availability and reduce the cost of credit to the private sector, settle the
payment schedule on outstanding external commercial bank credit,
and increase the supply of housing (World Bank unpublished
document). The government was to use a number of tools to achieve
these euds, including a public investment program (in conjunction
with the Emergency Social.Fund); a fund for private sector credit; a
tax rebate of 5 percent on traditional exports and 10 percent on
nontraditional exports; changes in banldng regulations. (for example,
some limits on lending and restructuring of the three state banks); a
debt swap arrangement for outstanding external commercial bank
debt using aid money; and the creation of some new institutions,
including the National Council for Social Policy, and others in the
areas of housing and export promotion (World Bank unpublished
document). The government that came to power in 1989 also issued
policy statements concerning structural change similar to those of the
preceding government.
It is too early to assess the effects of the adjustment program. The
stabilization part of the package was immediately successful. Inflation
halted within a week or so (Morales and Sachs 1988), and has since
remained at 10 to 14 percent per annum, one of the lowest rates in
Latin America. The Emergency Social Fund has been relatively
successful at reaching its employment targets, if rather slow to begin.
By December 1987 it still employed fewer than 10,000 people in a
given month, although by June 1988 the figure had risen -to 20,000 in
a month (Emergency Social Fund unpublished data). This compares
to -an estimated number of unemployed of 110,000 in the nine
department capitals in 1988 (INE 1988b). The package was also
relatively successful at targeting poorer individuals (Newman and
others 1990).
However, the reactivation measures have had less immediate
success. Although the GNP began to nrse again in 1987 and 1988, the
110 Susan Haron
recovery has been relatively weak to date, and was largely
concentrated in construction and manufacturing. A number of
problems remain. The trade balance has worsened since 1985 despite
the devaluation, although estimates suggest that coca exports more
than compensate. The labor market statistics from 1988 and 1989 are
worrisome. There is also concern that the Paz Zamora government
may be finding it difficult to maintain tight fiscal control.
A number of important obstacles remain in the path of successful
adjustment. First, some analysts argue that the exchange rate is still
somewhat overvalued, causing Bolivian labor costs to be high relative
to those in surrounding countries. One possible cause of this is a
"Dutch disease" effect due to drug exports and possibly repatriation
of some of the funds from capital flight during the inflationary years
that are fueling the construction boom. The narcotics trade exerts an
umnreasurable but large effect on the economy: unofficial USAID
estimates for 1988 place coca exports at 56 to 87 percent of the value
of legitimate merchandise exports, and 6-0 to 9.3 percent of GDP in
direct effects. Overvaluation of the exchange rate is also problematic
for the trade balance.
Second, Morawetz (1987) cites the problems of low labor
productivity in Bolivia, due in part to low levels of human resources,
that further increase the problems of export competitiveness.
Third, the opening of the economy to trade led to large
contractions in sectors such as milk, paper, and chemicals (Afeha and
others 1988), serious in view of Bolivia's very limited manufacturing
base. Another problem for the trade sector is that people from
neighboring countries with foreign exchange controls and overvalued
official exchange rates are using Bolivia's free foreign exchange
market Bolivia's consumer goods market has been flooded with
cheap imports from Brazil, Chile, and Peru, whose exporters can
obtain bolivianos to exchange for dollars at favorable rates.
Comparing Bolivia's current set of policies with the disastrous
Southern Cone experiments of the 1970s to liberalize current and
capital accounts simultaneously is tempting.
Fourth, the banks continue to experience dificulties, and the tight
monetary policy exacerbates the problem of already scarce and
expensive credit to the private sector.
Bolivia I1I
Fifth, new institutions created as part of the structural adjustment
measures have been slow to commence activities. The Emergency
Social Fund took longer to get going than expected, partly because
the central bank, which administers the fund, underwent a major
reorganization and labor shakeout along with other government
agencies. The other institutions created by the NEP have been even
slower in getting off the ground.
Finally, additional external finance to support adlustment has been
limited. Bolivia is unfortunately considered too small and unimportant
to receive the same assistance with restructuring its debt that the big
Latin American debtors receive, although efforts to contrn] drug
trafficking may lead to greater leverage with some aid donuis. The
success of the debt buy-back was modest (World Bank unpublished
document), but a 1990 donor's meeting was more successful than past
exercises in obtaining aid pledges.
Some people are concemed that the rather unusual coalition that
formed the incoming government in 1989 may find it hard to
maintain the pace of reforms. Observers are carefully examining
indicators such as the govemment's fiscal position.
An in-depth evaluation of the structural adjustment polices will
have to await further developments. However, the political costs of the
measures undertaken have not been trivial: there have been numerous
marches of campesinos in the capital, demonstrators confronted by
tanks, strikes and blockades by groups such as teachers, and so on.
Some early indications of recovery are therefore important to
maintain political support for the measures.
The Bolivian Labor Market
Many studies examine Bolivia's labor market. This section
summarizes their conclusions (see appendix A for data sources). Note
that in the tables presented in this section the years covered are
dictated by data availability.
Maletta (1980) describes the evolution of the Bolivia labor force
based on the three censuses. Bolivia has undergone many of the same
employment changes as other developing countries, namely, a falling
share of employment in agriculture, -an increase in the size of firms, an
increase in the share of salaried employment, and a fall in the overall
112 Susan FTorton
participation rate as younger age groups stay out of the labor force
for education and older people retire. Increased participation of
women over time is not obvious, but data on ,this are only available
since 1976. The 1952 revolution occasioned some changes in
employment, particularly in that land reform caused the replacement
of hired labor by self-employed farmers. Subsequently a trend back
to hired agricultural labor has been evident as the focus of production
has shifted away from the traditional agriculture in the aitiplano and
toward commercial agriculture in the east.
The country's marked regional differences affect the labor market.
Bolivia is divided into nine departments. Although the departments do
not correspond exactly to different geographic zones, a crude guide is
that La Paz, Potosi, and Oruro are basically western mountain and
altiplano areas; Cochabamba, Chuquisaca, and Tarija are eastern and
southern valleys; and Beni, Pando, and Santa Cruz are eastern lowlands
(see figure 3.1).
Migration has followed the shift of economic activity. Between
1900 and 1976, the population decreased in three departments
(Cochabamba, Potosl, and Chuquisaca), increased in three (Santa Cruz,
Beni, and Pando), with little change in the other three. During 1900-
50, migration was mainly toward the four cities that were at the center
of traditional activities, namely, La Paz (the capital), Oruro and Potosi
(mining centers), and Cochabamba (an agricultural center). Since
1950, there has been a shift toward the provinces of commercial
agriculture in the eastern lowlands and a demise in the traditional
mining areas, a trend likely to be accelerated by structural adjustment.
Thus, in Bolivia, structural adjustment implies substantial geographic
mobility.
The cities of La Paz, Santa Cruz, and Cochabamba account for
most of the urban population, and together with Oruro are sometimes
known as the Eje Central (central axis). Since data tend to be
consistently available for these cities and not for some of the smaller
cities, much of the discussion is based on the Eje Central. The Eje
Central accounts for just under 90 percent of the employed
population of the nine department capitals.
Table 3.2 shows some of the aggregate labor force statistics for
Bolivia. Participation rates are similar to those in other Latin American
Table 3.2 Labor Market Indicators, Urban Centers, 1980-89
indicator 7980 198) 1982 1983 1984 1985 1986 1987 1988 1989
Seven cities (department capitals)
Population older than 10 1,321 1,363 1,416 1,470 1,626 1,686 n.a, 1,830 1,977 1,772
(thousands of people)
Eje Central (four cilies)
Population older than 10 1,155 1,196 1,246 1,279 1,454 1,501 1,535 1,625 1,730 1,606
(thousands of people)
Labor force participation rate 51.2 49.6 41.8 44.8 49.7 44.8 47.1 46.3 489 48.7
(percent)
Unemployment rate 7.5 6.2. 7.5 8.2 6.6 5.7 4.2 5.9 115 10.7
(percent)
Underemployment, rate 9.0 n.a. 4.8 3.4 4.2 2.0 2.2 3.5 - 7.0 5.3
(percent)
Pcrcentage of job leavers 62.9 68.6 n.a. 67.6 67.1 50.6 49.3 56-6 762 66.0
in unemployed
Percentage of salaried 58.4 58.9 58.9 58.2 53.8 56.9 57.5 53.8 543 n.L
employees
Hours worked per week 44.6 n.a. 42.0 n.a. 44.6 n.a. 44.1 44.8 445 na.
n.a. = not available
Notes: The participation rate Is the ratio of the economically active population to the population aged over ten years. Job leavers are
those who have previously held a job (the underemployed consist of both job leaver and new entrants to the labor force).
Underemployment is all those working 12 hours or less per week.
Sources: Hours worked based on author's calculations; other indicators INE (1988c, 1989).
114 Susan Horton
countries. Open unemployment rates (6 to 7 percent) are not
particularly high at least until 1988 (unsurprising in view of the lack
of any unemployment insurance system), and rates of under-
employment are also not especially high.1
Table 3.3 gives the sectoral composition of employment for
selected years. According to the World Bank (1988), Bolivia has a
somewhat lower share of its labor force in agriculture and a scmewhat
larger share in manufacturing than other countries with similar income
levels.
Table 3.3 Sectoral Composition of Employment, Selected Years
(percent)
Sector 1970 1976 1980 1986
Agriculture 50.6 48.1 46.5 49.9
Mining . 4.0 3.3 4.0 3.1
Hydrocarbons 0.3 0.3 0.4 0.5
Manufacturing 9.7 10.1 10.3 8.9
Construction 3.7 5.7. 5.5 2.6
Utilities 0.2 0.2 0.4 0.5
Transport 4.0 3.9 5.4 5.6
Commerce 7.2 7.4 7.4 8.2
Finance 0.6 0.6 0.6 0.8
Services 19.7 19.6 19.3 20.0
Tradables 64.6 61.8 61.2 62.4
Nontradables 35.4 37.4 38.6 37.2-
Total 100.0 100.0 100.0 100.0
Notes: Tradables include agriculture. hydroarbons, manufacturing, and mining. All
other sectors are considered nontradables.. Figures are rounded and therefore may not
add exactly to 100 percent
Source: Ministry of Labor unpublished data.
1. The INE's underemployment measure (working less than 12 hours per week) is
not identical with what labor economists might define as underemployment, as some
people voluntarily work short hours, while others who work longer hours may be in
very low paid jobs or jobs not commensurate with their abilities. However, the latter
concept is very hard to quantify empirically.
Bolivia 115
Table 3.4 provides an overview of different labor market sectors.
The state and capitalist sectors correspond roughly to the formal
sector, while small enterprises, family businesses, and domestic servants
correspond to the informal sector. The table illustrates the small
employment share of the private formal sector, which is a third or less
the size of the small enterprise and family business sectors combined,
and also employs fewer people than the state sector. Family businesses
account for the largest share of employment of any sector. There are
some variations between cities: La Paz has a larger state sector than the
other cities, as well as a larger formal private sector.
Table 3.4 also shows that the state sector has the second highest
average age of workers, the highest level of education, the second
highest proportion of migrants, the highest mean incomes, and the
best coverage by benefits. The domestic sector is the exact opposite,
with thc lowest age and mean income, and is evidently an entry sector
for young female migrants. Family businesses also have a
disproportionately large share of female employees, the oldest
workers, low wages, and low education. Those women who manage to
obtain employment in the state and both enterprise sectors require
higher levels of education than men in the same sectors.
Klinov (1987) describes the public sector. This sector is not
especially large in Bolivia (11.2 percent of employment) in
comparison to other Latin American and industrialized countries (18
percent of employment in 1980). However, the sector grew from 9.3
percent of total employment in 1971 to 14.0 percent in 1985 before
falling back below 12 percent in 1986 and continuing to decline in
1987. What is of more concem is the very large share of the public
sector in formal sector employment: 60 to 62 percent in 1980. During
1980-85 the growth in public administration alone accounted for 172
percent of the increase in salaried employment, thus indicating the
substantial weakness of the private formal sector.
The literature seems to have neglected the private formal sector.
Recent policy covcern has focused on small-scale enterprises
(Morawetz 1987; Sauchez 1988). The formal sector is relatively small,
was affected by the hyperinflation, and continues to suffer from the
small and fragmented internal market. It continues to have problems
resulting from the reduced tariff protection since 1985, which
Table 3.4 Employment, Socioeconomic Characteristics, and Gender Differences, by Labor Marrket Sector,
Selected Cities and Years
c'apilatlst Small Family
Category S .ai. enrterprise eiiterprise busi,sess Domesilc Toal
E&ployietnl (perceeii)
La Paz, 1980 23.6 17.2 16.9 36,8 5.3 100.0
Santa Cruz, 1980 21.0 14.9 27.7 27.7 8,7 100.0
Cochabamba, 1983 21.3 13.8 24.3 33A4 7.2 100.0
Santa Cruz, 1986 15.6 19.5 29.0 32.9 3.0 100.0
Cochabamba, 1986 23.3 12.5 21.7 41.0 1.5 100.0
Socloecallorllic characteristics, La Paz 1980
Avcrage age (years) 36.0 33,0 32,0 , 38.0 24.0 35.0
Percentage of migrant3 65.3 59.7 61.0 60.0 79.9 62.5
Education (years) 11.7 8,6 7.4 5,0 3.6 7.5
Income (pcsos) 1,200.0 997.0 861.0 564.0 227.0 835.0
Percntage without benefits 5.4 41.7 79.5 99.9 69.2 63.0
Genider differences, L.a Paz 1Q80
Males (percentage) 27,6 22.2 22.4 27.4 3,9 100.0
Femates (percentagc) 172 8.9 8.1 52. 13.3 100.0
Mean education (years)
Mcn 11.1 8.1 7,3 6,3 5.7 a.2
Women 13.2 10.4 7.7 3,8 3.5 6.3
Source: Cochabamba: Casanovas and Rojas (1988); Santa Cruz: Esscobar do Pab6n and Garcia (1988); La Paz: Casanovas (1987).
Note; Migrants are those who were not born In thc city In which they work. Capitalist enterprises rerers to the private formal sector.
Bolivia 117
combined with the overvalued exchange rate and low labor
productivity render domestic labor costs uncompetitively high.
The informal sector has been much more extensively studied,
perhaps due to its greater size (see Casanovas 1987; Casanovas and
Rojar 1988; CEDLA 1988; CET 1987; Escobar de Pab6n and Garcia
1988; INE and UDAPF. 1987). The sector is fairly heterogeneous,
encompassing small-scale enterprises ([NE's definition is fewer than
five employees), self-employed workers, family businesses, and
(depending on the definiti.. - Jomestic workers.
The informal sector is very large, both in terms of share of
employment and number of economic units.2 By sector of
production, the informal sector is concentrated in services and
commerce: 90.7 percent of informal compared to 47.4 percent of
formal establishments. are in this sector. However, the sheer number of
informal sector establishments implies that they nevertheless
contribute a large proportion (74.2 percent) of the number of
establishments in manufacturing,. including small enterprises and
family businesses (Casanovas 1987). Informal sector manufacturing
establishnments produce mainly clothing, textiles, and thread (over 50
percent of the establishments), with the other products in order of
importance being furniture, food, pottery, metalworking, shoes, and
other (Casanovas 1987).
Relatively little is known about Bolivia's rural labor markets.
Ormachea (1988) and Maletta (1980) synthesize what is available,
mainly from census data, although migration surveys yield some
information. Self-employed farmers on the altiplano practice
traditional agriculture with relatively small landholdings and
traditional farming techniques. The pressure for outmigration is
substantial. In the lowlands commercial agriculture is the dominant
2. Of the 63,289 economic units listed in the first national survey of economic
establishments, 95.1 percent were small (fewer than five employes). If one includes
market stands and ambulant commerciants (people who set up on street corners, not in
licensed spots, or who walk around selling things) as economic units, 97.8 percent of
the 142,469 economic units were small. However, the employment sharc of the
informal sector was smaller-. 44.1 percent of the 245,611 workers if market stands,
and so on are excluded, 58.4 percent of the 330,407 if they are included. This reflects
the much smaller scale of informal compared to formal. sector eaterprises: 1.4 persons
per establishment in the informal sector, versus 44.0 in the formal in 1985 (IN E and
UDAPE 1987)..
118 Susan Horton
form of cultivation. About 76 percent of production in the country as
a whole is traditional and 24 percent commercial (Ormachea 1988).
Migrant labor is important in commercial agriculture. The crops that
require seasonal labor include sugarcane, cotton, rubber, chestnuts,
and grapes. The migrant flows can be relatively large: for example, in
1987 seasonal labor requirements were estimated at 12,000 people in
Santa Cruz (Ormachea 1988) and around 1,900 in Tarija for
sugarcane and 10,000 in Santa Cruz for cotton. These seasonal labor
requirements are largely filled by migrants, who often travel fairly
long distances: of those in Santa Cruz, over half were estimated to
come from other departments. A sizeable fraction of those migrants
surveyed were nonagricultural workers.
There have been a number of studies of urban migration
(Casanovas and Rojas 1988; Escobar de Pabon -and Garcia 1988;
Maletta 1980; PREALC 1988). As discussed above, these flows have
been large in the past, as the geographic location of economic activity
has shifted. Some of the migration has been the usual rural-urban
flows accompanying economic development-the urban share of the
population was 27-3 percent in 1950 and 41.7 percent in 1976
(Maletta 1980)-but urban-urban migration has also been important
and has been growing in recent years. Rural-rural migration is also
significant: in addition to seasonal flows, farm familes have been
encouraged to colonize the eastern lowlands. La Paz and Santa Cruz
have been the main urban destinations for migrants. Little is known
about recent migration flows to coca growing areas, but USAID
unofficial estimates for 1988 suggest that 180,000-210,000 workers
are involved in production and another 30,000 to 50,000 in
transporting, processing, and exporting, presumably an increase over
traditional numbers in this sector.
The domestic service sector is an important entry point for young,
female, rural migrants. However, many migrants obtain salaried jobs:
77 percent in La Paz and 80.6 percent in Santa Cruz in 1987
(PREALC 1988), and until 1985, the state sector was an important
recipient of migrants. Mtigrants- also tend to be overrepresented in
construction PREALC (1988) argues that the onset of the economic
crisis in 1976-80 led to an Increased flow of migrants into services;
prior to this date services and the informal sector had not been an
Bolvia 119
important point of insertion of migrants contrary to many usual
theories of migration.
Stabiization, Structural Adjustment, and the Labor
Market
Relatively few studies on adjustments in the labor market in Bolivia
in the late 1970s and 1980s are available (see Afcha and others 1988;
CET 1988; INE and UDAPE 1987; Morales 1987). It is important to
try to separate two sets of factors, namely, the usual cyclical effects on
the labor market caused by economic conditions (1982-83 was a
recessionary period when the GDP fell over 11 percent) and trend
changes following stabilization and structural adjustment policies
introduced in 1985 However, separating these events is not easy,
especially since data are only available for a relatively short span of
years.
The figures in table 3.2 suggest that mainly cyclical factors were at
work in affecting participation and unemployment rates. In the
recession years (1982-83) participation rates fell (discouraged worker
effect), unraiployment rates peaked, the percentage of unemployed
who had previously worked increasedr and hours worked per week
fell- Detecting an effect due to stabilization or structural adjustment on
these labor market variables is difficult, at least until 1987. The only
discernible trend to 1987 is a decrease in the percentage of salaried
employees Thus, despite economic stagnation, unemployment
apparently did not rise. However, several of the labor market
indicators worsened in 1988 and 1989: unempIoyment,
underemployment, and share of job leavers in the unemployed. This
merits close watch in case it represents a delayed response to the
structurA adjustment measures.3
It is interesting to speculate on the role of the Emergency Social
Fund in alleviating unemployment The fund was first suggested in
1985 (FREALC 1985), at a time when the 1982-83 unemployment
3. However, some caution is required in interpretation as the [NE changed some of
its methods in 1988, including a switch from maiframe computcr processing to use
of microcomputers The fact that the urban population of working age fell in 1989
suggests data problems unless the shift to rural areas and coca growing areas was quite
dramatic.
Table 3.5 Urban Real Wages by Sector, 1970-80, 1982-88
btln sWry of Labor data Aft islry of abor data INE dara a
(1970 a 100) (Afar. 1982 = 100) (1982 e 100)
Decc. D ec. Dc ec Dcc.
Sector 1970 1971 1972 1973 1974 1975 1076 1977 1978 1979 1980 1982 1983 1984 1905 1986 1985 1987 1988
Mining 100.0 102.0 100.0 102,0 82.0 79.0 83,0 81.0 76.0 78.0 76.0 161.8 76.7 117.4 110.0 99.7
76.0 68.3 55.6
Hydrocarbons 100.0 102.0 100,0 102,0 82.0 142.0 189.0 209.0 263,0 248.0 249.0 74,4 48.5 199.8 66.0 66.4
Monufacturhng 100.0 113.0 114.0 123.0 99.0 98,0 114.0 111.0 111.0 114.0 102,0 81.3 113.6 136.3 51,8 58.3 52.5 42.9 31.9
Construction 100.0 119.0 116,0 109.0 97,0 90,0 flQ.0 115.0 115.0 131.0 121.0 51.5 47,3 92,5 51,9 73.5 101.9 84.7 76.7
UtilitIes 100.0 87.0 82,0 85.0 69.0 91.0 92.0 93.0 97.0 84.0 84.0 82.9 43.1 88.9 47.0 74.3 64,5 63.9 68.0
Transport 100.0 104.0 102,0 122.0 96.0 96.0 102,0 112,0 120,0 120.0 120.0 56.7 44.8 55.8 68.5 57.9 89.2 71.2 59.0
Comrmrce 100.0 108.0 105.0 117.0 83.0 79.0 101.0 107.0 114.0 110.0 91.0 69.0 56.9 76.4 48.2 62.5 43.5 43.8 23.5
Finance 100.0 98.0 94.0 73.0 46.0 43.0 42.0 53.0 60.0 59.0 53.0 89,8 65,9 108.9 95.1 129.5 69,4 64.7 47.6
Services 63.4 57.1 103.6 47.5 72.1 n.a. n.m. n,a,
100.0 102.0 100.0 104.0 87.0 70.0 76.0 92.0 89.0 84.0 80,0 69.3 27.9 50.5
Public
admlnlstration 51.9 45,8 64.8. 46.0 32.7
Average for nil
sectors 100.0 114.0 116,0 118.0 94,0 86.0 98.0 107.0 108.0 110.0 100.0 75.2 63.4 99,1 60.4 74,2 58.8 51.8 38.7
n.a. = not available
Note: Tradables include manufacturing, hydrocarbons, and mining.
a. INE data are for Eje Central only.
Sourrces: 1970-80: Delons and Bour (1988); 1982-86: Afcha and others (1988); 1986-88; authors' calculations from INE and
Encuesta Pernanente de Hogares (ElPH) data, deflated using consumer price Index of June.
Bolivia 121
data were becoming available. At that time unemployment was rising,
and the percentage of job leavers and heads of household among the
unemployed was increasing. At the same time, the government
anticipated that the labor market measures included in the structural
adjustment decrees would create further unemployment. However, the
fund did not become significant in terms of job creation until
1987188.
Table 3.3 shows some disturbing trends in sectoral employment.
During 1970-80 the sectoral shifts were of the kind usually associated
with economic development, namely, a shift out of agriculture and
into industry. This trend was abruptly reversed in the 1980s, with a
shift back out of industry into agriculture, commerce, and services.
This is of concern if it implies a crowding of displaced workers in
relatively low productivity and low remuneration activities. Another
way of interpreting these data is to group sectors into tradables
(agriculture, mining, hydrocarbons, and manufacturing) and
nontradables (all other sectors). On this basis, Bolivia's inward-
oriented economic policies pror to 1986 were associated with a
continuous shift out of tradables; thereafter the trend was reversed.
However, coca production may also be responsible for the increased
share of agriculture, and hence of tradables.
Table 3.4 shows the shifts between market sectors in two of the
three major cities over time. In both Cochabamba and Santa Cruz, the
family business sector gained.in relative employment share. In Santa
Cruz, which has tended to benefit from structural adjustment, both
enterprise sectors (capitalist and small) gained in employment share,
but they lost out in Cochabamba.
Table 3.5 presents data on real wages.4 Note that the data for the
hyperinflation years of 1984 and 1985 are particularly unreliable,
therefore, apparent sectoral differences in 1984 and 1985 should not
be ascribed undue significance. The net effect of the 1970-80 decade
was of no change in real wages: the onset of the economic crisis
4. I have been unable to obtain the original Ministry of Labor data and have had to
rely on indices calculated in secondary sources, namely. Delons and Bour 1988; Afcha
and others 1988; Muller and Machicado 1986. It is not possible to link the data for
the 1970s and 1980s, since the sectoral data for 1981 are not available in these
secondary sources, and it is not always clear which month of the year is being used.
122 Susan Horton
eroded the gains made up to 1978. As concerns sectoral differences,
the hydrocarbon sector did relatively well, while commerce and
services did relatively badly.
In the early 1980s real wages began to faR substantially. Using the
Ministry of Labor data, and assuming that the Muller and Machicado
index for 1981 represents a point near to the end of the year and can
be linked to the March 1982 base for table 3.5, then at a conservative
estimate real wages fell by 20 percent of their 1980 value by the end
of 1981, and by a further 2-5 percent of their March 1982 value by
December 1986, that is, by 1986 real wages were around 59 percent of
their 1980 (and hence also 1970) value. If instead we use the INE
series for the period 1982-86 (and arguably this is better, due to the
break in the Ministry series due to the consolidation of the salary
structure in 1985), then real wages in 1986 were 52 percent of their
1982 value, and thus 41 percent or less of their value in 1980. Both
sets of data agree that real wages fell substantially, and apparently far
exceeded the fall in per capita GDP during the same period (around
73 percent of its 1980 value by 1986).
Both the ministry and INE series show similar sectoral patterns of
change in real wages. The sectors that fared the worst were
manufacturing, transport, public administration, and (according to
INE) commerce. Manufacturing obviously suffered from the fall in
tariff protection, transport from deregulation, and public
administration from the drastic labor shedding. The explanation for
commerce is most likely the "crowding" one, namely, that labor
shifted into the relatively unremunerative sectors in the informal
sector.
Rough, separate estimates of indexes for real wages in the tradable
and nontradable sectors show that the tradable sector fared
consistently worse during 1982-88. Note, however, that no wage series
is available for the agricultumal sector, the one tradable sector that was
gaiig in employment share. Nevertheless, the data do suggest that
the shift in incentives for manufacturing due to the change in the
exchange rate was not large enough to offset the adverse effects of
decreased protection for this sector.
Table 3.6 provides some information on the formal and informal
sectors. The informal sectors are defined below the table. The formal
Bolivia 123
Table 3.6 Share of Employment in Informal Sector, Urban Areas, and
Median Earnings, Selected Years
(percent)
Sector 2982a 1983a 1984" 1985a -- 1988k
Employment share
Mining 16.1 n.a. 9.3 n.a. 20.6
Manufacturing 59.2 n.a. 66.7 n-a- 68 2
Construction 61.4 n.a. 63.5 na. 56-1
Util;ies 6.8 n.a. 14-Y n.a. 8.7
Transport 70.8 n.a. 67.7 n.a. 70.3
Commerce 83.6 n-a. 90.6 n.a. 89.9
Finance 40.9 n-a- 29.1 n.a. 40.7
Services 45.5 n.a. 42.8 n.a. 46.4
All sectors 56.9 61.1 60.5 .56.3 64.3
Median earnings
Informal as percentage
formal 54.3 n.a- 693 80.0 60.0
n.a = not available
Nores: The informal sector is defined as self-employed workers, excluding
professionals; domestic employees; and employees, employers, and unpaid family
workers in establishments with fewer than five workers. Earnings are for primary
occupation only. As a percentage of earnings in the formal sector, median eamings in
the informaI sector were 54 percent in 1982, 69 percent in 1984, and 60 percent in
1988.
a. Includes data for eight cities, all department capitals except Cobija.
b. Cities of Eje Central only.
Sources: 1982 and 1984: INE and UDAPE (1987); 1988: author's calculations.
sector includes those working in establishments of six or more people
and professionals. The data suggest that there has been a general
increase in informalization over time, particularly in manufacturing
and commerce. Casanovas (1987) confirms that the number of self-
employed workers rose much faster than the size of the labor force
during the 1980s: he estimates that the informal sector in La Paz
124 Susan Horton
increased from 47 percent of employment in 1976 to 53 percent in
1980 and 58 percent in 1984. We can combine this information with
INE and UDAPE (1987) data that indicate that the median size of
informal sector establishments actually fell during the 1980s from
2.06 people in 1982 to 1.75 people in 1985, while the median size of
private formal sector establishments increased from 13.0 people in
1983 to 40.9 people in 1985. One. interpretation is therefore that there
was increased concentration in the formal sector as smaller formal
sector establishments went out of business and people shifted.into the
informal sector, pactiularly into one-person enterprises.
Anecdotal evidence on the shift from formal to informal sector
employment particularly into petty commerce activities, is also
available. The mechanism for worker dismissal often encouraged such
a shift. The miners and some other public employees who were
dismissed were eligible for fairly large severance payments, extending
up to two or three years wages for miners (CET 1988). In some cases
these payments were made in dollars. Apparently a number of those
dismissed used these severance payments to purchase imported
consumer goods and set themselves up in trade and commerce
Figures on formal versus informal sector earnings (see table 3.6,
notes) suggest that informal sector earnings rose relative to formal
sector earnings during the hyperinflation, but had dropped back again
by 1988 (those on fixed wages and salaries would tend to lose out
during rapid inflation so this finding is not surprising).
Thus, the evidence suggests a partem of labor market adjustment
during economic crisis, with as yet little in the way of recovery and
resumption of economic growth. The consequence of prolonged
stagnation and very severe recession in 1982-83 was a fall in
participation rates, a fall in the proportion of salaried employment, an
increase in informalization of economic activity, and a reversal of the
sectoral shifts that usually accompany economic development.
However, on a more optimistic note, there seemed to be a reversal of
past trends in which labor had consistently shifted out of tradables.
Displaced labor from the formal sector has moved into self-
employment, into one-person establishments, and frequently into petty
commerce activities. The evidence does not suggest that the labor
market was rigid and impeded adjustment. In addition, the decline in
Bolivia 125
real wages is particularly important in a relatively poor country such
as Bolivia.
Implications of Labor Market Adjustment
LaborMarketAdjustment and lncome Distribution
Work on income distnrbution in Bolivia is hampered by the lack of
national survey data and the very fragmentary information available
for the rural sector. A number of studies are available that deal with
poverty (for example, R. Morales 1985, 1987). The INE household
surveys can be used to study the distribution of urban income, but the
literature is somewhat frustrating in this respect Afcha and others
(1988) compare data for 1982 and 1985, but do not state whether
they are using constant or current prices INE and VDAPE (1987)
compare 1982, 1984, and 1985, but in current prices, and only for the
informal sector. Finally, the INE is now publishing information on
income distribution (INE 1988b), but only since 1988. Moreover, all
these studies deal only with the distribution of earned income across
individuals.
To give some idea of trends over time, figure 3.2 presents Lorenz
curves for 1982, 1985, and 1988 for the distribution of earned income
on a personal basis. Variation in income distribution over short
periods of time is relatively unusual, yet noticeable changes are
evident here. The hyperinflation apparently improved the distnbution
of income for the vast majority of the population, although the bottom
15 percent lost out slightly and the top 10 percent gained.
Table 3.7 compares the personal distribution of earned income
with the household distnrbution of earned income and the household
distribution of total income for 1988. The results show that the
distribution of earned income on a household basis is slightly less
unequal than on a personal basis, particularly at the extreme deciles.
The inclusion of unearned income also tends to improve income
distribution slightly (however, data on unearned income are probably
unreliable, especially in the top deciles). Per capita and per household
distributions are not consistently different Apparently household size
is not systematically related to household income, unlike in some
other countries. Overall, the data suggest that the distribution of
126 Susan Horton
Figure 3.2 Lorenz Curves for Personal Distribution of Earned
Income, Selected Years
100
80
60
40-
20
0*
0 20 40 60 80 100
% Population
personal earings on an individual basis is a reasonable guide to urban
income distribution.
These data, however, only provide part of the story on income
distribution. The shift from labor earnings to other types of income
(as evidenced by the much larger fall in real wages than in per capita
GDP) would likely worsen income distrbution, but the shift in favor
of rural areas would have the opposite effect.
Bolivia 127
Table 3.7 Urban Income Distribution Data, 1988
(percentage of total income accruing per decile)
Earned income Total hicome
Earned income
Decile Per householda Per capitab Per housedolda Per capitab per individuatC
1 . 1.4 1.0 1.1 1.2 0.9
2 1.9 2.1 2.2 2.1 2.2
3 3.4 2.8 3.4 2.9 3.2
4 3.5 4.0 3.8 4.3 4.6
5 5.0 5.0 5.3 5.3 5.0
6 6.4 7.0 .6.2 6.9 6.7
7 8.5 8.1 9.1 8.2 8.5
8 11.7 11.9 11.4 12.4 10.4
59 16.4 16.9 16.6 .16.6 15.7
10 41.8 41.2 40.9 40.0 42.8
Bonom 40% 102 9.9 10.5 10.5 10.9
Middle 40% 31.6 32.0 32.0 32.8 30.6
Top 20% 582 58.1 57.5 56.6 58.5
a. Deciles calculated on the basis of household income.
b. Deciles calculated on the basis of income per capita, for households.
c. Deciles calculated on ,;: basis of individuals' own earnings.
Source: Author's calculations.
Labor Market Adjustmnent and Regional Development
As indicated earlier, structural economic changes in Bolivia
frequentIy imply regional shifts in employment and economic
activity. Between 1900 and 1950 the major shifts were toward the
altiplano, the center of mining and traditional agricultural activities.
Since 1950, and especially since the crisis in minerals, the eastem
lowland provinces have gained in importance, both for industry and
for commercial agriculture. This has important consequences, for
migration and for the political economy.
Table 3.8 provides some information on l'1or force indicators by
city. Larger cities tend to have higher participation rates and a higher
percentage of nonsalaried workers. Evidently the greater level of
Table 3.8 Labor Force Indicators by City, Selected Years
Mean monthly
earhinfs as
Percentage percenrtae
Popiilatiolf Participalion Utemploymtent Percenta e in informal of meanjfor
over age )0 ('000) rate (Cq) rate (%) salaried sector Eje Central
City 1980 1987 1980 1987 1980 1987 1980 1987 1988 1981 1988
Total
(7 cities)a 1,321 1,830 49.8 46.3 7.3 9.2 59,3 54*3 65.0b 100o.b 100.0b
La Paz 600 789 54,8 49.6 9,6 11.9 56.9 54.2 70.4 81.7 66.4
0 Cochabamba 193 264 47.4. 42.5 7.4 4.8 57.8 56.4 57.6 104.0 123.3
Santa Cruz 250 430 48.0 48.4 2.0 5.3 63.4 53.1 62.6 136.2 143.7
Oruro . 112 143 45.0 35.7 7.5 16.7 56.5 47.1 61.0 82.5 80.3
Sucre 58 69 41.3 41.7 11.3 8.2 65.6 60.7 n.a. n.a. n.a.
Potosi 69 83 34.7 38.0 1.2 9,1 70.4 60.5 n.a. n,a. n.a.
Tarija 39 53 48.5 47.6 4.9 5.1 63.4 55.8 n.a. n.a. n.a.
Trinidad 23 ri..&. 42.6 n.a. 1.3 n,a, 69.7 n.a. n,a. n.a. n.a.
n.a. = not available
Notes: 1981 data nccds weighting. Earnings data are for primary and secondary occupations combined.
a. Includes all department capitals except Cobija and Trinidad.
b. Includes only the four cilies of the Eje Centtral.
Sources: Population, participation, unemployment, and pcrccntagc of salaried workers: INE (1988c); femainder: author's
calculations,
Bolivia 129
overall economic activity in larger centers is better able to support
informal sector activities. The most prosperous cities of the Eje
Central are Santa Cruz and Cochabamba. The altiplano cities of La
Paz and Oruro have markedly lower levels of earnings per person.
The recession had different effects on different cities. The highest
unemployment rates in 1987 were in the three altiplano cities, namely,
Oruro and Potosi, the two worst hit by the problems in mining, and La
Paz, the immediate destination of most altiplano migrants-
Participation rtes in 1987 were the lowest in the two mining cities,
Oruro and Potosi, which might suggest discouraged worker effects.
Eanings trends suggest that La Paz fared the worst relative to the
other cities of the Eje Central, again implying substantial immigration
to the informal sector in La Paz.
These regional shifts imply some future adjustment problems. The
eastern cities are in an area where the transport infastructure is less
well developed and where exporting costs are higher. In addition, the
government has apparently. encountered difficulties in shifting the
burden of taxation to the newer economic activities in the east In the
past, taxing the nationalized mining sector was easier, notwithstanding
the occasional clashes between the mining unions and the government.
Another somewhat unusual feature is to have the capital city located in
an area of declining economic activity.
Labor Market Adjustm7ent and Women
Most experts believe that women tend to be especially
disadvantaged during the process of structural adjustment As more
marginal employees, they are more likely to lose their jobs during a
xecession, at the same time as recession may force more women into
the labor market to supplement falling family incomes. A
counterargument would be that women are less likely to be located in
the public and formal sectors that are worst hit by structural
adjustment
The literature suggests that Bolivian women have a worse position
in the labor market than men. Casanovas' (1987) data (table 3.4)
show that women are concentrated in lower paying sectors with fewer
benefits, and, that to- be able to break into the formal sector women
require more education than men. Fernandez (1988) reports that in La
Table 3.9 Urban Labor Market Indicators by Gender, Eje Centtral, 198088
Variable 1980 1981 1982 1983 1984 1985 1986 1987 1988
Puarcipadon tats <a)
Men 67.1 n.n. 58.3 n.a. 61,8 n.n. 58.0 59.6 58.6
Women 35.7 n,n. 26.9 n.a. 38.6 n.a. 32.8 35.2 40.0
Unemployment rate ($)
Men 6.4 n.n. 9,0 n.a. 7.8 n.a. 5.6 6.6 1 1.5
Women 6.2 n.n, 5.3 n.n. 4.9 n.ms 1.9 4.6 11.5
Underemployment rate (%)
Men 7,5 nAl. 4,4 n.a. 2.7 n.D. 1.5 2.4 5.1
Women 11.6 n,a. 5,5 n,a. 6.4 nal. 3.3 5.3 9.5
Percentage of saried employees
Men 66.1 n.a. 60.3 n.n. 59.8 n.a. 60.4 57.6 60.2
Womcn 46,7 nAa. 56.5 n.a. 45.0 n.a. 47.4 47.7 46.4
Men's eamings as percentage of
women's earnings n.n. 205,0 108.09 n,n 169.9 n.n, 169.7 17S,2 199.7
Percentage of women fn sector
Mining & hydrocarbons 4.9 n.a. n.,n na. n,a. n.a. n.a. n.m. 1t.4
Menufacrurlng 32,5 n.a, n.n. n.a. n,a. n,a. n0a. n.a. 31,5
Construcilon 2.0 n.a. n.e. n.a. n,a, n,D. n.e. n.a. 8.9
Utilitles 18.9 n.m. n,a. n.a. n,a. n.,. n.a. nA.l 2.5
Transport 6.6 *,a. n.a. n.. n.a. n.a. n.a. n.a. 7.1
Commerce 66.3 n.a. na, u.n. n.a. n.a. n.a. n.a. 65.6
Finance 28.9 n.n. na. n. n.n. n.a. n.a. n.a. 21.6
Scrvices 42.8 u.n. nAa. n.a. n.a. n,a, n.n. n.a. 45.6
Fennal n.a. na. n.a. n. n.e. n.e. n.a. n.a. 25,0
Infommial n.e. n.a. na. n.a. n.a, n,a. n.a. n.a. 49.4
Tolal 38,3 n.n. n.a. n,n. na. n.a. n,a. n,a. 42.8
n.a. = not available
Note: 1981 data need weighting.
Sources: 1980, 1982, 1984, 1986, 1987: INE calculations; 1981, '1988: author's calculations.
Bolvia 131
Paz, 40 percent of women work as individuals, three-quarters of these
in commerce. He also describes the precarious situation of women in
El Alto, the poorest suburb of La Paz and destination of most low-
income migrants to the city. In El Alto, a 1987 survey showed that 34
percent of working women earned less than Bs 50 a month, at a time
when the weekly minimum wage was Bs 200. The survey also showed
that women were engaged in additional work in the home designed to
supplement their inadequate household incomes. Of households in El
Alto, 6 percent grow food (quite a feat in a suburb), 18 percent keep
chickens, and 8 percent keep rabbits; all predominantly female
activities.
As regards the impact of the recession on women, CET (1988)
argues that those subject to dismissal from the formal sector were
more likely to be women because of the higher costs of maternity
leave and their allegedly lower productivity. However, the biggest
group among those dismissed were miners, of whom only 5 percent
were women.
Table 389 analyzes urban labor market indicators by gender.
Female participation rates tend to be lower than those for men, but
unlike the rates for men have not generally declined over time.
Possibly Bolivia is following (belatedly) the secular trend elsewhere in
Latin Amerca for incrased female labor force participation. Also, the
hyperinflation might have led to increased partcipation of secondary
income eamers so that families could make ends meet The table also
shows that female unemployment rates tend to be higher than for
men. The latter might be in line with the luxury unemployment view,
that is, secondary family workers can more readily afford to be
unemployed. Women work- on average somewhat shorter hours than
men, but this may reflect the choice of part-time work by some female
workers. In general, women are overrepresented in family businesses
and domestic service, where-hours are very long. Fewer women than
men are salaried, but unlike men, there was not a substantial fall in the
proportion salaried between 1980 and 1988. As far as the relative
male-female earnings differential is concerned, there is little evidence
of change over time (the 1982 data are an outlier and are perhaps
unreliable).
132 Susan Horton
The sectors that are most heavily female are commerce, services,
and manufacturing. In 1988, women accounted for half of informal
sector but only a quarter of formal sector employment, which explains
in large part the much lower average wages of women.
Thus, women are at a disadvantage in the labor market in Bolivia
compared to men, as they are in most countries, especially developing
countries. Undoubtedly women and children suffered substantially
during the economic recession and stagnation; however, there is no
evidence that structural adjustment per se has had especially advert
effects on women in the labor force in Bolivia.
Labor Market Adjustment, Political Economy, and Effects on Longr
Run Economic Growth
Political economy is very important in understanding the persistent
inflationary tendencies and problems in adjusting in many Latin
American countries, and this is definitely true in the case of Bolivia
We have already referred to the political economy's contribution to
the hyperinflation. The political economy of adjustment is a very
large topic that this paper can only touch on.
With a relatively strong right wing government in power between
1985 and 1989 that was willing to use displays of police and military
power where necessary (although using the Catholic Church as a
mediating factor), the population accepted the strong economic
measures undertaken in the stabilization and structural adjustment
programs- Despite strong unions, the govemment liberalized the labor
market substantially and real wages fell dramatically. However, only
time will tell whether the government can maintain the current
consensus unless more vigorous economic growth resumes soon. The
rise in unemployment in 1988 is a worrisome sign.
As regards the working of the labor market and its effects on the
resumption of long-mn economic growth, the labor market seems to
have been relativ ely flexible in terms of the substantial sectoral shift,
regional migration, relative wage changes, and real wage declines
observed. The blame for lack of resumption of sustained growth does
not seem to lie in this sector.
One way to analyze labor market functioning in more detail is to
use earnings functions. Table 3.10 contains some simple earnings
Bolivia 133
Table 3.10 Urban Earings Functions, 1981 and 1988
- ~~~~Eornbtgs funtons
cwfficicmt ~~Variable mean
Independent
variable. 1981 1988 1981 1988
Education (yrs) 0.127 0.0951 8.091 9.694.
(55399)*- (25.835)- (4.936) (4.819)
Experience (years) 0.053 0.0322 21.170 20511
(24.435)'*' (9:307)y* (14.682) (14.155)
Experience2 (years2) -.000640 -0.000308 663.724 621.029
(17-358)** (4.911)0-' (816.637) C759.063}
Female dummy -0327 -0.234 0.384 0A4i1
(16.314)--* (7.457)0*0 (0.486) (OA93)
Unmarried dummy -0.159 -0.022 0Q368 0339
(7.=)*** (0.639)*** (0.482) (0.474)
Informal dummy -0.156 -0.0558 02339 0.347
(7.430)" (1-704) (0.473) (0.476)
Onmi -0.0334 -0.262 0.849 0.162
(0t98) (6.032)-s (0.279) (0368)
Cochabamba 0.199 0.136 0266 0.248
(8.924)*' (3528)*** (0.442) (0.432)
Santa Cruz 0.W47 0339 0.214 0.233
(20.607)** (8.720)*** (0.410) (0.423)
Narial log
(haurly wage) - 2.853 0.183
(Caent sL) (1.002) (0977)
Intercept 1.238 -1-153 -- -
(29.-82)* (16.145y'*w
Adjusted p7 0.478 0253 - -
Fsatistic 66337410* fl7992gf
degrces of freedom 9,6502 9,3364 - -
- = not applicable
* implies significant at 10 percent level
* * implies significant at 5 percent level
implies significant at 1 percent level
a. Figures in parentheses are t statistics
b Figures in parentheses are standard deviations.
Source: Author's calculations.
134 Susan Horton
functions for 1981 and 1988. (Note that use of the semi-log
functional form implies that coefficients represent the effects of a
change of one unit in the independent variables as a percentage
change of wages.)
The equations are well behaved, with the usual positive effect of
education and experience, lower wages for women and those not
married, lower wages in the informal sector, and with substantial wage
differentials between cities. What is quite strikdng is the change in the
equations over time: between 1981 and 1988 the effect of all personal
characteristics declined: the rate of retu-rn to a year's dducation fell
from 12 to 10 percent, of a year's expenence from 5.3 to 3.6 percent,
the male-female differential fell, and the effect of marital status
declined and became less significant. The formal/mformal differential
declined from 16 to 7 percent. The explanatory power of the
equations also diminished. One possible interpretation is that the
1980s substantially disrupted the traditional working of the labor
market, and particularly of the formal sector (implying falls in the
wages of relatively privileged groups such as men, the educated, and
those married). The topic merits fiurther investigation elsewhere.
Conclusion
This chapter has covered a great deal of ground and includes a
substantial amount of new empirical work using microeconomic data-
Few previous studies have systematically attempted to study changes in
the labor market or income distnbution over the 1980s for Bolivia.
However, much work remains to be done. The magnitude of the
changes in economic institutions, for example, the somewhat daring
attempt to open the capital account, make Bolivia a very interesting
case for further study. The chapter has focused instead on miLcro
issues regarding the labor market, themselves of great interest in view
of the large changes that occurred in just a few years, and indicative of
substantial ability to adjust even in a relatively poor country with
substantial economic distortions.
Previous economic studies of Bolivia have tended to focus on the
macroeconomic aspects. Bolivia's economic crisis had a rather earlier
onset than for other Latin countries (in 1978), although it took seven
years for stabilzation to be successfully implemented, and the ensuing
Bolivia 135
structural adjustment packages have as yet to yield any worthwhile
benefits in terms of substantial growth, let alone employment and real
wage increases. The structural adjustment measures took an almost
textbook form: freeing of markets for foreign exchange, foreign
capital, trade, finance, goods, and (of most concern here) labor, severe
public sector employment cuts; and fiscal restructuring. Current
causes for concern are the high level of real interest rates, the
worsening trade and payments balance, sharply rising unemployment
in 1988, and the open capital account. However, the size of the illegal
(drug) economy makes it important to qualify analysis based solely
on reported economic transactions.
The structure of the labor market prior to the onset of crisis and
adjustment reflected previous government policies. Public sector
employment dominated the formal sector, and a large, and relatively
low-income, informal sector existed. The 1980s sequence of
prolonged recession interspersed with hyperinflation, a fairly drastic
stabilization. and far-reaching institutional changes designed to
encourage structural adjustment has had large effects on the labor.
market. Sectoral employment shifts, both out of secondary and into
primary and tertiary activities and out of formal and into informal
activities, have been large, and in the opposite direction from that
usually accompanying economic development. Despite prolonged
stagnation there was no trend increase in unemployment, fall in hours
worked, or fall in labor force participation rates, although a cyclical
trough occurred in 1982-83- The labor market adjustment took the
form of sectoral shifts and falls in real wages of 40 to 50 percent or
more since 1980. The government employment creation program can
be given relatively little credit for mopping up surplus labor before
1988. However, it may have served a more important role in 1988 and
1989 as unemployment indicators worsened sharply in 1988 despite
modest positive GDP growth, perhaps indicating a second round of
labor market ddjustments due to structural adjustment.
The changes in the labor market have ramifications elsewhere in
the economy. The distribution of earned income has worsened
(despite the very short time series for which data are, available), losing
any improvements that were caused by hyperinflation. Given that the
fall in real wages is about twice the magnitude of the fall in real GDP,
136 Susan Horton
the likelihood that overall urban poverty has increased is almost
inescapable. However, apparently women were not the big losers that
some have theorized to be the case in developing countries
undergoing structural adjustment. A wide disparity is evident in the
experience of different cities and regions of the country during the
economic difficulties. The declining areas, including the capital, have
fared badly relative to the newer eastern lowland cities. This in turn
has political economy ramifications, as the tax base and location of
population also has to shift. Finally, regarding the prognostications for
future growth, the evidence suggests that the labor market has
performed relatively well in adjusting, and that the blame for
inadequate recovery lies elsewhere, possibly in some macropolicy
decisions, bat mainly in the lack of external resources to fuel a
successful recovery. Bolivia has followed an almost textbook path in
the kinds of adjustment policies adopted. Unless growth resumes in
the near future, the social and human costs incurred to achieve these
adjustments will not have been worthwhile.
APPENDIX A
SOURCES OF LABOR MARKET DATA
Most of the data on which studies of the Bolivian labor market are
based are from urban areas; only the censuses of 1900, 1950, and
1976 cover rural areas as well. This is now changing as the current
urban labor force survey was expanded into a Living Standards
Measurement Survcy, which was expanded to cover rural areas
commencing in 1990.
The urban studies are based mainly on data collected by the INE,
whose surveys include a household labor force survey conducted
annually since 1980, a survey of small establishments conducted
several times since 1983, and some special surveys of self-employed
workers conducted in La Paz in 1983 and in Cochabamba and Santa
Cmz in 1986. Aggregate data collected by the Ministry of Labor used
to be published in an annual yearbook. There are also a number of
one-time surveys on particular topics, such as the Ministry of Labor's
1980 study (in conjunction with the International Labour
Organisation) of migration, studies of migration in particular areas or
industries, and studies of displaced workers.
Caution is needed when interpreting the data. The Ministry of
Labor's data are based on information from different sectors, and the
coverage of the informal sector is particularly weak. Hence its
unemployment data (which show a large increase in the 1980s) are
particularly suspect, much of the so-called unemployment probably
represents a shift into informal sector activities that the ministry simply
misses. Similarly, the ministry series on real wages is highly suspect
after 1985: as the ministry collects mainly basic salary data (excluding
bonuses), the incorporation of the bonuses into the basic salary in
1985 means that the ministry data tend to underestimate the fall in ieal
wages.
The INE household surveys seem more reliable on unemployment
and wage data, but also have a couple of disadvantages. One is that the
137
138 Susan Horton
surveys took place at different times of the year, hence an
unmeasurable seasonal effect may be present. Another is that the data
only go back to 1980; data from earlier surveys between 1976 and
1979 are not readily available.
References
Afcha, G., G. Huarachij, R. Pereira, and F. Valverde. 1988. La
Politica de Shock Antinflacionario y el Mercado de Trabajo:
El Caso Boliviano. La Paz: Unidad de Analisis de Politicas
Econ6micas. Discussion paper for presentation at workshop
on anti-inflation policies, Programa Regional del Empleo para
America Latina y el Caribe. Santiago, Chile.
Casanovas S., RI 1987. "El Sector Familiar en la Ciudad de La Paz."
In J. P. Perez S., R. Casanovas S., J. Alvarado, J. C.
Ribadeneira, and M. Chiniboga, eds., Familia y Trabajo en la
Ciudad Andina. Quito, Ecuador. Centro Andino de Accion
Popular.
Casanovas S., R, and A. Rojas R. 1988. Santa Cruz de la Sierra:
Crecimiento Urbano y Situaci65n Ocupacional. La Paz: Centro
de Estudios para el Desarrollo Laboral y Agrario/Centro de
Informaci6n y Documentacion de Santa Cruz.
CEDLA (Centro de Estudios para el Desarrollo aboral y Agrario).
1988. El Sector Informal Urbano en Bolivia. La Paz: Beatriz
Cajias.
CET (Centro de Estudios del Trabajo). 1987. Sector Informal y-
Movimiento Obrero. Temas Laborales No. 2 La Paz.
- 1988. La Relocalizaci6n. Temas Laborales No. 5. La Paz.
Cottani, J. 1988. "Exchange Rate Trends, 1960-87." Washington,
D.C.: World Bank. Draft, processed.
Delons, J. R, and 3. L. Bour. 1988. Empteo, Recursos Humanos e
Ingresos en Bolivia: Una Propuesta para la Acci6n. La Paz:
Unidad de Anilisis de Politicas Econ6micas.
Donoso, S. 1988. "Politicas, Actividades y Estudios sobre Empleo."
La Paz: Unidad de An;iisis de Politicas Econ6micas. Draft.
Bolivia 139
Doria Medina, S. 1987. La Quimera de la Reacrivacidn: Balance y
Perspectivas de La Econ6mica Boliviana. La Paz: EDOBOL.
Escobar de Pab6n, S., and C. L. Garcia. 1988. Urbanizacidn
Migraciones y Empleo en la Ciudad de Cochabamba. La Paz:
Centro de Estudios para el Desarrollo Laboral y
Agrario/Centro de Informaci6n y Documentacion de Santa
Cruz.
Fernandez M., M. 1988. "Insercion Laboral, lngreso y Estrategias
Ocupacionales de la Mujer Popular de El Alto de La Paz." La
Paz: Centro de Promocion de la mujer "Gregoria Apaza?
Draft, processed.
IMF. Various years. International Financial Statistics- Washington,
D.C
INE (Lustituto Nacional de Estadistica) 1988a. Encuesta Permanente
de Hogares 1987. La Par Ministerio de Planeamiento y
Coordinacio6n, Area Sociales n. 1-88.
. 1988b. Encuesta Permanente de Hogares 1988. La Paz:
Ministerio de Planeamiento y Coordinacion.
- 1988c- Principales Resultados de la Encuesta Pernanente
de Hogares 1980-87- La Paz: Ministerio de Planeamiento y
Coordinaci6n.
- 1989. Encuesta integrada de Hogares 1989. La Paz:
Ministerio de Planeamiento y Coordinaci6n.
INE (Instituto Nacional de Estadfstica) and UTDAPE (Jnidad de
AnOlisis de Polfticas Economicas). 1987. Un Intento de
Medicicn del Sector Informal Urbano en Bolivia. La Paz:
UDAPE.
Klinov, R. 1987. "Public Sector Wages and Employment in
Bolivia." Washington, D.C.: World Bank. Draft, processed.
Maletta, H. 1980. La Fuerza de Trabajo en Bolivia 1900-1976:
Analisis Critico de la Informacidn Censal. Proyecto de
Migraciones y Empleo Rural y Urbano BOL178/P03. La Pazr
Ministerio de Trabajo y Desarrollo Laboral.
140 Susan Horton
Mercado S., A. F., M. Fernandez, and T. Reinaga. 1988. "La
Relacion Precios-Salarios: El Caso Boliviano (1982-1985)."
La Paz: Instituto Nacional de Estadistica. Draft, processed.
Morales A., J. A. 1987. Precios, Salarios y Politica Econdmica
Durante la Alta Inflacidn Boliviana de 1982 a 1985. Estudio
Diagnostico Debate. La Paz: Instituto Latinoamericano de
Investigaciones Sociales.
Morales A., J. A., and J. Sachs. 1988. Bolivia's Economic Crisis.
Working Paper No. 2620. Cambridge, Massachussetts: NBER.
Morales A., R. 1985. La Crisis Econdmica en Bolivia y su Impacto en
las Condiciones de Vida de los NiVios. La Paz: UNICEF.
Morales A., R. 1987. Bolivia: Efectos Sociales de la Crisis y de las
Politicas de Ajuste. La Paz: Instituto Latinoamericano de
Investigaciones Sociales. Estudio Diagnosuzco Debate.
Morawetz, D. 1987. Exportaciones de Productos Manufacturados de
Bolivia: Una Perspectiva mas Optimista? La Paz: Unidad de
Analisis de Politicas Econ6micas. Processed.
Muller y Asociados. 1988. Estadisticas Econ6micas 1988. La Paz:
Instituto Latinoamericano de Investigaciones Sociales.
Muller y Machicado Asociados. 1986. Caracteristicas Estructurales
del Empleo y la Evolucion del Salario. La Paz. Confidential
report Processed.
Newman, 3., S. Jorgenson, and M. Pradhan. 1990. "How Did
Workers Benefit from Bolivia's Emergency Social Fund?"
Washington, D.C.: World Bank. Processed.
Ormachea, E. 1988. Apuntes Relativos al Empleo Rural. Documento
de Trabajo Analitico DTA-0176/88. La Paz: Unidad de
AnAisis de Politicas Econ6micas.
Programa Regional del Empleo pan America Latina y el Canbe.
1985. Bases para la Formulaci6n de un Programa de Empleo
-de Emergencia Bolvia 1985-86. Documento de Trabajo
PREALC/268. Santiago, Chile.
Bolivia 141
* 1988. Migracicn y Empleo en Bolivia: Los Casos de las
Ciudades de La Paz y Santa Cruz. Documento de Trabajo
PREALC/321. Santiago, Chile.
SAnchez, C. E. 1988. "La Pequefna y Mediana Empresa Industrial en
Bolivia: un Diagnostico y Recomendaciones de Poiftica." La
Paz: Unidad de Anglisis de Poifticas Econ6micas. Processed.
Villegas Quiroga, C. 1987. Reactivacidn Econdmica en Bolivia:
Analists del D. S. 21660. La Paz: Centro de Estudios para el
Desarrollo Laboral y Agrario.
World Bank. 1988. World Development Report 1988. New York:
Oxford University Press.
. 1989. World Development Report 1989. New York:
Oxford University Press.
BRAZIL
M. Louise Fox
EdwardAmadeo
Jose Marcio Camargo
Although the external shocks of the early 1980s presented major
challenges for all middle-income countries, most entered the 1990s in a
stronger macroeconomic position than they had entered the previous
decade. The process of adjustment (painful for all, but especially for the
population's poorest segments) resulted in leaner public sectors, an
increased decentralization of economic decisionmaking, more
competitive economies, and a more favorable debt structure...
This was not the case in Brazil. By following a less orthodox
macroecononic program Brazil experienced signfficantly less pain than
most of its Latin American neighbors during the decade. Its economic
growth rate was one of-the highest in the hemisphere during the period.
However, Brazil ended the 1980s in a significantly worse position then it
had begun- because of its failure to adjust macroeconomic balances.
Inflation remains high, capital outflow continues, growth comes in spurts
and is unsustainable, and its debt structure (especially internal debt) is
much worse.
In many countries, the adjustment effort's success or failure has
depended on the labor markets flexibility and its ability to adjustL This
has been especially important in effecting the terms of trade changes
required. Was the labor market (or one segment, the labor unions)
responsible for the failure of Brazilian macroeconomic policy? This
chapter argues that the answer to this question is no. While the
democratic process and the role of labor unions in that process
complicated the formulation of economic policy during the period,
Brazilian labor markets were very flexible, and. indeed the extemal
143
144 M. Louise Foi EdwardAmadeo, andJose Marcio Camargo
adjustment required was achieved relatively painlessly. What hurt Brazil
was the failure of the political process to agree on the size of the public
sector and who would pay for the public consumption that segments of
the population sought.
Origins and Nature of Brazil's Adjustment Problem
In the two decades before the advent of the debt crisis in the early
1980s, Brazilians had become accustomed both to high rates of economic
growth and significant improvements in living standards. In the 1960s,
growth averaged 3 percent per capita annually, and between 1970 and
1979 grew at an astonishing 6 percent per capita per annum. During the
latter period, the incidence of poverty fell roughly 50 percent, and the
severity of poverty (the poverty gap) fell by 25 percent (Fox 1990).
These decades of high growth were also ones of structural
transformation. Brazil had started an industrialization and import
substitution program in the early 1950s that focused primarily on
consumption goods. While imports of consumer goods were reduced
drastically, foreign exchange earnings continued to depend on the coffee
crop- After 1968, Brazil sought to break its dependence on coffee through
a more open trade regime combined with a manufactured export
development program financed by large inflows of foreign capitaL The
result was a substantial increase in the size and diversity of Brazil's
international trade, such that manufactured exports grew from 8 percent
of total export receipts in 1965 to 30 percent in 1975, while coffee's
contribution declined from 45 percent to 10 percent during the same
period. Brazil ran a continuous trade suplus from 1968 to 1973.
After the first oil shock, payments for oil imports climbed sharply, up
from 11.0 percent of imports during 1967-72, to 22A percent in 1974,
and 44.4 percent by 1980. Inistead of reducing absorption, the governent
decided to try to grow its way out of this crisis. It opted to borrow abroad
at relatively low real interest rates and continue its investment push,
primarily in the state enterprise sector, designed to substitute domestic
goods for imported capital and intermediate industrial goods. Although
Brazil was able to continue to increase both exports and domestic
consumption, it was sowing the seeds for the adjustment problems of the
1990s. The heavy borrowing, combined with increases in world interest
Brazil 145.
rates, meant that interest payments rose steadily as a share of export
earnings, from 8.2 percent in 1974 to 312 percent in 1980.
The decision not to reduce absorption meant that the second oil shock
and the subsequent rapid rise in world interest rates in the early 1980s
caught the Brazflian economy in a very unstable and vulnerable position.
Even before the second oil shock, Brazil's debt had grown to enormous
proportions, and new lending was increasingly needed just to cover
interest obligations. Thus, the Brazilian growth machine would have
faced serious adjustment problems in the 1980s in any evenL
Although many people have criticized the policies prevalent during
the boom years of the 1970s for the simultaneous increase in inequality,
most of the population experienced significant increases in their standard
of living.1 These effects were not uniform, as poverty was reduced by
roughly 66 percent in urban areas compared with 50 percent in rural
areas, and roughly 70 percent in the southeast compared with just under
50 percent in the northeast Improvements in social indicators were also
more dramatic in the southeast Nevertheless, Brazil's record on poverty
alleviation in the 1970s, even in the least affected areas, is the envy of
many countries, and despite having one of the most unequal income
distributions in the world, the economic growth of the 1970s was clearly
acoompanied by significant social mobility (Morley 1982; Pastore and
others 1983).
Poverty reduction came primarily through expanded employment in
the urban formal sector, where average wages were close to three times
the wages in the rest of the economv by the end of the decade- The
growth in formal sector employment reflected the government's
continuing commitment to an industrialization strategy throughout the
1970s, and was heavily concentrated in the already more developed
southeast Much of this employment was in enterprises that depended on
either government subsidies, government protection from domestic or
1. Brazil's ability to decrease povert at the same time that income inequalitY was
rising does not imply that the criticisms of the unequal nature of Brazil's growth strategy
are without merit On the conrary, studies have shown that Brazil's progress in poverty
reduction would have been even greater if it had been able to have more distnbutionally
neutral growth (For 1990; World Bank 1990).
146 M.LouiseFox,EdwardAmadee,andJoseMardoCamargo
international competition, or government capital (in the form of equity
shares).
The effects of continuing economic improvement and of the expanded
opportunities available to the working class, especially the urban working
class, changed working class expectations. These changed expectations
were given voice as the dictatorship lightened its repression of labor
unions in the late 1970s and a politically active labor movement
reemerged. The expectations formed during the 1970s about economic
growth and the state's role played an important role in the political
economy of adjustment policies during the 1980s.
The debt cisis hit Brazil hard. Domestic absorption had to be cut by 4
percent of GDP merely to adjust to the cutoff in foreign capital inflows.
At the same time interest rates were rising in real terms, requiring even
more belt tightening. Nevertheless, despite the severity of the crisis, most
observers believed that Brazil, with its diverse economy and relatively
rich resource base, would eventually return to a growth path, less
dependent on external savings. In 1983, the World Bank estimated that
the required adjustment in Brazi would require a savings rate of about 20
to 25 percent of GDP over the next five years (a marginal rate of about 30
percent, assuming a return to growth after a brief period of austerity).
Compared with the marginal savings rates of 50 to 60 percent required
from Chile during the same period, Brazil was viewed asc the country that
could be a model for the region in terms of adjustment, growth, and
external trnsfer, with minimal tradeoffs between the three objectives.
What these projections could not highlight, which proved critical in
Brazil's failure to adjust, was that in Brazil all the adjustment had to take
place in the public sector (the actual owner of the debt), to avoid a large
public-private transfer problem and significant crowding out- This
adjustment in the public sector had to take place at the same time as the
country was opening up the political process to groups that had been
disenfranchised for 20 years. Politically, the task was to cut the size of
the pie by about 25 percent just as the group standing in line. to get a
piece was increasing dramatically and its power was growing. Brazil's
ultimate failure to stabilize and adjust effectively in the 1980s was in part
a result of the emerging democracy's failure to reach a political
consensus on who would pay the bil for the excesses of the 1970s.
Brazll 147
The Period of Adjustment: The 1980s
Brazilian macroeconornic policy in the 1980s and its outcomes can be
divided into three periods: (1) recession, 1981-83; (2) recovery, 1984-
85; and (3) boom-bust, 1986-89. Tables 4.1 and 4.2 summarize the
quantitative record.
Table 4.1 Macroeconomic Indicators, 1980-87
hnd&cusor 1980 1981 1982 7983 7984 1985 1986 1987 1988
GDPfactorcosx(1980=100) 1.00 096 L00 0.93 0-99 1.07 1.12 1.17 1.18
Agriculture 1.00 1.08 1.08 L07 LlI 1.22 1-12 1.28 1-30
Indusuy 1.00 091 091 0.86 0.91 1.00 1.11 lI2 1.09
Services 1.00 0.98 LOO 099 1.03 1.40 1.19 123 1.26
GDP markct prices (1980 100) 1.00 0.96 0.96 0a3 0o98 1.06 1.14 1.18 LIS
Fiscal policy indicatots
(patentage of GDP)
Revenuc 23.7 235 23.8 232 20.7 21.2 23.4 24A ZLP9
Interest 19 22 3.3 4.2 6.2 l.9 10.6 9.7 12,5
Gavrrnment saving 1.1 1.1 -0.4 -1.4 -2.8 -&O -7.0 -6.6 -126
Debt u.. 15.5 19.8 28A 34.3 36.1 22.7 40O) ua.
Inflation (annual rae, perccut) 90 108 106 141 215 235 144 210 673
Real exchange rate 1.0 0.9 0.9 12 1.2 1.3 1-1 1.0 0s
Real interest rate (working capal) -13.4 25.7 24.6 13A 36.4 32.1 6.4 30.7 na.
Implicit rate of return,
goveranncat debt, ovcenight
market (percent) 1.3 185 265 13.6 17.9 15.9 5.8 7A n.a.
n.a =not available
* GDP deflator, annual rate of change.
a Exchange rate deflated by the cost of living (Brazil) times U.S. WPI (increase =
depreciation).
Source: National Accounts: Conjwntura Economica (June 1990); Government debt World
Bank (1988, p. 313).
148 M.LouiseFoxEdwardAnmadeo, andJoseMarcioCamnargo
Table 4.2 Changes in Saving and Investment, 1980-88
(millions of 1980 cruzeeros)
Recossion Recovery Boom Slowdown Toarl period
Category 198083 1983-86 7985-86 1986-8 1980-8
Changes in
consumption -494 7n1 960 48 1,215
Government -28 152 234 347 704
Private -466 549 726 -298 511
Exports 290 432 -195 574 1,101
Imports -447 -28 265 -50 -261
Foreign saving -738 -460 460 -624 -1,361
Domestic saving (GDP
-consumption) -377 879 28 466 996
Ilvestment -1,115 418 488 -156 -365
GDP -871 1,580 988 514 2,111
Source: Fox and Morley (1_9O).
In light of the collapse of foreign exchange reserves and the increase
in the trade deficit, only a "hard-option" financial adjustment seemed
possible by 1981. During this period, Brazfl used tight money policies,
some fiscal restrai, quantitative import controls, and an active exchange
rate policy to lower demand and squeeze out the resources that were
needed for the external transfer.2 GDP fell by 5 percent between 1980
and 1983. The burden of adjustment fell primarily on the private sector,
2. Throughout the 1980s, imports were tightly controlled by a system of import
licensing and quantitative restraints. Thus, the exchange rate was used primarily as an
export promotion tool, and Brazil was able to generate the trade surpluses required with
smaUl changes in the real exchange rate. In this way, the Brazian experience contrasts
with that of other, more open, economies Large real depreciationsldevaluations were not
required in response to the external shocls of the 1980s, and exchange rate policy played
a relatively minor role in stabiization/adjustment programs.
Brazil 149
as government savings began to turn negative with rising interest costs.
In an attempt to control inflation and limit the burden of adjustment on
the poor, the government also used a wage control policy of "cascading"
adjustment in the formal sector that permitted more than 100 percent
indexation of wages at lower wage levels, and less than 100 percent
indexation at higher wage levels. The result of these policies was Brazil's
deepest recession in 15 years, a 40 percent fall in investment and, by
1983, an annual tnsder abroad of 4 percent of GDP.3
Brazil was able to achieve external balance and earn the foreign
exchange necessary to service its debt quite rapidly. Exports increased
from USS20 billion in 1982 to US$27 billion in 1984, reacting to a slow
domestic market and better prices abroa& Imports declined from USS19
billion to US$14 billion under the tight regime of quantitative
restrictions. The balance of payments shifted from a deficit of US$95
billion in 1982 to a small surplus of US$0.4 billion in 1984. Brazil
achieved this rapid success largely because of the export promotion
policy of the 1970s.
During the recovery period (1984-85), Brazl began to ease interest
rates. At the same time, government expenditures on goods and services,
wages, and investment returned to the levels of the 1970s, when they had
been financed by extemal debt Unfortunately, this source of financing
was no longer available, and interest payments continued to grow To
finance the expenditures and the consequent deficit, the government was
forced to sell more government bonds-eventually forcing interest rates
back up-and to print money. These policies led to accelerating inflation
and no substantial improvement in investment.
As the recession had left many private sector firms with excess
capacity, the previous decline in investment was not yet much of a
constraint on growth, and the Brazilian economy responded well to the
fiscal stimulus. The government abandoned the policy of cascading wage
adjustments, and a policy of exchange depreciation maintained the trade
surplus even as internal demand began to expand, providing Brazil with
the foreign exchange to continue debt service payments. Once again,
3. One would expect that the cascading policy would lead to the compression of
salary difirentials In practice, it only had this effect in the public sector, as major private
sector and joint public-private companies simnply corrected for this policy by paying wage
supplements of various kinds to their higher level staff-
150 M. LouiseFox, EdwardAmadeo, andJose Mareo Camargo
earning foreign exchange was not a problem for Brazil. The trick was to
get the local currency equivalent of the trade surplus into the
governtment's hands so that the debt could be serviced; a feat that was
proving increasingly difficult. Nonetheless, Brazil achieved a marginal
savings rate well above the level required for debt service during 1984-
85, investment began to recover, and, except for the troubling inflation,
Brazil seemed to be emerging from the debt crisis on a "Baker" path.
By the end of 1985 (the start of the third period), the transfer problem
was becoming acute. Inflation was accelerating, the velocity of money
was increasing, and financing the govemment deficit by printing money
was becoming more and more difficult. Only two possible optbins were
available for solving these problems: reduced government consumption
or increased revenues. Both required political consensus. Unfortunately,
the coalition government Brazil's first democratically elected Congress
in 20 years led by a politically weak president, was not in any mood to
forge this belt-tightening consensus. The opposition, without control of
the public purse for so long, mostly sought to benefit its constituencies.
Neither was the establishinent, represented politically by the president, in
any mood to bear the burden of adjustment. All elected officials feared
recession and unemployment, the right, because it would strengthen the
leftist labor unions, and the left because their voters would suffer and
blame them.
This polifical stalemate dominated Brazil's macroeconomic policies
throughout the second half of the 1980s. The outcome was a period of
growth (1986), followed by recession (1987-88), followed by a further
growth spurt (1989), with inflation held in check only through
increasingly unsuccessful wage and price control programs inaugurated
roughly once every 18 months, and with private investment crowded out
The first and most famous of Brazil's stabilization plans was the
Cruzado Plan, initiated in February 1986. Its key elements included (a)
real wage increases to pacify organized labor, (b) monetary reform and a
price freeze; (c) a govemment-imposed deindexation of the economy,
including financial instruments and the exchange rate; and (d) an
exchange rate freeze (which implied an appreciation) and a more open
import policy to ease shortages. All these measures increased real
purchasing power in the short run, increasimg aggregate demand. Yet
despite the breathing room that the temporarily lower inflation brought in
Brazil 151
terms of interest savings and seigniorage gains, and the increased tax
collections stemming from the reverse Tanzi effect, the govemment
failed to cut govermment spending.4 On the contrary, the failure to
mcrease public sector prices prior to the freeze and the real wage
increases granted to government workers as part of the package
aggravated fiscal pressures. The disequilibrium in the balance of supply
and demand became evident by July 1986. Shortages developed, inflation
retumed, and the plan collapsed. In addition, the import buying spree that
was stimulated by the appreciated exchange rate (facilitated by
government import policy) had used up reserves. A debt moratorium was
finally imposed in 1987.
Brazil undertook two more shock stabilization programs in the 1980s.
While both appear to have averted hyperinflation-a constant threat to
Brazil as inflation begins to accelerate with each recovery in private
aggregate demand-neither permanently reversed the negative trend in
govemment savings. At the same time, a new debt agreement with
foreign commercial banks in 1988 led to renewed savings outflows. With
the debt service outflow and the govermnent financing needs eating up
savings, private investment remained stagnant after a short period of
increase during the Cruzado Plan.
Brazil never stabilized financially during the 1980s, constantly opting
for inflation over unemployment and consumption over savings. Table
4.2 shows the macroeconomic results of Brazil's economic.policy during
the second half of the 1980s. On the positive side, Brazil increased
domestic income by about 16 percent over the level in 1980, and met the
savings targets required to continue servicing its foreign debt, moving
quicldy from a trade deficit in 1980 to a surplus in 1982. That position
was maintained throughout the decade except during the Cruzado Plan
boom of 1986.
Both private and public consumption also increased, and although
public consumption increased almost 50 percent faster than private
consumption, the overall increase did help to protect living standards.
However, most of the savings generated during the 1980s were applied to
4. As inflation increases, real tax revenue tends to decline even as tax rates remain
unchanged because money Ioses value during the collection process. This is known as the
Tanzi effecL When inflation declines and real tax revenues increase, the process is called
the reverse Tauzi effect
152 M.LouiseFoxEdwardAmadeo, andJoseMarcioCamargo
debt service; consequently, the level of investment fell from 23 percent of
GDP in 1980 to 16 percent in 1984. The private sector's increasimg
unwillingness to finance government consumption (including debt
service payments) led to an inflation level of above 50 percent per month
by the end of 1989. This failure to invest can be expected to compromise
Brazil's growth prospects for the 1990s.
The Development of the Labor Market in the 1980s
Segmentation in tlhe Brazilian Labor Market
The distribution of employment by sector in Brazil reflects the
industrialization of the postwar period. Most Brazilians do not work in
agriculture, but in industry, commerce, or services. Out of a total labor
force of about 50 million, about 5 million people work directdy for the
government in public administration positions, and about 2 to 3 million
more work in state-owned enterprises.
The Brazilian labor market can be divided into three segments: the
formal sector, the informal sector, and the self-employed (table 423).
Labor law obliges employers to sign identification cards for each worker,
which makes the worker eligible for a number of benefits from the state-
Table 4.3 Composition of the Labor Force by Sector, 1986
(percent)
Formal Informal
Signed Nonsigned Self- Witho. Either
Sctor contraA contract enplayed remnwration employer
Agriculture 6.9 332 33.2 25.2 3.5
Industy 72.4 15.9 6.6 13 3.8
Constuction 39.3 Z79 29.6 017 2.5
Commerce 42.5 16.3 30.9 3.5 6.8
Services 21.9 40.7 33.2 15 2.7
Transport and
ComiMUication 60.2 1139 25.1 0.7 2.1
Public administration 52.5 47.0 0.3 0 0.2
Total 38.2 Z79 22.9 7.6 3.4
Source: Saboia (1989).
Brazil 153
These include extended health and social security benefits,
unemployment insurance, protection from minimum wage laws, paid
vacations, a maximnum normal working week of 44 hours with time and a
half for overtime, and so on. These benefits cost the employer 50 to 70
percent of the direct wage. Employers also pay a 20 percent payroll tax
on workers' earnings (matched by an 8 percent payroll tax paid by the
workers themselves).
Small business often cannot afford these additional costs, and do not
sign their worker' cards. Together with the self-employed, these workers
constitute the informal sector The extent to which workers fall into these
categories varies widely by industry. Only a tiny percentage of
agriculturaI workers and a slightly larger percentage of service industry
workers are in the formal sector. This contrasts sharply with industrial
workers, and above all public sector workers, who are much more likely
to be in the formal sector.
Regional differences are also important in the distnbution of formal
sector workers (table 4.4). Sio Paulo is the most developed state in the
Table 4.4 Distribution of Formal Sector Employment by State and
Sector, Selected-States, 1985
Told
Sector @pem) nowgrkcuntu
formal sctor
Public unpkoy=
Sine Indwuiy SerAw Commerc admradan (ih--zr-
Ceara (NE) 23.2 24.4 12.3 38.4 477
Paraiba.(NE) 20.8 18.2 7.9 52.7 235
Pemambuco (NE) 32.1 273 11.5 27.8 738
Bahia (NB 21.8 31A 14.0 31.1 834
Rio de Janeiro (SE) 24.9 39.0 14.2 21.7 2,711
Sio Paulo (SE) 41.0 31A 12.3 13.6 6,780
Total Brazil 32.1 30.9 13.1 22.3 20,172
Percentage offormal sector employment in sector
Sio Paulo 43.1 34.2 31.7 20.7
6 southeast states 81.1 76.0 75A4 57.0
Rest of Brazil 18.9 24.0 24.6 43.0
Source: Sabola (1989).
154 M. LouiseFox, EdwardAmadeo, andJoseMarcio Camargo
country, accounting for roughly 60 percent of GDP. Of its workers, 54.8
percent were in the formal sector in 1986, while that figure was only 21.8
percent in the poorer northeast. Similarly, 82.6 percent of industrial
workers were in the formal sector in Sio Paulo, compared to only 41.7
percent in the northeast.
Most of Brazil's poor, urban and ruraL live in households whose head
is not employed in the formal sector (table 4.5). Most heads of poor
families are self-employed or sharecroppers, earning income in the
agricultural or tertiary sectors, although in urban areas, heads of poor
households are also found in significant numbers in manufacturing and
construction. However, in the large cities of the southeast, poor southeast,
poor households do depend on formal sector earnings from the head
Average formal sector earnings are roughly three times those in the
informal sector (including agriculture) (table 4.6).
Table 4-5 Occupational CharacteTistics of Heads of Poor Households,
Selected Areas, 1985
(percentage ofpoor population in household)
Urban Urban Rural Rural
Occuation of head Brazil northeast soutet northeast souheat
Technical/administrative 4.4 52 6.8 2.7 3.1
Agriculture and mining 39.2 27.6 13.6 85.9 84.9
Manufacturing and consruction 10.3 25.8 33.2 55 5.1
Commerce and related actiities 8.6 12.1 6.7 1.7. 8
Transport and communications 4.6. 43 5.7 9 .6
Services 22.4 69 128 .7 2.6
Others 133 18.2 19.5 2.6 3.0
Formal sector employment 17.7 31.8 50.1 55 109
Share of the poor 100.0 202 17.2 33. 10.2
Source Fox (1990).
Brazil. 155
Table 4.6 Average Monthly Earnings of Heads of Households, 1987
(CZ$)
- onheas Southest South Nonheasti
Employment status (CZ$) (CzS) (CzS) southeast
Formal sector 23,225 33,812 27,753 0.69
Informal sector 10,511 22,000 20,664 0.47
Self-employed 12,180 30,383 25,263 0.40
Employer 58,121 81,647 82,150 0.71
lVote: Employees in the formal and informal sector do not include agricultural laborers or
unpaid workers.
Source: Fox (1990).
During the 1980s, rural and urban labor markets became increasingly
integrated. Thus, for example, 25 percent of the heads of poor households
in the urban southeast work in prinmary sector activities, and 15 percent of
heads of poor households in the rural southeast do not work -n
agriculture. The agricultural labor force has also become increasing
proletarianized during the decade: by 1987 over 50 percent of those
earing income in agriculture were employees (even in the northeast, the
comparable figure is 48 percent).
Roughly one-fifth of agricultural employees nationwide have signed
labor cards (formal sector employment), but this ratio also varies
significantly by region, with the level of formalization in the south twice
that in the northeast. While most earners in poor households are at the
bottom of the earnings distribution, not all low eamers belong to poor
households. In 1985, roughly 40 percent of those earning the minimum
wage in the formal sector were secondary earners in households with per
capita incomes in the top 40 percent of the distribution (Alemeida Reis
1989).
The Rise of the Unions
The reemergence of a populist, leftist labor movement in the 1980s
was one of the period's great ironies. One of the main goals of the
authoritarian military regime that came to power after the coup of 1964
was to subjugate the labor movement under federal control. During the
156 M. Louise Fox EdwardAmadec, andJoseMarcio Canargo
repression of 1964/65, a large number of labor leaders were jailed or
exiled, and government control over unions increased. Unions had to be
organized by occupation and geographical location, and needed the
formal approval of the Ministry of Labor. Approved unions were given
sole rights to represent given groups of workers, and national unions
were initially outlawed. Unions were tied to the state financially through
the provision of compulsory union dues mandated by the state, and
restrictive strike laws (which included the identification of "key" sectors
where strikes were entirely forbidden) blocked the emergence of
independent power centers. While unions did have the right to engage in
collective bargaining, these bargains were always subject to a national
wage policy, effectively rendering the collective bargaining process at the
local level impotent5
As long as the government was able to maintain authoritarian control,
the corporatist system worked reasonably well from a macroeconomic
perspective. Costly strikes were avoided, real wages rose with economic
growth, and inflation was moderate. However, the system began breaking
down as the political system began to liberalize during the late 1970s.
The very tool that had worked so effectively to control inflation during
the period of maximum repression began to be the undoing of the system
during liberalization, as unions continued to pressure the government to
set federal wage guidelines at a level that would allow workers to make
up for past inflation. As the goverment was both part or full owner of
many of the large employers of unionized workers and at the same time
was beginning to compete for the political support of the unions, it often
acquiesced, having no countervailing organized political pressures for
wage moderation.
At the same time, unions began to flout government wage guidelines
in their negotiations. The existing institutional framework of labor law,
which carried excessive penalties for minor infractions, became useless
during the period of liberalization, as the government could not afford the
political costs of invoking laws that were widely perceived to be harsh
and excessive. Thus, the government simply avoided the existing
5. Wage indexation became a key tool of incomes policy. Between 1965 and 1974, an
"expected" rate of inflation was imposed as the norm for adjusting nominal wages to
inflation. As expected rates invariably lagged real inflation rates, real wages fell, and
some downward pressure was exerted on inflation.
Brazi 157
structure, and costly and -violent strikes began to occur (table 4.7). Labor
leaders mobilized workers directly at the factory fcor, short-circuiting
the formal structure of corporatist state controls on union activity. As
mobilization increased, strong local leaders began to attract support from
th,I most important unions in the manufacturing regions, especially the
southeast. A Workers' Party (the PT) was formed in the early 1980s, and
two centralized national union federations were created, one of which,
the Central Unica dos Trabalhadores or CUT, was linked to the PT.
Consistent data on the growth of unionization and mobilization is sketchy
until the mid to late 1980s, so tracking the growth of this movement is
impossilble, but by 1986, roughly one-third of the nonagricultural labor
.orce in the industrialized southern half of the couutry belonged to a
union (table 4.8).
The unions also adopted a targeting strategy, seeking to bargain at the
enterprise level with the largest firms. Agreements rmade with these firms
were used as a basis for all -agreements, and became national bargains.
Negotiations occurred year round as each occupational group bargained
in a different month. Naturaly, the bargains made by the best organized
unions became the targets for the rest, while the support offered by the
central labor federations increased every union's ability to enforce its
Table 4.7 Number of Strikes and Total Number of Workers on Strike,
1985-89
Total number of
workers on strike
Year Number of strikes (millions)
1985 843 6.6
1986 1,493 7.1
1987 2,275 83
1988 1,914 7.1
1989 4,167 10.0
Source. Ministry of Labor, Brazil.
158 M. Louise Fox EdwardAmadeoa ndJose Marcio Camargo
Table 4.8 Membership in Workers' Associations, 1986
(percentage of nonagricultural workers)
Sector Membership
Manufacturing 29.10
Construction 12.30
Other industries 4336
Commerce 14.79
Services 5.61
Auxiliary-services 36.30
Transport and communication 43.37
Social services 25.85
Public administration 20.89
Others 4856
Total 2136
Notc Employers are included in the denominator, thus the level of unionism is
understated.
Sourwce PNAD survey data (1986).
demands. As bargains were usuaily made once a year, the high and very
uncertain inflation rate had a huge impact on bargaining. It ensured that
unions would aim at wage U-creases not only large enough to compensate
members for any nominal losses since the last agreement, but large
enough to ensure that members woild not suffer regardless of the
prevailing inflation rate in the coming year. The result was built in
inflationary pressure.
A bargaining structure in which bargains were made locally but
enforced nationally had poor macroeconomic results. The national
enforcement ensured that labor markets became very rigid, unable to
differentiate between successful and failing. firms, occupations with
strong demand and weak demand, and so on. Conversely, the local
negotiation meant that national macroeconomic concerns were not part of
the equation, in contrast to more centralized systems such as that of
Sweden.
Brazil 159
In sum, unionization became a powerful force affecting economic
outcomes during the 1980s. Equally important perhaps, the unions had
developed into a sufficiently potent force to ensure that adjustment
measures primarily affected those in the informal sector.
The Role of the Labor Market in the Adjustment Process
Brazil's macroeconomic policies of the 1980s produced modest
growth in per capita income, extemal balance, and high inflation. Brazil's
labor market structure, policies, or institutions did not prevent the
economy from earning the foreign exchange necessary to service its
external debt. However, achieving internal stability was another matter.
Throughout the 1980s, Brazil stmggled in various ways to achieve this
goal.
In general, the Brazilian labor market exhibited tremendous flexibility
throughout the 1980s (tables 4.9 and 4.10). During recessions, the decline
in formal sector output crowded workeis into the informal sector,
lowering average incomes in this sector. Daring expansions, wages and
employment in the formal sector rose. While unemployment was a major
problem during the early part of the decade, both because of the recession
and the drought in the northeast, which brought many heads of farm
families to the cities in search of work, it declined during the decade as
the informal sector was able to continue to absorb new enrants.
However, different parts of the labor market gained or lost during
different policy regimes.
During the recessionary period (1981-83), the formal sector clearly
gained at the expense of the informal sector, while the big loser was
agriculture, where roughly 29 percent of the labor force was employed.6.
Two government policies seem to have facilitated this outcome: (a) the
guarantee to formal sector workers in lower earnings categories that
wages would be overindexed every six months; and (b) the generous
government employment policy. Private sector employers did shed some
workers in response to falling demand, but some Iabor stockpiling also
occurred, as output fell faster. In contrast, government employment
6. Although agrculural output increased overall eaings in agrculture must have
been affected by the drought, which lasted through the 1982 harvest.
160 M. Louise Fox, EdwardAmadeo, andJoseMarco Camargo
increased during the period, so total formal sector employment did not
decline.
Table 4.9 Indices of Labor Market Outcomes, 198048
Category 1980 1981 1982 1983 1984 1985 1986 1987 1988
Employmenm growth
Total l.0 1.01 1.06 1.4 112 1.18 122 127 1.30
Agriculture 1.00 0.96 1.02 OSS 1.08 1.10 1.02 1.01 1.02
Formal 1.00 lJ01 1.03 10 1.03 1.11 1.21 1.25 1.31
Informal 1.00 1.S 122 127 138 1.49 1.54 1.69 1.67
Private foumal sector + 1.00 09S 0.95 0.89 092 098 1.02 nA. nua
Publicscctor+ 1.00 106 1.12 1.16 125 132 1.43 n.. na.
Open unemployment (PMD) (X)
Sio Paulo 720 7.20 550 6.80 680 5.D0 3.40 3.80 3.80
Average, 6 cities ni. U.L 6.30 6.70 740 520 3.60 3.70 3.80
Real wages
- Private industry
(Sic Paulo. FIESP) 1.00 1.07 114 1.06 099 1OS 1-17 1.08 0.95
Total formal sector+ I.C 1.01 1D7 0.92 087 0.98 1.05 u.n. nA
Governmentsector+ 1.0 0.97 1.03 0.86 0.78 0.99 1.16 IIn. na.
Minimumnwage 1.00 0.99 1.01 091 085 0.86 0.89 0.73 0.76
Real avcragc incomes"'
Formal 1.00' 0.85. 130 092 1.03 120 1.65 124 0.88
Informal 1.00' 0.85 1.30 092 1.03 1.20 1.I5 124 O08
Agriculture 1.00' 0.84 0.96 0.77 0.78 0.84 1.16 0.3 0.52
Inoome differetidals
Farmal/infcnal 3.08- 3.10 311 325 2.66 2.76 250 272 3.28
Informallagricrlturm 0.63 0.64 0.86 0.75 0.84 0.90 0.90 0.94 1.08
Factor incomes in formal sector
Labor 1.00 1.03 110 0.93 0.88 1.05 123 1.16 1.01
Interest 1.00 1.30 128 128 1.32 1.48 034 1.58 1.58
Profit (real) 1.00 0.89 0.86 088 0.96 0.99 1.13 1.12 115
Profit and interest 1.00 0.93 091 093 1.00 1.04 1.10 1.17 1.20
Labor productivity 1.00 092 09! 09 091 0.90 0.95 OSO 090
n.a. = not available
1979 +RAIS data.
Average earnings not corrected for hours worked, main occupation.
Note: Informal sector includes agriculture and is defined as labor force participants not
contnrbuting to the social security system.
Source. Fox and Morley (1990).
Brazil 161
Table 4.10 Poverty Indicators, 1981-87
Indicaor 1981 1983 1985 1987 1988
Incidence of poverty
by location
Brazil total 26.4 32.1 26.2 242 269
Urban 14.9 21.6 171 14.8 na.
Rural 46.8 542 47.1 463 na.
Poverty gap index 10.1 13.1 9.9 9.5 10.7
Index of GDP per capita 1D 0.9 1.0 1.1 1.1
Index of real bousehold
income
Mean 1.0 0.9 1.1 1.2 1.2
Bottom 10 percent 1.0 0.9 1.1 1.1 LO
Bottom 25 percent LO 0.9 1.1 1.1 1.0
Top 10 percent 10 0.9 1.2 1.3 L4
n.a. = not available
Sowrce Special Tabulation of Household Surveys. For a definition of the poverty line, see
Fox (1990).
The evidence on wages is more ambiguous& The minimum wage fell
by 10 percent from 1980-85, but real wages in the industrial sector
increased. The share of factor income going to labor increased sharply,
while the share of nonfiancial profits declined correspondingly. Within
the govemment sector, the employment increase was accompanied by
significant real wage compression, which caused average wages in the
formal sector as a whole to fall. As overall employment was stagnant in
the formal sector and shrinling in agriculture, the informal sector
absorbed all the natural increase in the size of the labor force during the
recession, and average value added per worker in this sector fell by one-
fourth. Reflecting this surge in employment (as well as the decline in
agricultural incomes), informal sector incomes fell by almost 10 percent
between 1980 and 1983. Somewhat surprisingly, the differential between
the formal and informal sector incomes remained roughly constant,
increasing by only 5 percent for the period. At the upper end of the
income spectrum, profits-specially nonfinancial profits-contracted
sharply, as owners of physical capital were hurt by the combination of
162 M, Louise Fox EdwardAmadeo, andJose Marcio Camargo
high interest rates, workers' ability to protect their wages, and sluggish
demand.
In short, the government's tight money policy, combined with a wage
policy that maintained real wages in the formal sector, in effect protected
the middle of the income distrbution against both ends. The protection of
the middle clearly also benefited the urban informal sector by helping to
support demand for its services. Nonetheless, the fall in incomes in the
agricultural sector, where the majority of the poor earn their incomes,
was combined with the crowding of new labor market entrants into the
informal sector, where average earnings are one-third of those in the
formal sector. This pushed a significant portion of the population back
into poverty, especially in the urban areas in the south and southeast,
where most of the urban population is located.
During the recovery period (1984-85), formal public sector workers,
holders of capital, and informal (including agricultural) workers
improved their positions at the expense of private formal sector workers.
The income differential between formal and informal sector workers feU.
Public employment continued to swell, while private sector employment
kept pace with output growth. In addition, in 1984, when inflation took a
sharp jump upward, formal sector workers were left behind. These
income losses led workers to demand (and receive in some sectors) a
halving of the indexation period in 1985. Although private sector workers
did not make real gains, government workers began to recover wages lost
during the previous period- The increase in informal sector incomes
combined with the increase in formal sector employment, which
automatically raises average wages in the economy as the formal sector is
the high wage sector, brought a significant decrease in urban poverty and
in poverty overall. In this period, growth did trickle down to the poor,
reversing some of the adverse effects of the previous period-
Tle Cruzado Plan resulted in short-run gains for all groups, but it
proved unsustainable. Under the plan, interest rates and prices declined
while profits and consumption increased. This generated increases in real
income across the board. In addition, employment rose, especially in the
higher earning formal sector. Labor markets tightened, the earnings
differential between the formal and informal sectors narrowed further,
and the increased demand relative to the supply of labor sharply
increased real earnings in the informal sector. Agricultural incomes also
Brazil 163
jumped as employment in the sector dropped in response to the urban
boom. The poor immediately benefited from the real income gains as
poverty dropped below pre-crisis levels.
After the boom, inflation retumed in 1987 as the government tried to
force the private sector to finance the fiscal expansion. Prices rose and
real incomes-both labor earnings and profits-fell. Higher inflation
clearly hurt labor incomes, especially in the less organized parts of the
formal sector where average earings fell 35 percent. Informal sector
earnings also dropped (almost to 1984 levels), and the incidence of
poverty increased again, eroding the gains of the previous period.
From a poverty perspective, the boom and bust of the Cruzado Plan
ultmately hurt the poor, as the slowdown that followed the plan lasted
through 1988. In addition, the Cruzado Plan's excesses also exacerbated
the stabilization and adjustment problem by adding to the debt burden. If
Brazil had actually stabilized in 1986 (and this was not an absurd
possibility), the poor might have recovered what they had lost relative to
the middle class during the recession by the closing years of the decade.
The longer stabilization and adjustment was postponed, the worse off the
poor became.
During the 1980s, the labor movement had gained a critical voice in
policies determining the distribution of burdens resulting from
adjustment. The case of Sio Paulo State is instructive. During the
recession, industrial workers were the only group in Sio Paulo able to
avoid real wage reductions while other groups suffered severe cuts and
the incidence of poverty increased by 50 percent. In contrast,
nonunionized workers did much better, relatively speaking, during the
recovery and especially under the Crnzado Plan, when they benefited
from the price freeze. Unionized workers were clearly much better able
to avoid the pain of adjustment, and they did so through the usual means
open to unions: militancy and activism- However, organized labor was
.not the obstacle to successful stabilization and successful adjustment in
1985; it was the failure of the system as a whole to reach agreement on
the distnbution of consumption. No doubt the unions' emergence as a
powerful actor onto the political scene at a time when increasing
democratization made the political sistem much more open to influence
complicated the political economy of adjustment. Unfortunately, at the
same time, Brazil realized that an agreement w.th the labor movement
164 M. Louise Fox EdwardAmadeo, and Jose Mrrcio Carnargo
was critical, the political institutions, and perhaps the political will, that
would make such an agreement possible were lacking.
Consequtences of Adjustment for the Poor
During 1980-87, Brazil's macroeconomic policies hurt the poor less
than they might given BraziL's lackluster growth performance. Despite
data difficulties and ambiguities, we can conclude with some confidence
that if the incidence and intensity of poverty did worsen, it did not
worsen very much. The main reason for this appears to be the protection
of formal sector wage incomes during the recession, and the expansionist
fiscal policies in the postrecession period. In the 1980s, as in the 1970s,
output growth was strongly related to poverty reduction. The major factor
keeping the econorny afloat was government consumption, a significant
portion of which was public employment. This fiscal stimulus helped to
maintain employment and stimulated some growth in real output. The
stimulus appears to have trickled down to the poor most rapidly in 1984-
85, when output in the private formal sector also expanded rapidly. But
during 1986-88, the poor were not as fortunate, as negative distributional
shifts overwhelmed overall income growth, reducing the average incomes
of the poorest 10 percent of the population.
By 1988, Brazfl's poor were already beginring to pay the costs of the
failure to exercise macroeconomic restraint during 1985-87, and these
policies will likely bring high future costs for the poor as well. First, the
public sector deficit absorbed a large share of private sector savings,
crowding out the private sector investment needed for accelerated growth
and labor productivity improvements in the 1990s. Second, the high
interest rates the govemment paid on its internal debt constituted a
significant and regressive income transfer, as the share of national
income going to debt service rose to 10 percent of GDP by the middle of
the decade. The household survey data does a very poor job of recording
capital income, and thus the effect of this transfer on income distribution
is not well documented. Nonetheless, the size of this transfer to holders
of government bonds may have been a factor in the deterioration of
Brazil's already unequal income distribution that occurred during the late
1980s.
To analyze further the effects of adjustment on the poor, they can be
disaggregated into three groups: the nrual poor, who are found primarily
Brazil- 165
in the northeast and who constitute 50 percent of the overall poor
population; the urban poor in the northeast, most of whom continue to
depend en agriculture and/or the informal sector for their inoDmes; and
the urban poor in the south and southeast, who are much more tightly
linked to the formal sector. Rural poor households are overwhelmingly
agricultural in occupation, and very few have the signed employment
cards that provide access to the formal sector (although the rate in the
southeast is twice that in the northeast). Urban poor households are much
more likely to be headed by persons with formal sector jobs: in the
southeast. 50 percent of the poor have a main source of income in the
formal sector.
The incidence of poverty changed for all three groups during the
1980s. First, for the first time in two decades, the percentage of the poor
living in urban areas increased bubstantially: by 1985, -!ss than 40
percent of the poor lived in households where the main source of income
was agriculture. Second, the percentage of the poor living in the northeast
also increased, even though the actual incidence of poverty increased
faster in the south and southeast (from a lower base). Apparently the
northeast was simply not as efficient in using its (generaUy higher)
growth during the early 1980s to alleviate poverty.
Poverty in the rural northeast is much more strongly affected by
supply-side factors such as investment than by adjustment. Agricultural
policies that have steadily favored investment and the building of
infrastructure-irrigation and so on-have helped to improve the
productivity of large farms, as has the switch from crops to livestock.
Increased investment and technical change have also encouraged the
proletarianization of the rural labor force, as wage labor has in some
regions, such as the sugar plantations of the Zona de Maua, completely
replaced the historically dominant sharecropping relationship. This
process has also led to a rise in the use of permanent labor contracts,
membership in trade unions, and the incidence of signed employment
cards. Although no data is available on the economic effects of these
changes on the poor, they are likely to have tightened the link between
the economic condition of the country as a whole and the lot of the rural
poor, who are now more closely, linked to the product markets in Brazl,
and even abroadL
166 M. Louise Fo; EdwardAmadeo, andJoseMarcdo Camargo
In contrast, poverty in urban areas is strongly and negatively related to
economic growth. This is especially so of the private formal sector,
which produced 70 percent of GDP in 1980 and employed 47 percent of
the nonagricultural work force. This relationship is especially strong in
the south and southeast. Thus, while the incidence of poverty increased
by 50 percent during the 1981-83 recession in the south and southeast, it
increased by only 30 percent in the northeast. Similarly, poverty declined
much more slowly in the northeast during the boom years of the Cmzado
Plan.
Despite the recession and persistent crises, Brazil continued to realize
improvements in social indicators. In part, this was because the heavy
investments in social infrastructure of the previous decade were
delivering their payoff in the 1980s, but also because adjustment did not
stop the expansion of social infructure in Brazil- Overall mortality
rates continued to decline, and the infant mortality rate fell 40 percent
between 1980 and 1986. By 1987, over 90 percent of urban households
had access to potable water and to electricity. However, the
disadvantaged northeast remained far behind, especially in the rural
areas.
Conclusion
In the end, Brazil never permanently adjusted internal demand to the
exteral shocks of the 1970s and 1980s. In the 1970s, it borrowed its way
through the oil crisis, waiting for oil prices to fall while continuing to
invest, increasingly in the public sector. In the 1980s, oil prices did fall,
but interest costs increased fast and external sources of finance dried up.
Initially, Brazil responded by reducing demand. From 1985 onward, the
government switched policies and sought instead to borrow from
domestic sources. To do so, it had to pay very high-real interest rates, and
interest payments became a growing proportion of overall government
expenditures.
In the fieId of trade and the external balance, Brazil's quantitative
controls over imports and substantial export promotion incentives
allowed the government to switch expenditure as needed. Thus, unlike
most middle-income countries, earning foreign exchange was not a
problem. Brazil's success in this area is due primarily to its diversified
industrial structure and large internal market, which allowed Brazil, more
Brazil 167
than most middle-income countries, to achieve external balance through
expenditure-switching policies instead of expenditure-reducing ones. A
large real wage decline was not necessary for Brazil to achieve external
balance.
However, some decline in consumption was needed to achieve
internal balance. The evlidcnce from the labor and capital markets is that
none of the actors with any influence over government policy were
prepared to sacrifice their own consumption to achieve this goal, nor
under the emerging democracy was capital able to force labor to sacrifice
as it did in the 1960s. Capital demanded high real interest rates to stay in
Brazil while profits recovered from the recession; workers in the formal
sector managed to defend real wage levels more or less continuously until
1988; and wages in the informal sector did fall during the recession of the
early 1980s, but then rebounded. Only in agriculture did incomes fall
overall, but the difficulty of measunrng incomes in this sector makes this
comparison problematic.
While Brazil's labor markets have shown significant flexibility and
can be expected to do so in the future, the political economy of Brazil's
labor market institutions suggests that achieving stability and growth in
the future will continue to be problematic. The government has little
credibility with the unions, who do not believe that an adjustment
program will treat them fairly. Yet adjustment will almost certainly
demand some kind of social contract involving both the unmons and
holders of capital. The alternatives-such as lowering real wages through
a recession-are expensive, especially given the decentralized .bargaiing
structure, and politically highly unpopular- Such a policy would also
favor financial capital if it were implemented through the standard
approach of a tight money policy.
The difficulties of putting together such a social contract are
formidable. Experience in Europe suggests that key ingredients in such a
deal include (a) a belief that the process itself is fair -and that burdens are
being shared fairly; (b) a long-term commitment to the process on all
sides so that inequities in one deal can be compensated in the next; (c) a
crisis serious enough to engage the attention of all parties; and (d) a
government that does not seek to load the process of bagaining with too
many reforms, for example, a sectorally-oriented industrial policy.
168 M.LouiseFo EdwardAmadeo, andJose Marcij Camargo
Whether Brazil has the political and institutional resources to
manufacture a policy that fits these requirements is an open question.
References
Almeida Reis, Jose Quilherme. 1989. "Salario Minimo e Distribuicao de
Renda." In Perspectivas da Economia Brasileira 1989. Rio de
Janeiro: IPEAJIN.
Fox, M. Louise. 1990. "Poverty Alleviation in Brazil, 1970-87."
Internal Discussion Paper (IDP-072). Washington, D.C.: World
Bank, Latin America and the Caribbean Region.
Fox, M. Louise, and Samuel A. Morley. 1990. "Who Paid the Bill?
Adjustment and Poverty in Brazil, 1980-85. Background paper
for World Development Report 1990. Washington, D.C.: World
Bank.
Morley, Samuel A. 1982. Labor Markets and Inequitable Growth: The
Case of Authoritarian Capitalism in Brazil Cambridge, U.K:
Cambridge University Press.
Pastore, Jose, Helio Zylberstajn, and Carmen Silvia Pagotto. 1983.
Mudanca Social e Pobreza no Brasil: 1970-1980. (O que
Ocorreu com a Familia Brasileira?). Sao Paulo:
FIPE/PIONEIRA
Saboia, J 1989. 'Dualism e Integracao do Mecado de Trabalho."
Discussion paper. Rio de Janeiro: Federal University of Rio de
Janeiro-
World Bank. 1988. Brazil: An Assessment of the Current
Macroeconomic Situation. Report No. 7540-BR. Washington,
D.C
1990. World Development Report 1990: Poverty. New
York: Oxford University Press.
CHILE
Luis A. Riveros
Like other developing countries, at the outset of the 1980s Chile
faced persistent internal imbalances and an unsustainable external
deficit. This was partly due to internal factors, especially the economic
policy of the late 1970s, which sustained a revaluation of the real
exchange rate, stimulated low savings and permitted growving external
indebtedness. However, the lending cutback of 1982, a sharp drop in
the terms of trade, and an increase in intemational interest rates also
played crucial roles in creating a deep recession. The economy was
unable to generate a trade surplus quickly in response to these
external developments. After a deep recession, export promotion and
other policies dealing with the debt problem allowed the Chilean
economy to resume sustained growth and internal balance, which was
complemented by a return to a democratic form of government in
1990.
The labor market played a key role in the adjustment Chile's
economy underwent in the 1970s in the post-recession period of the
1980s. As predicted by standard adjustment models, a decline in real
wages occurred as a result of expenditure-reducing and expenditure-
switching policies. However, due to rigidities that hindered labor
mobility, the economy suffered from persistent high unemployment.
These rigidities were mainly associated with labor market
segmentation and expectations associated with the lack of an
The author is indebted to Bela Balassa, Albert Berry, Erik Haindl, Susan Horton,
Ravi Kanbur, Dipak Mazumdar, Ricardo Paredes, and the participants of workshops
beld at the University of Toronto and the University of Chile for valuable comments
on earlier drafts, and to I Charoenwattana and J. Lackman for efficient research
assistance.
169
170 LuisA. Riveros
institutional framework for the labor market. As a result of the post-
1984 export-led adjustment program and the introduction of more
adequate labor market reforms, open unemployment declined,
employment in tradables iincreased strongly, and real wages started to
recover.
Without the profound structural economic reforms of the 1970s,
rapid achievement of growth and macroeconomic equilibrium after
the financial crash of 1982-84 would have been nearly impossible.
These reforms permitted flexible and competitive markets, as required
to achieve macroeconomnic adjustment. 5imilarly, understanding the
Fecarious situation of labor market variables in the late 1970s is
essential in comprehending the effect of the 1980s' crisis on such
variables as open unemployment and real wages.
This chapter reviews the Chilean labor market during the structural
reic ms of the 1970s and the crisis and recovery period of the 1980s.
To analyze the effect of macro policies-and given that existing wage
differentials and the behavior of both unemployment and investment
are central in interpreting the role of the labor market in the
adjustment-it adopts a segmented labor market model to study the
effect of exchange rate policies on labor market variables. The chapter
analyzes the wage determination process in the formal and informal
sectors using a model that examines the effect of typical macro
policies. The implication is that if labor markets were less segmented,
unemployment during the adjustment would have been lower. Given
that segmentation is linked to labor market policies, less intervention
would have been advisable.
The Economic Setting
Chile's achievement of deep economic reforms in the 1970s
radically changed relative prices and reduced state intervention in the
economy. During 1970-73, economic policies inspired by socialism
produced substantial economic and political strain. A military
government took power in September 1973 and instituted
deregulation aimed at correcting major price distortions during 1973-
75. In a second phase during 1975-76, the government placed greater
emphasis on price stabilization while it continued with the structural
reforms. Appreciating real exchange rates, high domestic interest rates,
Chile 171
and labor market friction cre.ated macroeconomic problems during
1976-80 and promoted higher growth of nontradable relative to
tradable production. During this third phase, the economy was
characterized by growing real wages, high growth, and high
unemployment. The 1982-84 financial crisis resulted in a sharp
economic decline amid serious balance of payment problems.
Following the crisis, sharp devaluations combined with reduced
expenditure and other policies aimed at affecting expectations,
increasing savings, and promoting exports led to a notable export-led
economic expansion. This sequence of phases is paramount in
explaining observed labor market outcomes.
The Socialist Experiment, 1970-73
The government of Dr. Allende aimed at making profound
changes in Chile's economy that would achieve key improvements in
distributive results and growth records. The installation of this
economic program, however, involved serious macroeconomic
imbalances. During 1971, the budget deficit increased from 2.7 to
10.7 percent of GDP, and credit from the central bank to the public
sector increased by more than 110 percent. On the external fiont, and
partly as a result of a sharp drop in the world price of zopper, Chile's
major export at that time, international reserves dropped dramatically
from US$390 million in 1970 to US$161 .millic" in 1971, and the
trade balance went from a surplus of US$156 million in 1970 to a
deficit of US$16 million in 1971. On top of traditionally high tariff
rates, the government decided to introduce significant quantity
controls on imports, thereby generating a more recessionary
productive adjustment Moreover, as a result of changes in relative
prices, consumption increased by 12 percent and investment dropped
some 2 percent of GDP in 1971. This increase in consumption was
associated with a sharp growth in real wages of more than 22 percent
during. 1971, which was achieved mainly through traditional
government policies and at the cost of rwoductive investment.
The overheating of the economy that began in 1971 created
increasing problems during the next two years. The government
considered maintenance of its progrcssive and revolutionary image to
be more important than reducing disequilibria (Larrain and Meller
Table 5.1 Macroeconomic Indicators, 1970-88
Output Outplut Gross Fiscal Current Real
GDP grosvth in growth in CPI Ml investmnent deficit account deficit exchange
Year growvth tradaables nontradables Inflation grol/vai (% of GDP) (1 of GDP) (% of GDP) rate
1970 2.1 1.4 2.9 32.5 66.2 16.4 2.7 -1.2 38.5
1971 9.0 9.2 8.8 22.1 113.4 14,5 10.7 -1.9 35.3
1972 -1.2 -0.8 -1.1 260.5 1S1.8 12.2 13.0 -4.0 36.7
1973 -5.6 -7,3 -3.7 605.1 362.9 7.8 24.7 -2.7 56.7
1974 1.0 6.6 -0.4 369.2 231.2 21.2 10.5 -2.6 83.8
1975 -12.9 -16.6 -8.4 343.3 257.2 13.1 2.6 -6.B 100.0
1976 3,5 5.3 1.6 197.9 189.4 12.8 2.3 1.5 91.5
1977 9.9 7.8 9.4 84.2 113.5 14.4 1.8 -4.1 83.5
1978 8.2 4.5 9.6 37,2 65.0 17.8 0.8 -7.1 101.6
1979 8.3 7.0 10,0 38.9 57.8 17.8 -1.7 -5.7 105.9
1980 '7.8 5.5 l,O0 31,2 64.0 21.0 -3.1 -7.1 94.0
1931 5.5 3.8 5,4 9.5 -3.8 22.7 -1.7 .14.5 74.S
1982 .14.1 . -1.2 -10.8 20.7 7.3 11.3 2.3 -9.5 81.1
1983 -0.7 0.5 *6.1 23.1 27.7 9.8 3.8 -5.6 98.1
1984 6,3 7.9 5.3 23.0 12.1 15.3 4.0 -10.7 100.8
1985 2.4 2.5 2.4 26.4 11.3 13.9 6,3 -8.3 123,0
1986 5.7 6.7 5,0 17.4 41.4 15.0 2.8 -6.5 139.9
1987 5.7 3.5 6,6 21,5 9.8 17.9 0.1 -4.6 143,5
19398 7.4 6.9 7.7 12.7 82.4 18.1 1.7 -3.0 149.9
1989 10.0 8.1 10,9 17.1 13.2 22.0 0.4 -4.7 151,2
Notes: Tradables includes agriculture, fishing, mining, and manufacturing. Nontradables includes construction and services. CPI
inflation corresponds to the December to December change in the corrected CPI. The real effective exchange rate is the real
multilateral exchange rate in terms of the wholesale prices of trading partncrs and Chile's cpi. The average ratio of investment to GDP
durng 1960-69 was 14.9 percent. MI growth is the December to December growth of MI.
Sources: Banco Central de Chile (1987); Corbo (1985a); Cottani (1988); Cortazar and Marshall (1980)j IMF (various years); World
Bank (1990). For 1988 and 1989: Banco Central de Chile (1990).
Chile 173
1990). The fiscal situation deteriorted because the government made
no effort to reduce expenditures, while the need for transferring
resources to a growing number of public enterprises placed a heavier
burden on fiscal expenditures. At the same time, tax collection
dropped dramatically, making the fiscal situation even worse. In
response to these developments, the total quantity of rnnrc.y increased
by 152 in 1972 and 363 percent in 1973 (table 5.1). As a
consequence, yearly inflation rates skyrocketed to 261 and 605
percent, respectively, during the same years. Another outcome was the
significant drop in GDP of 1.2 percent in 1972 and 5.6 percent in
1973. This was accompanied by a drop in real wages of 11.3 and 38.6
percent, respectively. The current account deficit rose from US$189
million in 1971 to US$387. million in 1972 and US$295 million in
1973, despite heavier import controls. The government insisted that
the problem was a plot organized by entrepreneurs and insisted on
nationalizing private firms. This increased the political turbulence in
1973, which was already high due to severe shortages and a very
critical opposition party.
During the socialist experiment, union activities reached a peak.
The increase in real wages in 1971 and the permanent efforts to keep
pace with price inflation in 1972 and 1973 were facilitated by a
centralized system of wage fixing in which the national union
confederation (CUT) played a key role. Unemployment rates reached
the lowest historical level in 1972 (3.1 percent), and increased slightly
to 4.8 percent despite the significant drop in GDP in 1973. This was
due mainly to a significant increase in public sector employment.
The Economic Reforms of the 1970s
The military government that took office in 1973 embarked on an
intense program of economic reforms aimed at improving efficiency
in the framework of an open economy. Many observers have analyzed
the specific targets and policy tools used to achieve those reforms (see
Corbo 1985a; Edwards and Edwards 1987; Walton 1985). However,
the direct and indirect impacts of the reform program on labor market
outcomes have received relatively little attention despite their crucial
political impact
174 LuisA. Riveros
The government deemed that major changes in labor market
institutions were necessary in the context of a freer, more deregulated,
open economy. The presence of some intervening policies
notwithstanding; these changes directly affected wages and
employment. Deregulation of output markets, the opening of the
economy, and the reduction in the state's economic size affected
employment and wages indirectly as vital shifts in the skill
composition of the labor demand called for higher labor mobility. A
more detailed description of the main refojrms follows.
TRADE REFORMS. Chile, like many other developing countries, pursued
industrialization based on the creation of a sizable import substituting
industry. Major outcomes of this policy were high trade barriers,
considerable inefficiency, discrimination against agriculture, and
growing govemment intervention in economic management (Corbo
1986). Paradoxically, employment growth in manufacturing was
affected negatively (Corbo and Meller 1984), and labor market
segmentation occurred due to the parallel need for labor protection
created by the appearance of a strong labor movement. As the failure
to achieve an efficient industrial sector required progressively higher
protective barriers, average tariff rates reached as high as 105 percent
by 1973 (Torres 1982).1
The reforms initiated by late 1973 aimed at sharply reversing
import substitution through a far-reaching opening up of the
economy.2 The trade opening was expected to produce more
investment and employment in sectors with comparative advantages.
Two causes account for failure to obtain higher exports and growth of
labor-intensive industries after 1973: (a) the tariff reduction program
did not begin with p precise final target, thereby creating uncertainty
among investors (Riveros 1986); and (b) the exchailge rate was used
L. Ad-valorem tariff rates ranged from 0 to 750 percent,.while import prohibitions
applied to 187 tariff classifications, a 90-day import deposit requirement was in effect
for 2,800 others, and 2,300 categories required special approval from the central
bank.
2. Tariffs were planned t^ Yeach an average 60 percent by 1977. By 1975, average
tariffs had reached 57 percent while almost all quantitative restrictions were
eliminated. In a second stage, a new structure with tariffs ranging from 10 to 35
percent was achieved during the third quarter of 1977. Finally, a more radical reform
allowed the average nominal tariff rate to reach a uniform 10 percent by late 1977.
Chile 175
as a stabilization device, particularly after 1978, thereby permitting
substantial overvaluation (Corbo 1985b; Edwards and Edwards
1987J.3 However, amid high internal interest rates and growing peso
appreciation, the economic authorities decided to open up the capital
account In addition to the macro imbalances so generated (Edwards
and Edwards 1987), this policy allowed large- frms to adopt more
capital-intensive techniques, which also affected the prospects for
employment creation in expanding activities.
PUBLIC SrOR REFORMs. Another set of reforms after 1973 aimed at-
reducing the state's economic size by reducing both government
expenditures and privatization of public firms. The huge fiscal deficit
that existed in 1973 is an indicator of the economic importance the
state had attained. Another indicator is the share of parastatals in total
production, which had been 14 percent of GDP in 1965, but reached
39 percent of GDP in 1973 (Hachette and Luders 1987). Still another
indicator is the state's importance as an employer, especially during
the socialist experiment of 1.970-73, when public sector employment
reached about 15 percent of total employment. Total public sector
employment grew by 38 percent between 1970 and 1973.
To reduce the state's economic importance, the government
implemented a privatization program together with policies aimed at
making the central government more efficient. Private firms
nationalized during the socialist regime were immediately privatized:
by the end of 1974, 202 out of 259 had been returned to their owners
(Larrain 1988). In addition, the government quickly sold some state
assets, with most bank shares (US$171 million) and a significant part
of industrial property (US$58 million) being sold by 1975.4 As
3. After unification of the exchange rate in 1973, an initial 300 percent
devaluation, and a series of niinidevaluations, the real exchange rate reached a peakl
value by late 1975. Subsequently, the real exchange rate declined sharply concluding
in a 10 percent appreciation in June 1976. Real peso appreciation continued being
used as a stabilization device through a system of devaluations that allowed for certain
real appreciation. However, inflation did not drop as expected. From June 1979 a
nominally fixed exchange rate was implemented and maintained until mid-1982, when
dramatic peso devaluations took place in the wake of the world recession. In this
analysis, appreciation is represented by a decline in the real exchange rate.
4. This was a year of unprecedented economic decline (GDP fel by 12.9 percent),
thus sales were not profitable. At the same time, firns were sold below their book
values, although above their stock markcet value (Larmain 1988).
176 Luis A. Riveros
parastatals' share in GDP dropped from 39 percent in 1973 to a still
high 24 percent in 1981, the govermnent initiated further privatization
at the time of the 1980s crisis. Owing to active job creation in the
public sector in 1971-73 and redundant employment, major job cuts
occurred in both the civil service and parastatals between 1973-77,
when total public sector employment declined by 24 percent,
implying an increase in-the unemployment rate of about 3 percent of
the labor force.
MRcErr DEREGULATION. A top priority of the reform program after
1973 was to improve resource allocation through an efficient price
system. Hence, and countering a historical tradition of price fixing,
price regulation was almost completely eliminated. After an era in
which more than 3,000 prices- were set and eventually controlled by
the authorities, only 33 commodities remained under government
control, most of them utilities. Likewise, interest rates were also
deregulated and quantitative constraints for capital market operations
were eliminated before 1979. Entrepreneurs facing increased market
competition had to improve their productive efficiency, which resulted
in several bankruptcies during the period of tasformation.
The wave of deregulation also reached the labor market, which was
considered a crucial area after a period of acute govemment
intervention and union activism. Deregulation resulted in a
fundamental change in wage setting mechanisms and encouraged
labor dismissals aimed at eliminating overemployment. Before
September 1973, the Labor Law- (Codigo del Trabajo) govemed the
institutional functioning of ihe labor markei, and according to this law
wage bargaining could be made at the most aggregated level, thereby
authorizing unions to form confederations and negotiate with
entrepreneurial associations. Immediately after September 1973,
collective bargaining and union activities were simultaneously
eliminated, which also ousted traditional wage bargaining procedures.
Similarly, traditional regulations on job security, which made labor
dismissals relatively expensive, were eliminated. After 1973, massive
labor dismissals only required a simple administrative authorization
from the government. Thus, the new government gave the private
sector full power to implement its desired employment wage strategy.
Chile 177
The importance of institutional changes in the labor market after
1973 is revealed by the fact that once almost all unions were virtually
eliminated, the government hand picked the leaders for the remaining
ones. The most striking result was, however, that during the entire
period 1973-79, no new legal provisions governed labor relations,
including those in connection with wage bargaining, job security, the
right to strike, negotiation of working conditions, and union activities.
Although a set of rules for industrial relations was enacted in 1975
(Decree 1005), it did not introduce any significant changes as
concerned the functioning of the labor market
After 1973, minimuim wages and nonwage cost regulations were
upheld and a wage indexation system was implemented, although it
had little effect given the absence of enforcement mechanisms
(Edwards and Edwards 1987; Riveros 1986). A1s disc.ussed later, a
major failing of the program was the lack of a labor law during 1973-
79, and even of signals about projected legal changes, which created
expectations of high future firing and hirng costs, and affected
private sector decisions on employment.
Between 1979 and 1982, the govemment adopted key changes in
labor market institutions. In 1979, a new law (DL 2756 and DL 2758)
was enacted, establishing new guidelines for unionization and
collective bargaing. Countering a tradition in Chilc, more than one
union was allowed per enterprise, wage barginig could only be done
at the firm level, and the right to strike was curtailed by granting firms
the right to hire temporaries. In addition, the law did not restore labor
courts that had existed before 1973. Instead it established the principle
of voluntary affiliation to umions and banned public employees' right
to strike. Finally, the law also instituted a 100 percent wage indexation
to past inflation as a floor for any negotiation.
After 1982 the govemment made several change; ao the 1979 labor
law. The most important was the elimination oi the full indexation
clause, which had apparently creat.e, subAsan-tA'a problems in
connection ;vith ihe economy's resptse tcK the 19S2 recession. Other'
changes included eliminating special a-trhc`iza;&s required to hold
certain jobs, such as, actors, m!lsicians, and bus drivers. The most
important change was the elimination of special.privileges traditionally
awarded to dock workers, which pemiitted more competition in hiring
178 LuisA. Riveros
labor. Finally, the government eliminated employers' fights to dismiss
workers without justification and established severance compensation
of one month per year of service for workers hired after 1981.
THE STABLIZAnON PROGRAM. The initial phase of the structural reforms
was accompanied by a sharp stabilization effort in 1975-76. As table
5.1 shows, as late as 1975 inflation was still above 300 percent per
year. To-deal with it, the government managed to reduce the fiscal
deficit (measured against GDP) from 24.7 in 1973 to 10.5 in 1974
and 2.6 in 1975, mainly by reducing public sector expenditures. This
was accompanied by tighter monetary policy, and later on by a
deliberate appreciation of the real exchange rate. Inflation was
significantly curbed between 1975 and 1978, and continued dropping
until 1981 as a result of exchange rate management and the creation
of a fiscal surplus instead of the traditional deficit The effect on
unemployment of the across the board. reduction in aggregate demand
was as important as the drop in real wages. The stabilization effort also
affected long-run growth due to its effects on investment and wealth
(Edwards 1985).
The 1982 Financial Crisis and the Policy Response
Domestic economic policies introduced at the end of the 1970s
were primarily responsibe for creating macrocconomic disequilibria
that led to unsustainable, high expectations, and ultimately to a
disruption in the economic recovery initiated in 1976. The full
opening of the capital account when the exchange rate was fixed with
the aim of ccutrolling inflation produced substantial difficulties. This
combination of policies was still nore burdensome in the presence of
a binding full indexation of wages to past inflation. Large capital
inflows financed an otherwise unsustainable expansion of private
consumption and investment- Optimism about future trends was based
on the economy's stable growth since 1976, the balanced public
budget, the sustained improvement in real wages, and the opening of
the economy- However, the economic boom of 197941 was basically
financed with foreign credit, with few resources being allocated to
productive investment In addition, a poorly controlled financial
system was weakened by a deterioration in the quality of its loans to
Chile 179
the corporate sector. This was aggravated by the normal practice of
granting credits to firms whose ownership was interlocked with that of
the lending institutions.
Some :ndicators reveal the magnitude of the crisis the Chilean
economy faced in 1982. Total external debt increased from 2.7 times
total exports in 1979 to 4.6 times in 1983. As a share of GDP, the
external debt increased from 40 percent to 100 percent during the
same years, while yearly interest payments increased from 3 to 10
percent of GDP. At the same time, and due to an early policy reaction
to the crisis, the value of imports in. real terms declined by more than
40 percent (table 5.2), which precipitated a GDP drop of more than 15
percent in GDP (table 51). This GDP drop was accompanied by an
Table 5.2 The External Sector, 1979-87
.(billions of current US$)
Caregory 1979 1980 1981 1982 1983 1984 198 2986 1987
External debt
Total 849 11.80 1550 17.10 17.40 18.90 1930 19.40 19.10
Interest 0.67 0-93 1.46 1392 1.75 2.02 1.90 1.89 1.70
Debt/exports 2.70 2.40 4.10 4.60. 4.60 5.20 5.10 4.60 3.70
DebtIGNP 0.40 0.40 O0. 0.80 1.00 1.10 1.40 1.30 1.20
Exports (aob.)
Traditionals 2.16 2.62 2.18 2.12 2.34 1396 2.12 2.10 2.60
Nontraditional L68 2.09 1.66 158 150 1.69 ..68 210 2.62
Total 3.84 4.71 3.84 3.71 3.83 3.65 3.80 4.20 5.22
Total 4.33 &71 4.41 4.87 4.78 4.92 532 5.92 na.
Imports (cI.L)
Conmergoods 1.33 2(7 7.73 1.48 1.02 1.04 751 754 9.01
Capila! goods 0.95 1.7 1.45 0.70 0.39 0.60 0.65. 0.74 1.10
Toi irnpons. 4.71 6.15 732 4.09 3.17 3.74 3.27 3.44 4.40
Total imports 5.81 6.19 6.25 4.25 3.36 4.04 3:59 3.65 n.a
Terms of tade
(1980 = 100) 11850 100.00 84.30 80.40 87.50 83.20 7850. 82.00 77.00
ELao = not available
Series is expressed in cvnstant 1980 billions of dollars.
Notes: Data for ;1987 are preliminary. Traditional exports include copper and muiing.
Nontraditional exports are agricultural and industrial products.
Sources: Bolelin Mensual, Banco Central de Chile: World Tables UBRD); Corbo &
Sturzeneggcr (19881.
180 LuisA. Riveros
even larger decline in aggregate investment, which affected future
growth. The magnitude of the external shock is demonstrated by the
drop in terms of trade (from 119 in 1979 to 88 in 1983) and the
increase in real (LIBOR) interest rates from 2.6 to 4.6 percent (see
Corbo and Sturzenegger 1988). The slharp curtailment of capital flows
in early 1982 amplified the problemn and led the economy into a deep
recession in 1982-43.
During 1979-S1, the authorities also shifted their attention away
from stabilization policies and toward structural adjustment. During
this period, the government implemented a series of reforms aimed at
changing traditional practices with regard to social policies and
administration of social welfare. In 1981, the government made a key
change to the social security system, changing it from a pay-as-you-
go system to one in which benefits depended only on individual
contributions. Also in 1981, the government introduced health system
reforms aimed at promoting a private health care system. It also
decentralized the education and health systems to make thiem more
responsive to local needs. These changes were important during the
recovery from the world recession, as they permitted L.tetter targeting
of fiscal social expenditures on the poor.
In 1984, a so-called "adjustment without recession" approach
resulted in huge reserve losses and firther external indebtedness. This
policy stimulated aggregate demand to encourage output growth.
However, the availability of external financing posited a tough
constraint to the planned expansion i aggregate expenditures.. The
current account deficit almost doubled between 1983 and 1984, while
the fiscal deficit also increased considerably (table 5.1). Although
unsustainable in the mediuni run, this approach resulted in a GDP
growth of more than 6 percent in 1984, but also resulted in declining
exports and a large growth in external payments (table 5 2). What the
economy needed to face the recession was sensible policies aimed at
encouraging exports and a continuation of the structural reforms.
After 1985, the adjustment program focused on taining a high
real exchange rate, privatizing public firms, controlling fiscal
expenditures; creating mechanisms that allowed conversion of extemal
debt into investment (which has yielded a drop in the total external
debt of more than 10 percent), introducing specific incentives for
Chile 181
exports, and targeting fiscal social expenditures to the poor. This
combination of policies resulted in a resumption of strong economic
growth (more than 5.5 percent per annum in 1986-88 and about-10
percent during 1989), with a large expansion of nontraditional
exports, investment growth, low inflation, and sharply declining open
unemployment. The year 1988 was a culmination of a successful
adjustment achieved by accentuating the role of markets in allocating
resources.
Labor Market Effects of the Adjustment Program
The economic reforms of the 1970s had major effects on the labor
market. One of the most important was the increase in open
unemployment rates from an average of about 6 percent of the labor
force during the 1960s to more than 16 percent during 1974-81
(table 5.3).5 Moreover, even with high GDP growth between 1976 and
1981, open unemployment remained at relatively high levels (table
53). A related result was the decline in average real wages. In
addition, during 1976-81, employment in nontradable activities
expanded more rapidly than in tradables, a result not concordant with
the outward orientation of the economic program, but explainable in
the context of the signals provided by an appreciating real exchange
rate (table 5.1). Finally, traditional labor market institutions-like-
wage bargaining, unionization, and job security laws-were not legally
reinstated until 1979, which probably effected expectations and
countered employment creation in expanding tradable activities.
Employment and Unemploymrent Tre-nds
Unemployment rates increased significantly after 1974, causing
concern a-bout the social impact of adjustment policies.
Unemployment was proportionally higher for the relatively more
skidled labor force, as revealed by unemployment rates broken down
by education (Riveros and Diaz 1987). Similarly, unemployment rates
were higher for older people, possibly because those entering the
5. Thia average includes persns in emergency employment programs (EEP). As
discussed later, this calculation yields an economically meaningful unemployment
level. If EEPs are not included, the average uncuployment-in 1974-81 reaches about
13 percent of the labor force.
Table 5.3 Employment and Unemployment, 1970-89
Caieqory 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979
1. Populalton (ihousands of persons)
(a) 12 years and over 6,455.6 6,636,3 6,815,4 6,992.5 7.164,2 7,339.1 7,515,0 7,691.5 7,866.7 8,057.1
(b) Total labor foxce 2,932.2 2,978,8 3,000.8 3,039,0 3,066.8 3,152.9 3,216.4 3,259.7 3,370,1 3,480.7
(c) Paticipation ratce 45,4 44.9 46,8 43,5 42.8 43.0 42.8 42.4 42.8 43.2
2. Emptoyrenat (thousands of persons)
(a) Totalemployment(UCHI) 2,766.1 2,865.6 2,907,8 2,893.1 2,784.7 2,727.3 2,705.0 2,796.8 2,391.5 3,000.4
(b) Totalcriployment(INE) n.a. 2,880.5 2,901.8 n.a. n.a. 2,777.3 2,820.5 2,981.3 3,003.3 3,257.1
(c) Emergencyemploymentprogram - - - - 71.5 172,0 187.6 145.8 133.9 191.0
3. Une,nployment (percenw)
(a) Castaneda (UCH) 5.7 3.8 3.1 4.8 9,2 13.5 15.9 - 14.2 14.2 13.8
(b) Correced (a) 5,7 - - - - 15.5 20.6 19.2 18.0 17.2
(c) INE n.I. 3.7 3.3 3.3 n.a. n.a. 12.7 11,8 14.2 13.6
(d) Caorected(c) - - 17.4 16,9 17.9 17.0
4. Sectoral employment (ihousands of persons)
(a) Tradables 1,206.0 1,223.0 1,178.0 1,151.5 1,128.2 1,088.2 1,017.1 1,060.0 1,058.1 1,086.4
Agpiculure 625.6 588.0 S30.5 500,0 510.0 534.6 486.9 517.4 514.7 512.1
(b) Nontrdables 1,560,1 1,642.6 1,729.8 1,741.6 1,656.2 1,575,0 1,536.4 1,571.4 1,705.0 1,799.1
(c) Public sector employment 280.0 325.3 342.0 387.2 360,2 325.5 314.3 295.9 293.3 315.7
(d) Rado of tradable employment to
nontradable employment 0.77 0.74 0,68 0.66 0.68 0.69 0.66 0,67 0.62 0.61
J. Population (ihoutsands of persons)
(a) 12 years and over 8,207.4 8,369.7 8,527.0 8,681.9 8,888.6 9,096.0 9,309.6 9,502.7 9,710.2 9,922.2
(b) Total labor force 3,539.8 3,669,3 3,729.5 3,797.1 3,937.1 4,071.8 4,160.3 4,288.3 4,455.0 4,620.6
(c) Pariclpation tale 43.1 43.8 43.7 43,7 44.3 44.8 44.7 45.1 45.9 46.6
2. Eniploymen: (thousands of persons)
(a) TotalemploymenitCUCH) 3,122.1 3,269.3 2,971.5 3,091.2 3,185.1 3,420.3 3,582.0 3,748.0 3,911.5 4,163.2
(b) Totalemployment(INE) 3,270,9 2,943.1 3,215.8 3,349,4 3,537,4 3,895,7 4,010,8 4,110.8 n.a. n.n.
(c) Emergency employment progmm 175.6 226.8 502.7 336.3 324.3 233.5 148.5 4^ 2 - -
3. Unemployment (percent)
(a) Casianeda (UCH) 11.8 10.9 20.4 18.6 19.1 16.0 13.9 12.6 12.2 10.1
(b) Corrected(a) 16.5 15.1 25.7 30.1 22.9 20.9 18.0 15.2 13.1 10.1
(c) NE 10.4 11,3 19.6 14,6 13,9 12.0 B8, 7.5 6.3 5.3
(d) Comcrced(c) 15.0 15.5 25.0 26.2 21.4 19.0 13.6 10.0 7.2 -
4. Sectoral enmployment (thousands of persons)
(a) Tradables 1,113.3 1,164.6 1,037.1 993.5 1,021.6 1,035.9 1,173.6 1,273.9 1,374.7 1,491.6
Agriculture 518.3 546.9 531.9 510.1 532.7 561.0 570.9 605.2 647.3 657.6
(b) Nontradables 1,840.7 1,950.2 1,734.8 1,655.2 1.867.6 2,060.1 2,212.9 2,343.4 2,496.1 2,671.6
- (c) Public secloremployment 258.9 281.3 230.5 357.6 356.2 381.6 360.6 3;' 1 .302.7 n.a.
(d) Ratio of tradable employment to
nontradable employment 0.60 0.59 0.60 0.57 0.57 0.52 0.53 . 54 0.56 0.56
n.a. = not available
- = not applicablc
Note: The Emergency Employment Program cxisted only between 1974 and 1987,
Sources:
I(a). 1970-83 from Castaneda (1983), 198489 projected on the basis of UCH surveys;
1(b). 1970-83 from Castaneda (1983). 1984-89 estimated with the growth rate (March-March) in UCH surveys;
2(a). Banco Central de Chile (1987);
3(a). Castaneda's figures using [NE's unemployment rate corrected for rate of participation; 198489 based on UCH surveys;
3(b). is 2(n) adJusted by employment emergency program during 1975-88: U-corrected = (tU) + (EEP#0,88)/(LF)+(EEP*.05), where LF total labor
force, U = uncorrected unemployment rate, 1EP = number of members;
4(a), estimated based on Riveros (1985a) and Banco Central de Chile (1987);
4(b). 1970-83 from Budnevich and othcrs (1986), 1984-89 based on UCH surveys; figures exclude 88 percent of EEP members (see text);
4(c). Paredes (1987).
184 Luis A. Riveros
labor market were more flexible about accepting lower wages. At the
same time, employers considered younger job searchers easier to train,
and thus they were more likely to obtain a job. The duration of
unemployment also increased dramatically, from an average of six
months in the 1960s to more than a year in 1975-79 in the Greater
Santiago area (Riveros and Diaz 1987).
Explanations of the persistent high unemployment of the 1970s
occupied a prominent place in the literature (for a review see Meller
1984; Riveros 1985b). After a protracted debate, economists reached
consensus that alternative, explanations have to be combined to
provide a consistent theory on the unemployment figures shown in
table 5.3.6
* A first explanation refers to the higher labor force growth seen in
the 1970s compared to the 1960s. Although participation rates did not
increase, and even declined sclightly in 1974-80, labor supply growth
was triggered by the post-World War II baby boom (Castaneda 1983).
Some estimates showed that this supply effect may have accounted for
no more than 3 percent or' the higher open unemployment seen in the
1970s (Riveros 1986).
A second explanation refers to the effect of policies on public
sector and trade reforms. Figures on public sector employment (table
513) show a dramatic decline after 1973, which probably raised total
unemployment in the short run. As already mentioned, between 1973
and 1977 public sector employment fell by almost 3 percent of the
labor force, while the public sector's wage bill declined by about 3
percent of GDP in 1973-76 (Larrain 1988).7 Moreover, the
elimination of job security amid increased competition in product
markets gave nse to a drastic reduction in redundant employment in
the private sector (most entrepreneurs declared to a survey carried out
through a World Bank study that one of the most beneficial reforms in
6. Statistics presented in table 5.3 adds to total observed unemployment those
included in EEP but declaring themselves as "employed" to the surveys (see Riveros
1986). The economic int-rest of this corrected series is that it permits observation of
actual supply pressures on the labor market.
7. According to Paredes (1987), with data taken from Marshall and Romaguera
(1981), public employment in 1973 reached about 388,000 persons. Cortes and
Sjaastad (1981) suggested a more drastic decline between 1973-76, with public
employment declining by more than 6 percent of the labor force.
Chie 185
connection with firms' adjustment was that referring to labor laws).
Hence, the array of reforms of the 1970s would have transformed the
hidden unemployment that existed before 1973 into open
unemploymenit.
A third explanation refers. to the existence of labor market
imperfections. The two explanations described above assume that
wages did not play any significant role in accommodating a larger
labor supply or a decline in labor- demand. Although this is a likely
short-term outcome, a prolonged period of high unemployment may
be ascribed to market imperfections. The basic hypothesis here is that
the existence of protected/ unprotected sectors in the labor market-
and -possibly the expansion of the informal sector during the 1970s-
led to increased quasi-voluntary unemployment associated with'
queuing by informal sector workers for formal jobs. The observed
evolution of wage differentials supports this -hypothesis, particularly
because of the increase in the ratio of minimum wages to average
unskilled labor after 1974 (Riveros and Paredes 1989).
A fourth explanation refers to skill mismatches derived from the
major productive shifts associated with the structural reforms, which
would have produced significant shortages and/or surpluses of
industry-specific skills. This is suggested by the presence of a growing
tradable/nontradable wage gap for both skilled and unskilled labor,
and by increasing returns to the general human capital in economic
sectors undergoing expansion (Riveros 1986). Prolonged friction was
associated with the wron]g signals emanating from both- the lack of
labor laws and an appreciating real exchange rate. This did not allow
expansion of more labor-intensive sectors like agriculture and export
manufacturing, which are less intensive in specific skills than typical
import substituting activities. The result was a stubborn persistence of
wage differentials and lack of labor mobility toward expanding
industries.
While employment in tradables did not expand significantly in
-1976-81 (table 5.3), employment in nontradables was much more
dynamic, growing at an average of 5.8 percent per annum.8 This
8. Employment in tradables grew at an average of only 13 percent per annum and
employment in agriculture at 0.8 percent per annum.
186 LuisA. Riveros
higher growth was mainly associated with construction activities,
private services, and the financial sector. As table. 5.1 showed,
nontradable production also expanded relatively more during the
1970s (at 7.7 percent per annum in 1976"81, while tradables grew at
only 5.7 percent per annum), thus revealing a major.problem with the
signals provided by exchange rate policies, and later by the opening
of the capital account.
Unemployment skyrocketed during the financial crisis, when the
corrected rate reached as high as 30 percent of the labor force (table
5.3). This result was basically demand-driven,.as participation rates did
not change significantly with the recession. In addition, employment
* in tradable sectors dropped relative to total employment, an outcome'
.probably associated with the increase in urban informal jobs. This
caused a decline in urban informal wages, as suggested by the increase
in relative unskilled wages. Furthermore, real wages declined as
unemployment was increasing.
-As a result of the policy stance of 1984,. total corrected
unemployment fell significantly as employment grew by about 10
percent and emergency.employment declined. Nevertheless, as the
1984 program-was short-lived, real wages declined by more than 4
percent in 1985, at the time that the real exchange rate further
depreciated (table 5.1). The unemployment rate experienced a steady
decline since 1984, in a way apparently correlated with the recovery in
GDP, which was accompanied by expanding employment and
production in tradable industries, and by a notable increase in
nontraditional exports (table 5.2). Employment in agriculture and
manufacturing mining grew at 6 and 11 percent per year, respectively,
in 1985-88, when total employment grew at 6 percent per year. Even
though the economy underwent a major adjustment program during
1985-87, output and employment grew considerably in agriculture
and manufacturing as a result of strong exports. At the same time, a
more deregulated labor market, in which the institutional framework
.was clearly defmed, and the maintenance of high real exchange:rates
allowed the labor-intensive sectors to expand.
Chile 187
The Performance of Real -Wages
There are two basic data sources on wages. First, the National
Bureau of Statistics (INE) wage index, which is prepared on the basis
of firm-based surveys, provides information about formal sector
activities. Second, the University of Chile (UCH) labor force survey
for the Greater Santiago area, which collects information on labor
incomes once a year, covers both the formal and the informal sectors.
Yanez (1987) constructed sectoral wages based upon this information,
which are the ones used in this study.
The official (INE) wage index shows a decline in real formal sector
wages during the recessionary years of 1975-76, following the
dramatic drop of 1973 -caused by spiraling inflation (table 5.4,
column 2). Figures from the University of Chile's surveys (table 5.4,
column 6) indicate a very similar trend, though the earlier decline is
more dramatic and the recoverv in 1975-81 is stronger, probably
because of the more procyclical nature of labor earnings in informal
activities. The real minimum wage was relatively stable in 1976-80,
but declined sharply during the postcrisis period (1983-87) Note that
the minimum wage grew significantly relative to both the equilibrium
wage for unskilled labor in the informal sector and the average wage
in the economy. Hence, minimum wages were probably important in
affecting both average wages (Paldain and Riveros 1987), and the level
of employment of less skilled workers (Riveros and Paredes 1988,
1989). The argument here is that inaeases in the minimum wage
increased open unemployment and-caused the withdrawal from the
labor force of low-skilled people, particularly women and young job
seekers.
Average real wages resumed growth during the 197641 expansion
despite unemployment rates higher than historical averages.
Nonetheless, during the growing phase (1976481), unemployment was
declining, which would suggest adequate, though probably slow,
allocative work of the labor market In interpreting observed wage
growth in 197641, some observers- (for example, Cortazar 1983)
have suggested a binding role of the indexation mechanism
implemented by the military government, whereby the private sector
was advised to award workers the same wage increase given to public
188 Luis A. Riveros
Table 5.4 Real Wages, 1970-89
(index, 1980 = 100)
INE University of Chile
Total
Minimum Average Manufactured Unskilled Skilled Average Manufactured labor
wage wage wage wage wage wage wage cost
Year (1) (2) (3) (4) (5) (6) (7) (8)
1970 81.0 109.7 97.1 124.1 144.0 12.4 113.9 0.69
1971 106.8 134.1 106.6 133.7 182.9 117.8 135.3 1.09
1972 101.7 119.0 101.2 138.4 157.7 104.3 92.9 1.06
1973 46.2 73.1 68.4 99.0 943 58.4 49.6 0.58
1974 90.6 70.2 62.2 94.5 78.4 56.2 53.6 0.53
1975 105.1 62.5 58.3 77.3 66.7 62.8 72.0 OA3
1976 100.1 78.9 7EL6 77.0 77.7 715 80.4 0.60
1977 84.6 79.9 7EL6 80.3 895- 81.3 84.9 0.8
1978 97.4 85.0 87.4 824 97.6 91.0 101.7 1.06
1979 983 92.0 94.2 102.6 1033 99.6 1081 1.28
1980 io.0 100.0 . 100.0 100.0 100.0 100.0 100.0 1.45
1981 111.0 108.8 1155 120.4 117.7 114.4 131.5 2A42
1982. 122.2 108.6 11a7 126.6 133.3 133.2 146.7 2.08
1983 93.2 97.0 102.9 80.9 95.2 95.1 111.5 1.30
1984 82.1 97.1 99.0 77.8 89.8 89.4 101.3 1.13
1985 743 93.0 97.1 75.6 74.4 75.5 85.1 0.77
1986 .78.8 95.0 101.9 64.4 66.6 72.6 7&7 0.78
1987 66.9 93.1 103.2 63.1 652 71.4 81.1 0.81
1988 74.2 98.6 109.8 64.5 67.1 72.1 86.3 091
1989 75.1 100.6 114.0 65.1 71.2 79.7 91.4 0.97
n.a. = not available
Note: (8) is expressed nominal- dollars per hour.
Sources: (1): INE (two legal minimums existed in 1970, one for white and one for blue
collar workers; the latter is used here; (2)-(3): INE; (4)-(5): Paredes (1987), May of
each year. Unskilled wages are a proxy for informal sector wages.and correspond to an
average for self-employed workers with less than eight years of schooling. Skilled
wages are a proxy for formal sector wages and correspond to an average for blue and
white collar workers with more than 8 years of schooling.. (6)-(7): Yanez (1987),
May of each year. All the wage data have been defated by the average corrected CPI
based on ENE, Yanez (1979), and Cortazar and Marshall (1980). (8): Riveros (1988).
sector workers for the period up to 1979. Thereafter, there was 100
percent indexation. The hypothesis that indexation resulted in a
growth trend in real wages is a very unlikely explanation, at least for
the period 1973-79, when labor laws were lacking, unions were
Chite 189
banned, wage bargaining was suspended, job security did not exist,
and unemployment was relatively high. The wage indexation scheme
prevailing in 1973-79 was ineffective not only because of poor
enforcement, but also because the past CPI inflation was normally
higher than mandated adjustments (Edwards and Edwards 1987), and
the actual (corrected) inflation was even higher than the official level
(Cortazar 1983). Moreover, the mandated adjustnents were not a legal
obligation that could be enforced by the authorities. Possibly, a
neoclassical model, that is, one based on the role played by
unemployment, inflation, and labor productivity; would be more
suitable to explain observed nominal wage trends during this period.
However, in 1979-82 when the new labor law was enacted that
included a full wage indexation rule, the explanation of a binding
indexation postulated by Cortazar is more acceptable, because labor
market institutions were allowed and the wage indexation mechanism
was established by law.
Real wages declined strongly between 1982 and 1983, partly
because of the recession, but also because of the elimination of the
legal wage indexation established in 1979. After 1983, and
particularly after 1985, average real wages remained basically
constant, and increased in the case of manufacturing. This contrasts
with wage behavior during the- economic expansion of 1976-81.
During this period, high unemployment coexisted with growing real
wages, probably because in the absence of a legal framework for the
labor market, entrepreneurs were unwilting to hire more labor and
preferred to expand production on the basis of increasing the number
of hours worked and providing more incentives to increase labor
productivity. In addition, as other studies' have shown (Riveros 1986),
the labor market during this period was characterized by substantial
skill mismatches, which probably resulted in higher wages for those
employed and owners of the specific human capital. In the post-1985
period, by contrast, employment expanded significantly, particularly
in the case of tradable industries and agriculture. This was facilitated
by the existence of a legal framework and a less severe skill mismatch,
given that the activities were basically labor intensive.
190 Luis A. Riveros
Wage Indexation
Analysis of wage indexation in 1976-82 is important in connection
with the effect of wages on oWier macroeconomic -variables,
particularly inflation and the competitiveness of domestic production.
It is also crucial to-understand the role of institutions in determining
the path of real wages. Our strategy to statistically analyze the effect of
wage indexation is to distinguish the period when indexation was
unaccompanied by enforcing mechanisms (1974-79) from the period
when legal wage indexation was established in combination with a
more appropriate legal framework (1979-82).
'We used quarterly INE. data on wages-the same information
Cortazar used for the entire period 1974-81-to estimate a regression
equation in which mandated wage increases (M) "explain" the actual
change- in average wages (W). This model assumes that the
performance of wages is entirely explained by government policies.
Regression results reveal that in 1973-79 the presumed causality from
mandated wage increases to average wage growth is debatable. As
table 5.5 shows, the correlation between these variables is high, and the
parameter associated with M reaches a value of 1.0. However, when the
quarterly inflation rate (P) is included, the effect of M on W does not
have any interpretation: it has a negative sign. When the past rate of
inflation (P-1) is included instead of actual inflation, the parameter
associated with M is still significant, but its effect on average wages is
Telatively smaller than that of inflation.9 Hence, the causality role of.
mandated adjustments in terms of observed wage changes is not easily.
identifiable using statistical analysis.
The observed effect of mandated adjustments on average wages
seems to reflect only the effect of price inflation on wages. Therefore,
the performance of wages in the Chilean economy during 1973-79
may be better explained by an economic model that takes inflation
and the existence of open unemployment into account. A standard
short-run wage setting equation was adopted to test this hypothesis, in
which nominal wage growth is explained by price inflation,
9. When we included laggped adjustments (M-1) and actual inflation, both
parameters were significant and the values& were 0.70 and 0.26, respectively.
Alternative lag structures were used with regard to both M and P.
Chile 191
Table 5.5 Wage Indexation Results, 1973.3-1979.2
Specification Constant M P P-I R2 DW
(j)S 4.57 1.00 0.72 1.74
(1.12) (7.00) (51.20)
(2)* 6.05 0.60 1.37 0.97 1.98
(4.20) (-3.65) (10.50) C93.50)
(3)* 0.21 0.40 0.65 0.88 2.16
(0.78) (3.00) (5.75) (71.70)
* = Correetion for fist-order serial correlation was implemented.
Notes: The method of estimation was OLS. The value of the t-test is presented (in
parenthesis) under the corresponding parameter. The value of the F test is presented in
the f2 colulmn.
unemployment (U), and the growth in average labor productivity (q).
In estimating a regression model, we assume that only the cyclical
portion of total unemployment (UC) is actually able to affect market
wages (Lopez and Riveros 1988; Riveros and Paredes 1990).10 We
found a significant effect of both unemployment and inflation on
observed wage changes in the period 1974.2-1979.2 (table 5.6, row
1). In addition, when the variable M (mandated wage increases) was
included in the equation, it produced a nonsignificant coefficient. This
evidence suggests that wage setting in this period was driven by
economic rather than institutional forces.
Quite another stor fits the period 1979.3-1982.2, when wage
indexation was included in the labor law enacted in 1979, which also
gave rise to unionization and fornal wage negotiation. As table 5.6
(row 3) shows, the effect of lagged inflation is significant, but the
overall fit is much poorer, while the parameters of unemployment are
statistically equal to zero. This suggests that nominal wage behavior in
this latter period was probably dictated by institutional indexation
rules rather than by economic forces.
10. The estimate of cyclical unemployment was obtained from a regression
discussed later, in which total unemployment is set as a function of both structural and
cyclical variables. This is a version of the model proposed by Lopez and Riveros
(1988) and applied by Riveros and Paredes (1989).
192 Luis A. Riveros
Table 5.6 The Unemployment-Wages Tradeoff
Period Constant P UC q R2 DW
1974.2-79.2
M1* 24.1 0.918 -21 l -0.02 0.97 1 5
- (3.98) (24.60) (-3.4) (-0.32) (220.30)
1979.3-82.2
(2) 9.1 0.23 -4.7 0.04 0.10 1.9
(0.68) (0.48) (-0.4) (0-14) (2.92)
(3)*. --1.2 097a -1.4 0.37 0.42 1.8
- . (0.15) (337) (-0.2) (1.61) (4.41)
* Correction for first-order serial correlation was performed.
Notes: The equations were estimated with OLS. The t-values are presented under the
corresponding coefficients. The F value is presented under the R2.
a The parameter corresponds to the variables P-1.
This wage analysis establishes the importance of institutional labor
market intervention policies. Wage indexation created substantial
rigidities to accommodate production and employment to a changing
external environment, because it was binding at a time when the
exchange rate was nominally frozen- Thus, indexation caused a
notable increase in wages in terms of the price of ;radables (Corbo
1985a)J1 In addition, indexation further segmented the labor market
during a period of output and employment growth. If the wrong
signals had caused deteroration in employment growth prior to 1979,
in 1979-82 overvaluation and the blow to tradables production were a
further reason for persistent unemployment
Relathze Wages
The ratio of skilled to unskilled wages is one proxy for the relative
wage of the formal to informal sectors. The wage of unskilled workers
corresponds to urban self-employed workers with less than eight years
of formal schooling, which is a good proxy for the typical member of
the informal sector. The ratio between skilled and unskilled wages
dropped during the 1970s recession, but increased with the recession
11- Data on dollar labor costs in table 5.4 (column 8) indicate the substantial
increase associated with the period under discussion.
Chik 193
of the 1980s (table 5.7). During the former period, the demise of job
security probably affected skilled labor more. During the economic
recovery period of 1976-81, this ratio increased until 1978, but
dropped significantly in 1979-81, probably due to an expansion of
informal sector activities. The ratio between the average-wage indices
of INE and UCH may also be taken as representative of formal-
informal wages, given that the former include only relatively large
enterprises. As table 5.7 indicates, this ratio aIso declined with the
recession of the 1970s, although not as much as the ratio of skiled-
unskilled labor. During 1976-79, the INE-UCH ratio dropped, then
increased in 1980, thereby showing a different trend than that
displayed by the skilled-unskilled ratio, probably because the skilled-
unskiled differential within the formal sector changed.
The ratio of minimum wages to average wages increased
significantly in 1973-75, mainly due to an increase in the fonner.
After that period, minimum wages suffered periods of decline (for
example, 1975-77, 1978-80, 1982-7) and expansion (1977-78,.
1980-82). The behavior of this ratio was basically driven by periodic
mnmum wage adjustments, which were probably very relevant for
unskilled labor employed in the formal sector. This is also suggested
by the decline of the relative wage of unskilled labor-which
corresponds to the informal sector of the labor market-during
periods when the minimmTn wage was increasing, for instance, 1973-
75 and 1978-81, due to -the spillovers of unskilled labor from the
formal to the informal sector.
The ratio of public to private sector wages declined dramatically in
1970-76, and stayed relatively constant until 1980. After 1973 this.
result must be linked to the privatization of public firms, which
produced a downward bias in the public sector average wage as
obtained from labor force surveys. With the recession of the 1980s
public sector wages carried a more substantial burden of the economic
adjustment. With regard to the ratio of white collar to blue collar
wages, an interesting point is that the decline suffered during the
recession of the 1970s was more significant than during the 1980s
recession. The ratio's behavior in the early 1970s is also associated
with the government's practice during 1970-73 of promoting
194 LuisA. Riveros
Table 5.7 Relative Wages, 1970-89
Average wage Public sector White collar
Skilled Minimum wage JNE wages wages Tradable"
unskilled average wage average wage private eCtor blue collar nontradable
INE UCH wages wages
Year (1(2) (3) (4) (5) (6)
1970 116.1 73.8 97.6 144.4 97.4 119.9
1971 136.7 79.6 113.8 138.2 106.6 1062
1972 113.9 85.5 114.1 126.8 90.5 109.3
1973 95.0 63.2 125.2 116.9 76.2 119.1
1974 83.0 129.1 124.9 106.1 73.8 1324
1975 86.3 168.2 99.5 109.8 81.5 130.4
1976 101.1 126.9 110.4 99.1 93.7 11&8
1977 111.7 105.9 983 106.1 1033 118.7
1978 119.4 114.6 93.4 106.0 99.7 116.0
1979 1012 106.8 92.4 108.2 93.0 980
1980 100.0 100.0 100.0 100.0 100.0 100.0
1981 98.6 102.0 95.0 91.7 95.7 124.8
1982 105.3 1125 81.5 91.5 103.0 118;6
1983 118.5 961 102.0 6&8 105.0 110.9
-1984 115.0 84.6 108.5 107.7 110.3 122.3
1985 981 79.9 109.3 106.0 104.0 121.1
1986 103.4 82.9 130.9 101A 100.0 10&1
1987 103.3 719 130.4 104.7 98.3 107.7
1988 104.0 68.6 128.6 103.4 99.1 10&4
1989 1093 68.1 129.1 102I. 99.6 109.1
na. not available
a. Tradables = manufactring, nontradables = construction.
Sources; Columns 1-3: table 5.4; columns 4-6: Yanez (1987) and UCH surveys.
relatively higher wage increases for blue collar workers along the lines
of traditional distributive policies.
Earning Functions
With the purpose of qualifying the trend displayed by average
wage data, it is important to analyze earning functions based on data
provided by the University of Chile's labor force surveys. A standard
Mincer-type earnings function reveals that the average social rate of
return to schooling has increased over time, possibly due to the
Chlile 195
selectivity bias created by higher unemployment (Riveros 1990).
During 1978-81, however, that rate of return did not change
significantly. Analysis based on cost-benefit comparisons leads to the
conclusion that the private rate of return to schooling has declined
over time, particularly in the case of primary and secondary education
(Riveros 1990). The rate of return to experience (table 5.8) suggests
very stable behavior during most of the period analyzed except for a
notable increase in 1978.
Distributive Results
Income distribution deteriorated sharply as a result of the i-form
program of the 1970s. The social cost of the adjustment was
associated with the appearance of higher unemployment and declining
real wages (Riveros 1985b). The Gini coefficients for the family
income distribution (table 5.9), for instance, increased notably in
1974-76 to decline only slightly afterwards (between 1976 and.
1979). Strikdngly, during the expansionary -years 1979-81-amid a.
financial boom-the observed income distribution deterorated further
(table 5.7), which is consistent with the persistence of unemployment
and low real wage levels that still existed during that period. By
Table 5.8 Earning Functions
(dependent variable: Ln of income, selected years)
(LnY= aco +alS+a2X+a3X2 + c4LnH+p)
Category 1968 1972 1976 1978 I982 198$5
Schooling (S) 0.1374 0.1280 0.1331 0.1572 0.1507 0.1512
Experience(X) 0.0560 0.0518 0.0520 0.0640 0.0581 0.0572
Experience sq. (X2 -0.0006 -0.0007 -0.0007 -0.0008 -0.0007 -0.0007
Ln hours worked 0.2227 0.1275 0.3915 0.3498 0.3874 0.4320
Constant 1.3687 0.7684 0.7390 l.9110 3.2151 2.9710
Adj. R-sq. 0,4810 0.4300 0.4210 0.4850 O.4610 0.4720
Notes: Method of estimation is the OLS. All parameters are statistically significanL.
Source: Riveros (1990).
296 Luis A. Riveros
Table 5.9 Social Indicators, Selected Years and 197488
Fiscal social Pcr capita
income shtare of expenditure GDP
Gini coefficient 40%poorest (% of GDP) (1976 US$)
Years (1) (2) (3) (4) (5)
1960 0.4590 n.a. 13.59 n.a. n.a.
1965 0.4750 n.a. 12.87 n.a. n.a.
1968 0.4980 n.a. 11.70 n.a. 1,114
1970 0.5010 0.4345 11.50 n.a. 1,137
1974 0.4499 0.4232 n.a. 11.08 1,090
1975 0,4710 0.4127 12.78 10.30 0,933
1976 0.5380 0.4886 n.a. 9.99 0,950
1977 0.5260 0.4762 n.a. 10.56 1,026.
1978 0.5197 0.4662 n.a. 10.16 1,091
1979 0.5179 n.a. n.a. 9.2S 1,162
1980 0.5257 n.a. 10.88 10.29 1,231
1981 0.5220 n.a. .11.24 12.80 1,277
1982 0.5390 n.a. 9.95 15.76 1,078
1983 0.5420 n.a. 10.07 15.10 1,052
1984 0.5550 0.5151 9.33 15.40 1,100
1985 0.5320 0.5011 10.13 15.12 1,108
1986 0.5390 0.4997 10.00 14.30 1,119
1987 0.5310 0.4950 10.22 13.97 ;.n.a.
1988 0.5301 0.4897 10.37 14.01 n.a.
n.ea. = not available
Notes: The Table includes two alternative Gini coefficients for the family income
distribution. To calculate the one in column 1, we ranked households by total
household income. To calculate the second one (column 2), wve ranked households by
the per capita income (Riveros and Weber 1987). However similar conclusions are
reached on analyzing both coefficients.
Sources: Columns 1-4: Riveros and Weber (1988); column 5: World Bank (various
years).
contrast, the decline seen in the Gini coefficients after 1984 is most
likely related to the improved performace of employment and wages
under the export-led growth strategy.
The existence of fiscal social expenditures in the fonm of direct
monetary subsidies to the poor makes estimates of Gini coefficients
based on only labor incomes subject to debate because they may be
ChAie 197
downward biased. However, in 1976-79 the need to control inflation
and to tighten fiscal expenditures caused a severe decline in social
outlays, which apparently increased poverty.-After 1980, the increase
in social outlays (table 5.7, column 4) is associated with the
privatization of the social security system and -the emergency
employment programs, and does not necessarily, mean that more
resources were devoted to deal with poverty. Declining social outlavs
after 1985 are associated with fiscal restraint and improvement in
some labor market results, particularly the decline in open
unemployment. Although social outlays may have declined in the
1980s, there was an improvement in targeting the poorest groups,
which was facilitated by the social sector reforms implemented during
the early 1980s. In general, however, problems of access to and
financing of the health and education systems are still of paramount
importance.
A Model for Adjustment Policies and Labor iMarket
Response
This section discusses and presents estimates of a model aimed at
analyzing observed wages and unemployment in the presence of labor
market segmentation. In this model, segmentation is defined on the
basis of protected and unprotected sectors with regard to the coverage
of typical labor market regulations. In considering that protected
sectors usually consist of large urban firms and the public sector, the
protected(unprotected breakdown overlaps significantly with the
traditional formal/informal dichotomy. Given that the enforcement
(and enforceability) of certain regulaTions may prompt asymmetrical
wage effects across sectors, this approach to segmentation is appealing
from an analytical viewpoint.
The existence of a growing protected/unprotected wage gap during
a period of substantial macro adjustment may explain persistent
unemployment. The existence of queuing unemployment-informal
sector workers queuing for formal sector jobs-is likely to increase
during adjustment if the relative degree of labor protection in the
formal sector increases. For instance, as suggested by the increase in
minimum wages relative to the wages-of unskilled-informal labor
during the reform period, the relatively larger protection awarded to
198 LuisA. Riveros
formal sector workers may have resulted not only in more
unemployment, but also in inequitable effects in terms of wages in the
unprotected sector.
In a neoclassical labor market, one would expect flexible wages in
the face of aggTegate expenditure-reducing and expenditure-switching
policies. This condition would imply declining wages in terms of the
price of tradables that will, in turn, produce labor shifts away from the
production of nontradable to tradables. Rigid wages in a neoclassical
labor market will result in a lack of mobility, thus leading to standard
policies aimed at removing wage distortions, but will not necessarily
create systematic changes in wage differentials azross certain labor
force groups.
In a segmented labor market, adjustment policies would exert an
inequitable effect in terms of the observed protected/unprotected wage
gap. In particular, nominal wages in the protected sector will be less
responsive to a change in tradable prices, thus making a nominal
devaluation less effective in achieving a real devaluation. This will
hinder interindustrial labor mobility and increase total unemployment.
In addition, the deterioration in income distribution in terms of wage
gaps will affect the sustainability of adjustment programs. Thus;
persistent unemployment, deteriorating income of the poorest segment
of the labor market, a decline in the production of nontradables (due
to a drop in relative prices), lower employment in protected activities,
and implementation of repeated ineffective nominal devaluations are
probable outcomes of a segrnented market.
A Theoretical Model
The urban labor market is segmented into a protected (formal)
sector characterized by government and union intervention in wage
setting and by binding minimum wages, and an unprotected
(informal) sector, which is basically a neoclassical labor market. The
formal sector produces both tradable and nontradable goods using
skilled and unskilled labor, while the informal sector produces
nontradables with only unskilled labor. (This model follows the basic
lines presented in Lopez and Riveros (1989, 1990). Assumptions
about the informal sector satisfy two important characteristics of this
Chile 199
sector in developing countries (PREALC 1987): it is a low
productivity sector mainly concentrated in the services sector.)
The formal market for skilled labor determines a notional
equilibrium wage (ws*); the actual equilibrium wage (ws) includes a
distortionary factor fl associated with government and union
intervention. This wage-setting function is compatible with the case of
oligopolistic unions that maximize a utility function that depends on.
relative wages and membership (Lopez & Riveros 1990). Thus, the
actual nominal equilibrium wage is written as:
ws--Ows (1)
where e > 1 (a more general specification of this relationship is
explored in Lopez and Riveros 1989). The minimum wage (MW) is
binding for unskilled labor in the formal market and we assume some
degree of substitution of skilled for unskilled labor. Thus, the formal
sector demand for skilled labor (Lds) depends on minimum wages as
well as on output prices and wages. The function is:
Lds = Lds(ws,6,MW,PT,PN,K) +Lg (2)
where PT and PN are, respectively, prices of tradables and
nontradables, K is capital stock, and Lg is public sector employment.
This function is a homogeneous degree one in prices and wages.
The labor supply of skilled workers (Lss) can be written as shown
in equation (3):
Lss = Lss(ws,M,CPI,N) (3)
where CPI is the consumer pnrce index. (in tum, an average of PT and
PN), and N is the working age population.
The formal sector demand for unskilled labor (Ldu) is:
Lduf = Lduf(MW,ws,PT,PN,K) (4)
200 Luis A Riveros
which depends on ws because of the substitution possibilities between
skilled and unskilled labor.
The total supply of unskilled labor (Lsu) in the economy is:
Lsu =Lsu(MW,wu,CPIN,K) (5)
where wu is the equilibrium wage in the informal sector. Finally, the
demand for labor in the informal sector is:
Ldu = Ldu(wu,PN,MW,K) (6)
Given MW and Lsu, employment of unskilled labor in the formal
sector is determined by equation (4), thus leaving an effective supply
to the informal sector that, in combination with Ldu, determines wu.
This type of equilibrium follows the concept introduced by Harberger
(1971). Given that there will be persons with supply price above wu,
but below the (given) MW, there will exist queuing (quasi-voluntary)
unemployment.
For empirical purposes, thz equilibrium form of the system will be
considered. Thus, the system is reduced to two equilibrium waige
equations. This will permit concentrating the analysis on the effect of
policies on the formal-informal wage gap, thereby circumventing
estimation of the underlying strdctural demand and supply functions,
for which employment and labor force data are scarcer Equilibrium
in b oth the skilled (formal) market and the unskilled (informal)
market yields the following expressions:
ws.=wS(MW,PT,PN,K,LgN) m
wu = wu(MW,ws,PT,PN,K,N) (8)
Due to the homogeneity properties of the underlying demand and
supply functions, equations (7) and (8) are homogeneous degree zero
in prices and wages. Hence, for estimating purposes both equations
wil be written in terms of MW.
Chile 201
Unemployment in this model results from two sources: labor
market distortions (that is, wages above notional equilibrium in the
formal skilled market and queuing unemployment in the informal
market) and cyclical fluctuations in the economic activity (see Riveros
and Paredes forthcoming). Hence, a general formulation of an
unemployment equation may allow us to derive empirically both
components based on aggregate data. In equation (9) both structural
and cyclical factors are used to explain aggregate observed
unemployment. Give n their association with structural
unemployment-in turn due to the presence of labor market
distortions-the ratio MW/Wu and the value of the distortionary factor
C are included. The growth trend of the labor force (LFT) is also
included as a factor associated with the structural unemployment in
the economy (this variable is obtained by fitting actual labor force
data to a uime trend). Among the cyclical factors, we consider
unexpected changes in the following variables: GDP (Y), terms of
trade (TOT), and the labor force (LFS). Unexpected changes are
empirically proxied through the difference between observed values
and the fitted values obtaied from a regression of the respective.
variables against a time trend. Thus, the 'following unemployment
equation was estimated:
U =U(MW/Wu,0,LFT,Y,TQT,LFS,) (9)
Using equation (9) we estimate cyclical unemployment as the
difference between U (total unemployment) and US (structural
unemployment). US is calculated as the sum of the shift coefficient
and the parameters associated with MW/Wu, 0, and LFI obtained from
equation (9), multiplied by the values of the corresponding variable.
Thus, the third equation in the system corresponds to cyclical
unemployment (UC), which is:
UC= UCY,PN,PT,ws,wu4Lg, K,N) (10)
Equation (10) results from specifying equilibrium wages in both the
skilled and unskilled markets, and allowing for the presence of
202 LuisA Riveros
unemployment. This equation is homogeneous degree zero in income,
prices, and wages.
Prices of nontradables are endogenously determined. However, to
allow a better focus on the labor market issue, we do not include an
equation for PN, but we will account for its endogeneity in estimating
the structural system (Lopez and Riveros 1989 present a model in
which prices of nontradables are simultaneously estimated) PT is
determined using the small-country assumption, and is.thus equal to
the nominal exchange rate multiplied by the world price of tradables.
The model finishes with an investment equation that permits
connecting the short run and the long run. Investment responds to a
partial adjustment to a desired capital stock level, while the optimal
capital stock depends on the interest rate, wages, prices, and growth
(Ym> Hence, the following ivvestment function is estimated:
I = I(i,Pt,Pn,w,ws,1MW,Y) (1)
Given the price and income homogeneity properties of the model,
we arbitrarily chose to normalize by the minimum wage.12 Equations
(7), (8), (10), and (11) were estimated in rate of changes through
3SLS. The results are presented in table 5.10 below. The appendix
defines the variables.
Empirical Results
As a preliminary step, and to obtain the parameters to compute the
cyclical unemployment rate, we empirically estimated equation (9).
The 2SLS estimates presented below (table 5.10) indicate that the
distortionary factor MW/Wu is statistically significant to explain
observed unemployment. This result suggests the queuing
unemployment is an . important component of the structural
unemployment and, in turn, of total open unemployment. Labor force
and terms of trade are also significant explanatory variables.
12. The demand and supply functions are homogeneous degree zero in prices.
wages, and income. The wage equations are homogeneous degree one in prices. We
also assume the investment and the unemployment functions are homogeneous degree
zero in prices and income.
Chile 203
Table 5.10 Open Unemployment Equation (Dependent Variable:
Total Unemployment)
Constant B MWWu lIFT Y TOT L- S
870.3 4.57 5.55 0.82 -0.16 0.58 0.09
(1.40) (0.92) (1t98) (2.17) (-0.85) (1.68) (1.99)
R2 =.94 F=27.7 DW_ 223
Notes: The method is 2SLS. Instruments: government expenditures, working age
population, domestic credit and lagged values, endogenous variables: MW/Wu and
Y*. 6 was proxied by the monetary value of nonwage labor costs. LFI is the fitted
value of a labor force series regressed against time; LFS is the "shock" defined as the
diffcrence between the observed labor force and LFIT Y* is the output shock, which
was obtained similarly to LFS. TOT is the stucturl trend in terms of trade, which is a
fitted value against a time trend.
Using these econometic results, we decompose total unemploy-
ment into its structural and cyclical components- The results are
presented in figure 51. An interesting feature is that structural
unemployment has been relatively high. After 1975, structural
unemployment increased, but then fluctuated less. Cyclical
unemployment was historically low, and at times negative, indicating
excess vacancies probably produced by skill mismatch. The relative
importance of cyclical unemployment increased in the 1980s.
The econometric results for the system of four equations indicate.
satisfactory overall fits and right sign of the key coefficients (table
5.11). This suggests that a segmented labor market is an appropriate
way to analyze the statistical information concerning wages,
unemployment, and the role of macro policies.
In the case of the skilled wages equation, most parameters are
significant using a 90 percent confidence interval. The effect of the
distortionary factor 0 appears prominent, which suggests the potential
impact of exogenous intervention in raising effective market wages.
The effect of changes in the price of tradable and nontradable goods.
(PT and PN) indicates the positive response of formal sector nominal
wages to inflation. This contrasts with the negligible effect observed in
the case of informal sector wages. This is particularly important with
204 LucsA. Riveros
regard to PT, which reflects the direct effect of nominal devaluations.
This finding suggests the extent of relative wage rigidity in the formal
sector, which is at the root of an inequitable impact of exchange rate
policies in the presence of segmentation. This observed asymmetric
effect cannot be easily explained in the context of an integrated labor
market and suggests the need for appropriate coneective policies.
Income growth and the price of nontradables exert a. positive effect
on formal sector wages, possibly due to unions' ability to transfom
most of the increases in market output prices into wage gains. The
Figure 5.1 Components of Total Unemployment, 1963-85
22-
20-
16-
14-
- B
6 N:
4 - ~~~
2 - ~ ~ ~ ~ ~ -
0- ~ ~ ~ ~ ~ V
-2 -I
-4
1963 1965 1967 1969 1971 1973 1975 1977 1979 1981. 1983 1985
Year
- Tocal Unemployment Strucruxal Component of Total Unemployment
- - Cycical Component of Total Unemployment
Chile 205
Table 5.11 A Segmented Labor Market During Adjustment, 1960-85
(2SLS estimates)
ws = 6.70 + 0.99 (6) + 0.50 Yg + 0.53 PTg + 038 PNg -2.78 I1 -1.06 Lg
(1.66) (5.25) (1.34) (2.41) (1.41) (-1.15) (-3.72)
R2= 079 DW = 2.13
F 112
w=-11.4 + 2.00 Yg -0.94 PTg -0.004 PNg + 3 74 l -0.02 Lg -0.72 LCg
(-1.42) (3.23) (-0.59) (-0.01) (1.56) (-0.07) (-1.61)
R2 = o073 DW =1.79
F = 8.74
UC= 7.32 -17' Yg + 0.92 PTg + 0.52 PNg -079 I -0.51 Lg + 0.32 LCg + 0.67 wu
(1.39)(-3.38) (2.40) (1.81) (-1.52) (-1.68) (3.04) (1.99)
R2 OA7 DW =1.99
F =255
I = 10.4 + 0.27 1TIPN - 0.02 ws-1 + 022 wu-1 + 030 Mlg -0.34 MID -0.11 YS-1
(-134) (132) (-0.04) (0.06) - (353) (-3.68) (-239)
2- = 052 DW =1.82
F = 5.12
Notes:
Prices, wages, and income are defined in growth rates (relative to minimum wages).
Capital growth, nonwage labor costs, and money are expressed in real terms.
ws = wages of skilled workers Yg = aggregate expenditures
wu = wages of unskiled workers Kg = capital stock
(6) = real nonwage costs of labor Lg = public sector employment
PTg = price deflator of tradable goods Mlg = Ml
PNg = price deflator of nontradable MID = dummy (I = 1960-73)
goods Ig = investment
LCg = total labor costs [ws + (0)] UC = cyclical unemployment
ws-l = lagged (1 year) ws YS-1 = lagged income shock
wa-i = lagged (I year) wu
Insrunents: public expenditures, working age population, lagged values.
growth in capital stock affects ws negatively, possibly because skilled
workers are substituted for new capital. However, the parameter in this
case is not statistically significant. Finally, the effect of public secior
employment on ws is negative, which indicates that the public sector's
share in the wage index is high, and that expansion in public sector
employment takes place only at the cost of lower wages in terms of- the
minimum wage.
206 LuisA. Riveros
The wu equation indicates that growth of prices and formal sector
wages is not relevant in explaining change in equilibrium wages in the
informal market The factor relatively more important is income
growth, which displays a highly significant elasticity. In comparing the
coefficients obtained in both the ws and wu equations with respect to
aggregate income, we conclude that contractionary policies would
affect unskilled labor in the informal sector relatively more than labor
in formal activities. These results are in line with those Lopez and
Riveros (1989) found in a comparative study covering four
countries.13
In the case of the cyclical unemployment equation, the results
indicate that expenditure growth as well as the. expansion in public
sector employment and the growth in the capital stock negatively
affect unemployment A positive impact derives from both tradable
prices and wages; moreover, when the regression was performed with
the ratio PT/N (the real exchanger mte), the parameter was significant
and positive, indicating that switching policies create more cyclical
unemployment likely due to both rigidities in moving labor across
sectors and higher wage rigidity in the formal sector. The total effect
of a devaluation on unemployment must also account for the effect on
skilled wages and their impact on total unemployment.
The investment equation reveals a significant impact of the growth
in Ml, a variable we included as a proxy for the real interest rate.'4
Furthermore, a dummy variable on this coefficient (MlD = 1 for the
period 1960-73) was significant and negative, implying that due to
higher government interventon, the role of the interest rate on
aggregate investment was probably very low or zero The other
variable that affects investment is the lagged income shocks, defined as
the difference between observed aggregate income and a fitted time
trend. The relative price of tradables to nontradables is positive and
13. In both wage equations we included the rate of growth of the labor force, but.
the results were not significat. We tested for structral differences in the equations
for the periods 1960-73 and 197-S198, but we did not find evidence in support of
the idea of a different distrbution of the data.
14. Due to control of the interest rate during most of the 1960s and early 1970s,
this variable is not reliable for measuring the opportunity cost of capitaL We also use
the investment deflator, but the result was not significant.
Chile 207
significant at 90 percent. However, wages do not appear to affect
investment growth or output prices when included separately.
A devaluation increases the formal/informal wage gap. A similar
result is associated with an output decline. The implication is that due
to the prevailing labor market structure, typical adjustment policies
exert a negative equity impact. This evidence also indicates that.
adjustment policies based on nominal devaluation increase cyclical
unemployment. Likewise, contractive policies negatively affect
investment and a real devaluation seems to encourage it, but labor
market variables do not play a direct role. However, if labor market
segmentation makes a nominal dtvaluation more ineffective in
reaching a real devaluation-fr the case of prices of nontradables
highly responsive to the increase in formal sector wages-the labor
market would play a direct role.15 In sum, the adoption of a
segmented labor market approach allows us to highlight the
distributive consequences of adjustment policy, which may also
contribute to more persistent unemployment. In general, this evidence
suggests that the political sustainability of typical adjustment programs;
is strongly related to the prevailing labor market structure and labor
market intervention policies-
Conclusion
This chapter has analyzed the key role of the labor market in the
adjustment of the Chilean economy in the 1980s. To explain the
performance of the labor market during the structural adjustment of
the 1980s, consideration of the deteriorating situation in terms of
unemployment and wages in the late 1970s was deemed necessary.
The impact on labor market variables of a series of structural reforms
aimed at reducing the state's economic size, deregulating product and
factor markets, and opening the economy to foreign trade were of
paramount importance in the 1970s. Moreover, the absence of labor
laws and the use of the exchange rate as a stabilization device gave
signals that prompted lower employment growth, higher growth of
15. Lopez and Riveros (1989) measured this efect, and calculated the elasticity
displayed by PN with respect to prevailing labcr market distortions. In the case of
Chile it was found that the degree of ineffectiveness associated to a nominal
devaluation due to labor market distortions is small.
208 Luis A. Rieros
nontradables, decreasing savings, and increasing external
indebtedness.
After 1984. the Chilean economy underwent a major macro
adjustment, whose success was partly expedited by the deep reforms of
the 1970s. The postcrisis policy was characterized by the achievement
of significant real devaluations, further privatization, targeting of
social expenditures to the poor, and export promotion and financial
policies to deal with the external debt and to increase investmenL The
results in terms of the labor market were a dramatic decline in open
unemployment, a slight increase in real wages, and significant growth
of employment in tradables-
The characteristics of thre Chilean labor market provide support to a
model of labor market segmentation associated with the degree of
protection awarded to formal sector workers. Econometric analysis
indicates that expenditure-switching and expenditure-reduction
policies reduce real informal sector wages relative to fornal sector
wages thereby negatively affecting income distribution during periods
of adjustment. Relative wage rigidity in the formal sector and the
increase in the formal-informal wage gap hinder labor mobility and
make unemployment more persistent. This suggests that the
persistence of open unemployment is associated with the prevailing
labor market structure, which is also at the root of a slower adjustment
to macro policies.
Three important implications can be extracted from this case study.
First, macroeconomic adjustment should be accompanied by
deregulation of the labor market to minimize the adverse and
inequitable effects stemming from expenditure-reduction and
expenditure-switching policies. This essentially implies that wage
indexation and other policies aimed at protecting formal sector
incomes should be carefully considered. Second, the role of skills is
important in terms of the structure of the labor market and its
response to macroeconomic policies, which demonstrates. the
importance of including skill enhancement in structural adjustment
policies. This also implies that an increase in labor mobility may be
achieved not only through legal and institutional reforms, but also by
providing informal sector workers with skills- Third, high open
unemployment is strongly linked to queuing for formal sector jobs,
Chile 209
-which in turn derives from relatively high formal sector unskilled
wages. In periods of transition, open unemployment will be linked not
only to shifts in production, but also to higher queuing
unemployment due to a relatively larger wage distortion. This suggests
that policies aimed at dealing with unemployment must pay attention
to the regulatory framework existing in formal labor markets.
APPENDX-
EMPIRICAL DEFNTION OF VARIABLES
Wages of Skilled Workers (Ws):
Index computed from the labor force surveys of the University of
Chile, which considers blue and white collar workers with more than
eight years of schooling-
Wages of Unskilled Workers (Wu):
Tndex- computed from the labor force surveys of the University of
Chile, which considers self-employed workers with less than eight
years of schooling.
Nonwage Labor Cost (0):
Fringe benefits, social security contnbutions, regular bonuses, and
vacation periods expressed as a proportion of wages (source: labor
law and Price-Waterhouse: Doing Business in Chile, 1979, 1983).
Price of Tradabes (PT):
Price deflator obtained from national accounts (central bank) for
agriculture, manufacturing, and mining
Price of Nontradables (PN):
Price deflator obtained from national- accounts (central bank) for
services and constructionL
Aggregate Expenditures (Y):
GDP at market prices (national accounts).
Investnent (I):
Fixed gross domestic investment (national accounts).
210
Chilc 211
Unemployment (U):
Unemployment rate (number of unemployed divided by the total
labor force). Table 5.3. Unemployed people are those that declared
themselves to be involved in job search during the week of reference
for the surveys.
Labor Force (L):
Employed plus unemployed population 12 years and older. Table
5.3.
Terms of Trade (TOT):
Ratio of export prices to import prices.
212 Lads A. Riveros
References
Banco Central de Chile. 1987. Indicadores Economicos y Sociales
1960-85. Santiago.
- -. l__1990. Boletin mensual No. 748. Santiago. June
Budnevich, C. and fthers. 1986. "Trimestralizacion de las series
nacionales de Emples." Technical Note, Estudios de
Economica, University of Chile, 13(1):155-165.
Castaneda, T. 1983. Evolucion del Empleo y Desernpleo y el Impacto
de Cambios Demograficos Sobre la Tasa de Desempleo en
Chile: 1960-1983. Documento Serie Investigacion No. 6et.
Santiago: UniversnLy of Chile, Department of Economics.
Corbo, V. 1985a. "Reforms and Macroeconomic Adjustment in
Chile During 1974-1984." World Development 13(8).
. 1985b. "The Role of the Real Exchange Rate in
Macro-Economic Adjustment: The Case of Chile 1973-82."
Discussion Paper DRD145. Washington, D.C.: World Bank.
-_______ .1986. "Problems, Development Theory and
Strategies of Latin America." Discussion Paper DRD190.
Washington, D.C.: World Bank.
Corbo, V., and P. Meller. 1984. "Trade and Employment in Chile in
the 60s." American Economic Review 69(2).
Corbo, V., and F. Sturzenegger. 1988. "Stylized Facts of the
Macroeconomic Adjustment in the Indebted Countries."
Washington, D.C.: World Bank. Processed.
Cortazar, R. 1983. "Wages in the Short Run: Chile 1964-1981."'
Notas Tecnicas No. 56. Santiago: Corprociao Investigaciones
Economics Pro Latinoamerica (CIEPLAN).
Cortazar, R., and J. Marshall. 1980. Indice de Precios at Consumidor
en Chile 1970-1978. Santiago: Corprociao Investigaciones
Economics Pro Latinoamerica (CLEPLAN).
Cortes, H., and L. Sjaastad. 1981. "Proteccion y Empleo."
Cuadernos de Economia (54-55). Universidad Catolica de
Chile.
Chile 213
Cottani, J. 1988. "Exchange Rate Trends for Fifty-One Developing
Countries. 1960-1986." World Bank, Debt and
Macroeconomic Division. Washington, D.C.: World Bank.
Processed.
Edwards, S. 1985. "Economic Policy aid the Record of Economic
Growth in Cbile 1973-1982." In Gary M. Walton, ed., The
National Ecot nmic Policies in Chile, Contemporary Studies in
Economic and Financial Analysis, vol. 51. New York.
Edwards, S., and A. Edwards. 1987. Monetarism and Liberalization,
The Chilean Experiment. Cambridge, Massachusetts: Ballinger.
Hachette' D., and R. Luders. 1987. "Aspects of the Privatization
Process: The Case of Chile 1974-85." Washington, D.C.:
World Bank. Processed.
Harberger, A. 1971. "On Measuring the Opportunity Cost of
Labor." Intemational Labor Review 130.
IMF. Various years. International Financial Statistics. Washington,
D.C.
Larrain, F. 1988. "Public Sector Behavior in a Highly Indebted
Country: The Contrasting Chilean Experience 1970-19851"
Washington, D.C.: World Bank Processed.
Larrain, F., and P. Meller. 1990. "The Socialist-Populist Chilean
Experience 1970-73." Santiago. Processed.
Lopez, R., and L_ Riveros. 1988. 'Wage Responsiveness and Labor
Market Disequilibrium." PPR Working Papers WPS85.
Washington, D.C.: World Bank.
1989. "MLacroeconomic Adjustment and the Labor
Market in Four Latin American Countries." In G. Standing,
ed., Towards Social Adjustment: Labor Market Concerns in
Structural Adjustnent. Geneva: ILO.
____-____._ 1990. "Do Labor Market Distortions Cause
Overvaluation and Rigidity of the Real Exchange Rate?" PRE
Working Papers WPS 485. Washington, D.C.: Wodd Bank.
Marshall, J., and P. Romaguera. 1981. La Evolucion del Empleo
Publico en. Chile 1970-78. Notas Tecnicas no. 26. Santiago:
2)4 Luis A. Riveros
Corprociao Investigaciones Economics Pro Latinoamerica
(CFPLAN).
Meller, P. 1984. "La Evolucion del Empleo y Desempleo En Chile."
Santiago: Corprociao Investigaciones Economics. Pro
Latinoamerica (CIEPLAN). Processed.
Pakdam, M., and L. Riveros. 1987. "The Causal Role of Minimum
Wages in Six Latin American Labor Markets." DRD270.
-Washington, D.C.: World Bankl
Paredes, R. 1987. "Stylized Facts on Adjustment and Labor Market
Trends in Chile."' Santiago: University of Chile. Processed.
PREALC. 1987. Modelos de Empleo y Politica Economica. Santiago:
International Labor Office.
Riveros, L. 1985a. "Desempleo, Distribucion del Ingreso y Politica
Social." Estudios Publicos 20.
. 1985b. 'Una Revision de la Literatura sobre el
Mercado laboral Chileno en.los Anos 1970." Estudios de
Economia 12(2):17-36.
._ .1986. "Labor Market Maladjustment in Chile:
Economic Reforms and Friction Among Sub-Markets."
Analisis Economico 1(1).
. 1988. "International Comparisons of Wage and
Non-Wage Costs of Labor," Washington, D.C.: World Bank.
Processed.
. 1990. "The Economic Return to Schooling in Chile.
An Analysis of its Long Term Fluctuations." Economics of
Education Review 9(2):111-121.
Riveros, L., and A. Diaz. 1987. "Desempleados y Tiempo de
Desocupacion: Gran Santiago 1970-1983." In C. Clavel, ed.,
Empleo Recopilacion de Articulos. Santiago: University of
Chile, School of Economics.
Riveros, L., and R. Paredes. 1988. '"Veasuring the Impact of
Minimum Wage Policies on the Economy." PPR Working-
Papers WPS101. Washington, D.C.: World Bank.
Chile 215
. 1989. "Political Transition and Labor Market
Reforms." Washington, D.C.: World Bank. Processed.
. 1990. "Factores Estructurales y Ciclicos y la
Composicion del Desempleo Abierto en Chile." Revista de
Analisis Economico 5(l), Instituto Latinoamericano de
Doctrina Economicos y Social (ILADES). Santiago.
Riveros, L., and C. Weber. 1987. "Structural Economic Reforms,
Financial Stress and Targeting the Poor: Trends in Income
Distribution and Social Expenditures in Chile 1974-1981."
Washington, D.C.: World Bank. Proessed.
Torres, C. 1982. Evolucion de la Politica Arancelaria: Periodo
1973-1981. Report No. 16. Santiago: Central Bank.
Walton, Gary M., ed. 1985. The National Economic Policies in Chil
Contemporary Studies in Economic and Financial Analysis,
VGI. 51. New York.
World Bank. 1990. "Chile's Country Economic Memorandum."
Washington, D.C.
. Various years. World Tables. Baltimore, Md.: Johns
Hopkins University Press.
Yanez, J. 1979. Una Correccion del Indice de Precios at Consumidor
Durante el Periodo 1971-I973. Serie Investigacion No. 34.
Santiago: University of Chile, Department of Economnics.
1987. "'Un Indice de Remuneraciones para el Gran
Santiago." Santiago: University of Chile, Department of
Economics. Processed.
COSTA RICA
T. H. Gindling
Albert Berry
The Costa Rican economy fared better during the 1980s than many
others faced with the need to adjust to serious balance of payments
and fiscal crises. Its performance during the last few years of the
decade suggests that it may be on a new sustainable growth path.
Wages have reattained their precisis levels and unemployment is once
again low. Although the crisis caused both wages and unemployment
to deteriorate rapidly, the recovery was relatively quick. The policies
introduced to produce structural adjustment were deliberately
"gradualist' in nature, and the public's reaction was tolerant and
relatively supportive. As the outcome qualifies Costa Rica's
experience as a success story, at least in a relative sense, an
examination of the economic, social, or political structure and setting
or the steps taken to deal with the crisis should be of interesL
Costa Rica's Precrisis Experience
This section reviews the relevant precrisis experience of Costa Rica
in the areas of macroeconomic performance and labor market
institutions and functioning.
The Macroeconomic Side
Costa Rica is in many respects unique in Central America and
unusual in Latin America as a whole. While its per capita GDP puts it
The authors would like to thank Susan Horton, Dipak Miazumdar, Juan Diego
Trejos, and participants in the Toronto and Warwick conferences on labor markets and
structural adjustment for helpful comments and Karen Tuner for excellent research
assistance. More detailed discussions of many aspects of the methodology used in
this paper are available in Gindling and Beny (1991).
217
218 Z H. Gindling and AlbtBerry
among the lower middle-income developing countries (World Bank
classification), many of. its social indicators compare favorably to
those of the upper middle-income bracket. Life expectancy is higher
and infant mortality and fertility rates lower than in most countries in
the upper ranges of the middle-income . developing countries
(Gindling and Berry 1991). The stability and competitiveness of its
political system is unique within the region. For most of this century,
the country's governments have been chosen in competitive elections,
and have typically had a high level of popular support and legitimacy
(Seligson and Muller 19387)-
Economists believe that the exceptional performance of social
indicators in Costa Rica is largely the result of social and economiic
policies begun in the early 1940s by the govenmment of Calderon-
Guardia, and fully institutionalized after a social democratic
"revolution" in 1948. Policies introduced or strengthened after 1948
include n m wage and other worker protection legislation, a
comprehensive social security system (which includes a government-
fimanced health care system as well as a .pension program),
nationalized banking and insurance systems, elimination of the army,
and a pattern of state intervention in the economy designed to
encourage a more equitable distribution of income.
From 1950 to 1980, GDP grew at an average annual mte of 6.2
percent (3.0 percent in per capita terms), placing Costa Rica among
the fastest growing -economies in Latin America- Much of this growth
was fueled by expansion of traditional primary product exports, chief
of which were coffee, bananas, meat, sugar, and cocoa. In the early
1960s these products accounted for over 20 percent of the GDP and
80 to 90 percent of all export earings (Cells and Lizano 1990) The
growth of export agriculture coincided with an inicrease in farm size.; a
decrease in the number of landowners and in the rural labor force, but
an increase in the share of workers who were landless; and substantial
migration from rural to urban areas, especially San Jose. By 1980,
Costa Rica's agricultural subsistence sector had virually disappeared,
and the share of the labor force in agriculture had fallen to below 30
percent Still, the economy remained vulnerable to shifts in the prices
of its principle exports (particularly coffee); an important focus of
concern and policy.
Costa Rica 219
In the 1960s, in part because of this perceived vulnerability, Costa
Rica instituted a policy of import-substituting industrialization. The
Central American Common Market (CACM), established in 1962,
enlarged the market for Costa Rica's import-substituting products by
lowering or eliminating barriers to trade between the. Central American
countries and instituting a common external tariff on imported
manufactured products. The import substitution undertaken at the
regional level pro;vided an engine for growth and industrialization in
Costa Rica (Bourguignon 1986). As the most developed member of
the CACM, Costa Rica was well placed to achieve a significant surplus
in manufactured goods trade within the CACM. Manufacturing output
rose from 12 percent of GDP in 1960 to 22 percent in 1979, due in
part to the increase in manufactured exports to other CACM countries.
Costa Rica's mnin exports within the CACM were processed foods,
domestic appliances, texfiles and clothing, and other metal products.
However, the growth of manufacturing within the CACM produced
its own problems. For one ihing, policies used to encourage that
growth (low import duties on capital equipment, an overvalued
exchange rate, low interest rates on loans from the nationalized
banking system, and so on) also encouraged capital intensity in the
sector, so that manufacturing employment grew at a much slower rate
than manufacturing output Second, the health of Costa Rican
manufacturing depended on the economic and political conditions in
the CACM. As the 1969 'football war" between El Salvador and
Honduras showed, the foundations of the political cooperation
necessary to maintain the CACM were weak. Third, the new industries'
were heavily dependent on imported inputs, so the contribution to net
exports was less than it appeared and ithe import-substituting
manufacturing sector was quite vulnerable to swings in international
prices, just as the exporting agriculturl sector had always been. Each
dollar of manufactring output required an estimated 600 (Cespedes
and others 1985) to 80e (Gonzalez-Vega 1984) worth of imported
inputs.
A looming structural problem lay in the rapid growth of the public
sector (Gonzalez-Vega 1984), a growth whose roots lay in the social
welfare state established in the late 1940s. When, as part of the import-
substituting industrialization strategy, the state became increasingly
Table 6.1 Selected Macroeconomic Indicators, 1970-89
Eyporu of Resource
GDOP 8n/loron :aod Inmoris of balance s
(millions Crown, rare: Growah rate: Growt, rate: rate" and' ExporLs joals and Inymor o %f GDP Relative
Y ef 16 GDP agrkudebre manmfactwur TermU 4/ 1% chango norror as % narifacnar as % of (currnl pr- ,rfe%
Year ceforacso ) (pireenh: (pgrcent) fpercenlJ rade In CPI) Absorpaton senlnc a/CDP servcfes CDP prices) UpW
1970 5,573.5 7.5 n.a. n.a. 93.5 4,7 5,877.3 1,904.2 34.2 2,208.0 39.6 -6.77 0.83
1971 5,951.3 6.8 n.a n.m. n.a. 3,1 6,268,5 2,059,3 34.6 2,376,5 46.5 A.a. n.a.
1972 6,438.0 8.2 m3.m. n.a. n.M. 4.6 6,412.3 2,410.1 37.4 2,385.0 37.0 n.a n.a.
1973 6,934.3 7.7 n.A, n.a. n a. 19.9 6,880,8 2,586,4 37.3 2,532.9 36.5 n.a. n.a.
1974 7,318,8 5,5 n.n. n l. nna. 30.6 7,315.7 2,774.2 37.9 2,771.1 37.9 .14.98 0.87
1975 7,472.5 2.1 3.0 3.2 73.7 12,4 7,321.1 2,, 9,5 36.4 2,568.1 34.4 -8.49 0.83
1976 7,885.1 5.5 5.3 5.6 86,2 3.5 8,061,5 2,866.9 36.6 2,983.6 37.8 -6.00 0.79
1977 8,586.9 8.9 2,2 12.0 102.2 4.2 9,217,9 3,100.7 36,1 3,731.7 43,5 -5.47 0.85
1978 9,125.1 6.1 6.4 7.9 92.0 6.0 9,727.9 3,408.9 37.4 4,011.7 44.0 *7.85 0.75
1979 9,575.8 4.8 0.5 2.6 84.9 9.2 10,184.5 3,520.0 36.8 4,128.7 43.1 -10.27 0.73
1980 9,647.8 0.7 *05 0.8 85.1 18.1 10,267.7 3.367.1 34.9 3,987.0 41,3 -10.34 0.76
1981 9,429.6 -2.3 5.0 *0,5 70.5 37.7 8,625.0 3,741.5 39.7 2,936.9 31 1 .4.91 1.9
1982 8,742.6 -7.6 -4.8 -12.1 72,7 90.1 7,610.2 3,537.3 40.5 2,404.9 27.5 0.41 1.10
1983 8,992.9 2,8 3.9 1.8 78,2 32.6 8,287.1 3,490.9 38.8 2,785.1 31.0 -0.75 0.93
1984 9,714.5 7.7 9.6 9.9 80.3 12.0 8,912.3 3,884.5 40.0 3,082.3 31.7 0.41 0.86
1985 9,784.6 0.7 .5.7 2.0 82.9 15.1 9,322.1 3,729.4 38.1 3,266.9 33.4 -1.75 0.81
1986 10,326.3 5.5 4.8 7.3 102.2 11.8 10,303.0 3,864.7 35.5 3,841.4 35.3 -0.85 0.84
1987 10,885.3 5.4 4.0 5.S . 93.3 16.9 11,100.0 4,546.5 40.6 4,761.2 42.5 -4.48 0.75
1988 11,204.4 2,9 5.4 2.1 92.5 20.8 10,992.4 - 4,987.5 43.2 4,775.5 41.4 -1.19 0.77
1989 11,540,5 3,0 n.a. n.a. n,a. 16,5 n.a. .n., r.m. n.0. n.a. n.a. S.D.
n.a. = not available
a. Percentage change In the consumer price Index or low- and medium-lncome familics In San JosE, calculated on the basis ot the annual averagc price Index.
b. The relative price of exports to the GDP, as derived from compasrison of the current price and constant (1966) prIce series for these variables Itn th atlional accounts.
Sources: ECLA (vatlous years); World Bank (1977, 1988), The source.of the data cited In all these publications Is the Central Bank ofCosta Rica (BCCR).
Costa Rica 221
involved in the production of goods, public employment increased
rapidly. The government was under political pressure to employ the
increasing flow of educated workers into the public sector at relatively
high wages (Gonzalez-Vega 1984). Along with increased government
spending went rising fiscal deficits, which were financed increasingly
by borrowing. on the international credit markets.
While Costa Rica's rapid growth during ihe 1960s and 1970s was
achieved with reasonably typical gross domestic investment rates and
marginal output capital ratios, the country's low rate of national
savings and high share of investment financed with foreign savings
was atypical, at least compared to other Latin American countries. At
the beginning of the 1970s, 40 percent of gross investment was
fmanced, by foreign savings, and by 1977-80, 46-percent was. The
ratio exceeded 50 percent in 1974 and 1975.
The 1974 oil price hike worsened Costa Rica's terms of trade
enough to threaten a recession. To offset this contracionary impulse
the government pursued expansionary monetary and fiscal policies,
financing the resulting fiscal deficits by borrowing on - the
intemational capital markets. Between 1975 and 1977 a sharp,
fortuitous rise in the price of coffee (by over 100 percent) pushed the
overall terms of trade up by 40 percent, raised export earnings, and
increased govemment revenues (see tables 6.1-6.2). This coffee boom
appeared ex post to "justify" the government's expansionary policies
in 1974-75 and also invigorated private spending, pushing absorption
up by 24 percent between 1975 and 1977 (table 6.1). Inports
skyrocketed during 1975-77, then remained high when the-terms of
trade returned to more normal levels. The result by 1979-80 was a
current account deficit exceeding 50 percent of exports of goods-and
services, up from an average of 33 percent during 1957-69. The
government again borrowed extensi'vely in the- intemational capital
market, encouraged by the easy availability of credit and the low real
interest rates (Gonzalez-Vega 1984, p. 315), and was soon facing a
major debt crisis.
For purposes of macroeconomic analysis, it is useful to divide the
period beginning in 1978 into three phases: expanding crisis writh
ineffective expansionary policy (1978-82), stabilization (1982-83),
and recovery with structural adjustment (1983 onward).
Table 6.2 Selected IWdicators of Internal and External Finance, 1975-88
AbsoDltefigutres (U5$ Aillions) As a percenlagje of the CDP
Exchoarige ratc'
* (colottes/US$) Current Nc; Net (it : Debt
accouills Balaonce of capit foreign O overam eJmi Fiscal Afoney
Year Yearly average Judy balance payments JpJglh' Investment USald spendinig deficit - sipply Public Private
1975 8.6 8.6 -217.6 20.9 n.a. 69.0 n.e. 13.6 .. -0.21 30.3 421.3
1976 8.6 8.6 -201.5 59.1 n. . 63,3 n.a. 13,7 -0,66 38.4 535.9
1977 8.6 B.6 -225.0 107.9 4.6 63.2 n.s. 13.3 0.05 39.4 725.4
1978 8.6 8,6 -387.7 -26.8 33.1 47.1 8.2 14.2 -0,82 43.6 949.9
1979 8.6 8.6 -601.6 -1004. -89.7 46.0 17.2 14.9 -2.50 41.4 1,309.1
3980 9,3 8,6 -658.6 -172.8 110.3 48,1 15.9 15.3 -3.10 41.6 1,697.4 411.9
6 1981 21.2 14.5 -420.3 -l I 1.S -277.1 66.2 17.4 15.2 -1,60 56.5 2,206.4 371.5
1982 39.8 655 -246,0 139.9 -2305 26,3 57.0 15.5 -1.10 42.0 2,429.2 381.2
1983 41.6 45.3 -309.7 -58.1 248.5 55.1 216.5 16,2 0.36 .43.4 3,226.2 348.2
1984 44.4 43.6 -263.9 97.7 -126.4 51.9 186.2 16.3 0.26 40.3 3,289.5 316.6
1985 50.5 50.9 -365,9 99.8 n.a. 63,2 220.1 15.2 0.94 38.4 3,579.5 301.6
1986 56.1 56.1 -187.8 54.8 n.a. 57,3 154.5 15,9 0.47 37.2 3,575.0 306.4
1987 62.8 62,3 -399.5 -76.7 n.,. 89.5 180,1 15,3 0.36 37,6 3,623.0 345.1
1988 75,P 75,9 105,7 n.a. n.a. n.a. n.a.
a. Free-rate for July. The yearly average corresponds to an averago over Ihe course of the year of the effective exchange rates between
current Inpaymenis, and oulpaynicnis In (ihe balance of payments.
b. Using a technique developed by Morgan Guarantee Trust. A negative entry mcans a net Inflow of capital, a positive cntry means a
net outflow of ieapltal (capital flight).
Sources: World Bank (1988, varlous years); Glower (1986); Sanford (1989); Central Bank of Costa Rica data.
Costa Rica 223
Between 1978 and 1981 Costa Rica's terms of trade fell by a third
to a little under the precoffee boom lows of 1974-75 as the price of
coffee fell by 61 percent between 1977 and 1981 and the 1979 oil
embargo increased the price of imports (see table 6.1). Political
turmoil and economic. recession cut demand for Costa Rica's
previously substantial exports to CACM countries. Despite s ng
export markets (especially for industrial products) and stagnant
banana exports, the export quantum continued upward until 1981, but
the purchasing power of exports fell by 20 percent. Balance. of
payments deficits became increasingly difficult to finance as
international interest rates rose, international banks curtailed credit
availability, and private direct foreign investment declined
Largely in response to popular pressure, the Carazo government
(1978-82) attempted to stimulate the failing economy using the same
recipe applied successfully in 1975-77: a high level of aggregate
demand and relatively stable prices. This attempt probably contributed
to the severity and, length of the crisis (Gonzalez-Vega 1984)_
Government spending was increased, especially ori heavy
dinrastructure, thereby increasing the fiscal deficit and public sector
borrowing on the international capitaI markets. limport duties on
consumption items were cut further,. which increased imports and the
balance of payments deficit, but held the consumer price index to an
annual average increase of only 11 percent during 1977-80. Refusal
to devalue the colon in the face of a looming balance of payments
crisis encouraged capital flight and speculative importing (table 6.2).
The public foreign debt, previously growing but manageable,
ballooned out of control, more than doubling between 1978 and 1981
(Carrillo 1988). The failure to reach an agreement with the IMF in
early 1980 increased capital flight, and prefaced a reluctant but
massive devaluation (from a rate of 8.6 colones per dollar to 21.8)
near the end of 1980, which in turn led to rapidly accelerating
inflation. With a devaluing currency, falling export earnings, and little
or no access to foreign public or private capital without an IF
agreement, a moratorium was declared on servicing the public sector's
external debt (Rodriguez 1987).
Between 1979 and 1982, the GDP fell by over 9 percent Industrial
production, which fell by 12 percent in 1982, was hit especially hard
224 T. H. Gindling andAlbertBerry
by the breakdown of the CACM and by the rise in imported input
prices when the devaluation finally came. As inflation increased from
9 percent in 1979 to over 90 percent in 1982, real wages fell by nearly
40 percent, the unemployment rate rose from 5 to 8 percent of the
national labor force, and the underemployment rate rose from 3 to 7
percent
The economy hit bottom ini mid-1982; 1983 saw a mile upturn, a
sharply reduced inflation rate, and a significant recovery of real
earnings. The reversal occurred as the new government of Partido
Liberacion Nacional's (PLN) Alberto Monge took office. Monge and
the PL1N, with strong ties to the unions and past social legislation, were
in a better position to command union and popular confidence and to
call for short-term sacrifices than was the business-supported Carazo.
In summer 1982, Monge instituted a "100-day stabiliztion plan"
that included an appreciation of the colon, together with controls on
the outflow of capital and unification of the official and free-market
exchange rates;' income, sales, and consumption tax increases of 60 to
170 percent; decreases in subsidies (by 50 to 80 percent in the case of
public transportation) and increases in the prices of public utilities
(that of fuel increased by 80 percent, of electricity by 80 to 90
percent); a public sector wage freeze; and a credit restraint/contraction
of the money supply (Latin American and Caribbean Contemporary
Record 1983; Nelson 1989). In December, 1982 a letter of intent was
signed with the IMF. Large infusions- of American aid, which rose
from virtually zero in 1979 to US$216 million in 1983, mitigated the
damping effects of the public sector's large foreign debt (Rivera
Urrutia and others 1986).
Despite the apparent harshness of the stabilization measures, there
was relatively little public protest. Seligson and Muller (1987) report
that optimism about future economic conditions was higher in 1982
than at any time during the previous seven years. Nelson (1989, p.
148) reports that by "mid-1982, all groups were paying dearly for
unplanned adjustment, and many were willing to acquiesce to any
1. Capital controls were accomplished by closing the exchange houses and
controUing all currency exchange within the nationalized banking system. AU
foreign exchange camings from exports or tourism had to go through the central
bank.
Costa Rica 2225
plausible.policy that signaled the resumption of government control
over the situation." Nelson. also notes that austerity policies were
biased in favor of the poorer segments of society. Businesses and
medium and large agricultural interests bore the largest share of the
substantial tax and utility price increases. According to the national
accounts, general government revenues as a share of GDP rose from
21.2 percent in 1982 to 263 percent in 1983. -
Terms of trade stabilized in 1982 and improved thereafter through
1986, then slipped back again. In 1988 they were at about the 1980
level. The price of coffee began to rise in 1984 and experienced a 47
percent increase in 1986, before falling 53 percent in 1987. Real
wages and earnings regained their 1979. levels between 1986 and
1988. GDP began to grow in 1983 and achieved rates of. at least 3
percent each year through 1989, except for a slump in 1985, caused
in part by decreased agricultural subsidies to basic grains with a
consequent drop in agricultural production, a temporary catoff of
U.S. aid, and a drop in banana exports (Standard Fruit abandoned
production on the Pacific Coast due to labor troubles, high export
taxes, and crop disease). Average growth between 1982 and 1989 was
4 perce-nt.
In 1984, under pressure from USAID, the IMF, and the World
Bank, Costa Rica initiated one of the most comprehensive structural
adjustment programs in Latin America (Rivera Urrutia and others
1986)- The program included the following:
* selling state-owned production enterprises (by 1988 only four
remained);
* a 1987 agreement with the IMF to limit the increase in public
sector salaries;
a new CACM tariff regime instituted in January 1986 to replace
the 1963 regime, whereby the. mean tariff was cut from 53 to
20 percent and the standard deviation from 62 to 21 percent
(according to a 1987 IMF stand-by agreement further tariff
unification is to continue until January 1991);
* a 1987 tax reform that lowered the maximum marginal income
tax rate to 25 percent and increased the threshold for
contributions, instituted a 30 percent flat corporate tax ratc, and
increased sales, property, social security, and stamp taxes;
226 T. H. Gindling and Albert Berry
a loosening of the monopoly of state-owned banks, and hence
of government control over the provision of credit, and a
reduction in subsidized credit (although the program for
subsidized credit to small farmers was expanded);
* a reduction of some agricultural subsidies, together with
elimination of all quantity restrictions on agricultural imports;
* a program to promote nontraditional exports to third markets
via creation of free trade zones, tax breaks, import duty
exemptions, and production and marketing extension services.
The free trade zones are designed to encourage "draw-back" or
maquila industries, which can import and export duty free and benefit
from tax breaks. Over 75 percent of the participating firms are in the
textile industry, and over 15 percent are in electronics (CENPRO
1986). According to the Central Bank of Costa Rica, value added in
draw-back industries now accounts for 12 percent of the value added
of nontraditional exports.
The administrations of both Monge and Arias emphasized that the
structural adjustment program should be implemented gradually in an
attempt to minimize disruption. If the results have matched the plans, a
gradual rather than a rapid shift in the structure of production toward
exports (in particular nontraditional exports) should have occurred.
After falling by 15 percent during 1980-83, the current dollar value
of exports (f.ob.) had by. 1989 climbed to 40 percent above the 1980
level, while the value of imports, just equal to exports in 1988, only
surpassed the 1980 level when they jumped by 20 percent in 1989.
(In 1980 the import/export ratio was 1.37.) Meanwhile, with interest
charges high, current account deficits and the public foreign debt
remained high. By the end of 1987, the stock of debt was roughly
equal to the GDP and interest payments amounted to 6 percent of
GDP and 18 percent of export earnings. This heavy debt overbang
was made tolerable by a continued high level of U.S.. aid and IMF
balance of payments support. The U.S. aid came with stringent
conditions' including Costa Rican support for the Reagan
administration's policy toward Nicaragua.
Recently, the debt burden has been eased somewhat. In 1990 Costa
Rica negotiated a buy-back of US$1.8 billion of its foreign debt This
buy-back was financed in part with a US$250 million "bridge" loan
Costa Rica 227
arranged and financed by central banks in the. United States, Taiwan,
Mexico, and Venezuela (Tico Times, May- 25, 1990).
In the crisis years of the early 1980s the (current price) gross
investment rate appears to have held up surprisingly well, never falling
below about 23 percent, although the fixed investment figures did.
show a more marked decline (from an unweighted average of 23.9
percent over 1977-80 to 19.5 percent over 1982484) and the
inventory change estimates may be invalid (table 6.3). The share of
investment financed by foreign savings fell from a local peak of over
50 perct nt during 1979-82 to under 40 percent by 1984-85. The
dramatic decrease in real investment (in constant 1966 prices) from a
peak of 28.5 percent of GDP in 1980 to a low of 14.6 percent- in 1982
was due to a sudden change in the relative price of investment goods,
not to a major decline in either the current price investment and
savings rates or to the national savings rate (which.in fact rose from
11.5 percent during 1975480 to 13.8 percent during 1981-85), but
when foreign savings fell and capital goods became much more costly,
maintaining a high real investment rate was not feasible given the
traditionally. very low national savings rate.
Labor Market Overview
From the early 1960s to the late 1970s, opportunities for the
average Costa Rican worker improved markedly. The share in
agricultural employment, the lowest paying sector, decreased steadily
from 49.1 percent in 1963 to 38 percent in 1973 and 30 percent in
1979, while the proportion of professionals and technicians rose
steadily, and the share in the high paying public sector increased from
13.3 percent in 1963 to 15.3 percent in 1973 and nearly 19 percent in
1980 (1963 census; 1973 census; 1979/80 household surveys). The
unemployment rate typically hovered around 5 to 6 percent. Inconrs
-and real wages increased by more than 60 percent between 1963 and-
1979. A temporary decline between 1972 and 1975 due to the
recession caused by the first oil embargo was more than offset by a 25
percent "coffee boom" increase between 1976 and 1979. - T!^
incidence of poverty fell. Improvements on all fronts came faster with
the coffee boom, which produced growth spurts in both agnculturi-
and manufacturing production (9.1 percent and 22 percent,
Table 6.3 Gross Savings and Investment as a Percentage of GDP, 1970-88
(constanit 1966 prices)
Fnreign
Invesineni In fixed capital Gross savings as
i Investment percentage Natianal savings
Gross Consiruc- Alachines & Changc in Gross In Jteti of gross Foreign
Year Invesmenal Total lion equipnent Inveniorkes Investment capiral savings savings Total Public Private
1970 20.23 19.31 9.23 10,11 0.91 20,53 19.46 37.8 7.8 12.7 n.a. a.
1971 23.52 21.08 n.a. n.a. 2.44 24.34 22,12 46.2 1i.2 13.1 2.7 10.4
1972 20.15 20.40 n.a. n.a. *0.25 22.03 21.91 .41.4 9.1 12.9 2.9 10.0
1973 22.20 20.54 n.a. na. 1.66 23.99 22.16 35.0 8.4 15.6 3.6 12.0
1974 22.84 21.36 9.64 1.72 1.48 26.74 24.02 62.4 16.6 10.1 5.4 4.7
1975 20,38- 20.66 9.70 10.96 .0.28 21.64 21,99. 51.3 11.1 10.5 4,7 5.8
1976 24.43 24.22 11.57 12.65 0.21 23.66 23.44 35.3 8A 15.3 4.3 11.0
1977 27.54 25.00 11.40 . 13.60 2.S4 24.27 22.36 30.3 7.4 16.9 4.2 12.7
1978 25,48 25.44 10.96 14.48 0,37 23.46 23.03 43.9 10.3 13.2 3.4 9.8
1979 26.87 27.9S 12.42 !5.53 -1.08 25.31 26.17 54.0 13.7 16.6 0.3 11.3
1980 28.54 25.13 12.56 12.56 3.41 26.57 23.90 55.8 14.9 11.7 -1.0 12.7
19U1 18.18 19.31 9.66 9,64 -t.13 29.03 24,06 52.3 15.2 13.8 1.2 12.6
1982 14.62 15.03 8.15 6.88 -0.41 24.69 20.32 49.9 12.3 12.4 -0.8 13.2
1983 18.91 15,82 8.28 7.54 3.09 24.71 17.99 44.0. 10.8 13.9 10.2 3.7
1984 19.42 18,48 9.88 9.29 0.95 22.61 20.0s 31.4 7.1 15.5 853 5.9
1985 20,78 19.33 9.50 9.83 1.45 25.42 19.32 29.5 7.5 15.8 n.2. n.a.
1986 25.-6 20.49 9.34 11.15 5.27 25,20 18.66 15.2 3.8 21.4 n.a. n.a.
1987 28.77 21.32 8.63 12,67 7.45 28.25 19.59 31.2 8.8 19.4 n.a. n.a.
19K8 25.69 19.S3 8.74 10.79 6.16 26.04 18,11 . na. 6.0 20.1 n.a. n.a.
n,a. = not available
Source: ECLA (vartous years).
Costa Rica 229
.respectively, between 1976 and 1979). Total employment increased at
an average annual rate of 4.7 percent per year and the labor. force at
4.0 percent, pushing the unemployment rate to a low of about 4.0
percent in 1979. Employment in the public sector increased rapidly
from some 16 percent of total employment in. 1976 to 18.3 percent.in
1978, or by an average of over 11.0 percent per year, before slowing.
in 1979.
The labor market began to show signs of a downturn in 1979.
Although open unemployment did not rise, underemployment moved
up sharply. The late 1980 devaluation was followed by a sharp
increase in inflation, a huge drop in real wages and earnings (about 35
percen: between 1980 and 1982), and an apparent 'additional
worker' effect as secondary family workers entered the labor force.
The proportion of the labor force not heads of households increased
from 50 to 51 percent during 1977-80 to 53 to 54 percent in 1981-
82, and the share of women appears to have risen somewhat more
rapidly than the average increase in other years. The number of new
entrants to the.labor force doubled between 1980 and 1981. During
1979-82, the influx of secondary family workers pushed the labor
force up at an average of 4.3 percent per 'year, while employment
grew at only 2.6 percent per year (table 6.4) so the unemployment
rate rose steadily to a peak of 9.4 percent in 1982.
While employment did not increase as fast as the labor force, that it
increased at all is surprising, given the contraction of both agricultural
and manufacturing output. An important. ameliorating factor as the
crisis set in was the 8.8 percent increase in public sector employment
(almost all of it in the central government) between 1979 and 1981,
which accounted for most of the employment increase in that interval,
and reflected the Carazo govemment's attempt to spend its way out of
the impending crisis and to keep the unemployment rate at
manageable levels (table 6.4). Salaried employment also increased in
small repair fims and for domestic servants (interview-with Juan
Diego Trejes). The increase in employment from 1979 to 1981 was
entirely composed of wage eamers (the number of self-employed
workers and owners actually decreased by 6.2 percent between 1970
and 1981).
Table 6.4 Labor Market Indicators, 1976-88
Perceniage of labor ofocc that wax.: Percentage of employment In: Real wagn Real earnings
Labor
force Employment Under- Ncmr Paid Paid
(seasonalty (seasonally employed hoas)wold Public Rural whtrArs All wuug.lis All
adjusted) adjusted) temployed (open) Female head secloD oreats Agrcutrlre anly wo,ers only kVtACn
Year (1) (2) (3) (4) (5) (6) (7) (8) 100) (11) (12) (1)
1976 664.7 635,3 4.42 9.43 23.8(2) 50.8(2) 15.9(2) 50.0 34.2 5.7 n.a. 1,061(2) n.m.
1977 697.3 662.6 5.07 10.30 23.6 51.0 16.7 54.2 34.4 6,4 n.a. 1,199 n.a.
1978 737.7 704.0 4.56 9.87 25.9 50.8 18,3 55.0 33.5 6.9 n.ma 1,314 n.m.
1979 748.9 718.3 4.09 15.50 25.7 50.8 1R.6 53.1 31.0 7.4 n.a. 1,401 n.a.
1 90 779.7 738.0 5.34 17.17 . 25,8 50.2 18.8 52.5 29.5 7.2 7.3 1,346 1,353
1981 819.3 751.6 8.27 20,30 26.9 53.0 18.6 55.1 31.9 6.3 6.3 1,141 1,161
1982 849.0 177.3 8,45 22.80 27.2 54.0 17.3 52.2 29.8 4.8 5.0 865 899
1983 844.3 776.0 8,09 17.70 25.6 52.1 18.8 50.9 30.3 5.7 5.8 1,030 1,051
1984(2) 853.1 790.7 8.00 16.83 26.8 51.3 19.7 52.2 30.4 5.9 6.3 1,162 1,223
1985(2) 901.9 846.6 6.14 15.46 26.0 52.7 18.8 49.9 31.2 6.9 7.1 1,230 1,286
1986(1) 962.2 909,8 5.45 16.63 28,0 53.1 20.1 50.8 26.9 7.4 7A4 1,339 1,375
1987(3) 992.7 941.5 5.35 8.82 28.3 55.0 15.9 46.1 28.1 8.1 8.2 1,433 1,491
1988(3) 3,021.0 970.0 5.00 8.91 29.2 53.8 17.1 46.2 28.1 7.6 7.7 1,414 1,427
Souirces anid ntet:hodology Except where indicated, data are averages across observations for March, July, and November. Numbers in
parenthcses indicate years or variablcs In which observalions were only available to us for one or two rather than. the usual three
monihs, The data are from the household surveys of cmploymenit and unemployment. For columns 1-8, 12, and 13, missing figurcs
for any of the three months were filled in on the basis of.the observed average scasonal diffcrences over 1977-83 (when data were
available for each of March, July, and November), as follows, and in cach case sctting July = 100.
Labor forcec Mar. 97.5 Nov, 107.0
Employment; Mar. 97.6 Nov. 108.4
Female share of labor force:- Mar. 98.1 Nov. 109.1
Nonhouschold head share of labor force: Mar. 102.9 Nov. 106.1
Public sector share of labor force: Mar. 98.9 Nov. 93.3
Rural sector share of employment; Mar. 101.6 Nov. 104.6
Agriculturc share of employment (July data not nvailable): Mar. 100.0 Nov. 113.3
Costa Rica 231
By the time of the devaluation, the Carazo administration seemed to
have given up its holding action against the recession, and public
sector employment fell between July 1981 and the end of 1982.
Growth in total employment between July 1981 and July 1982 was
driven by a 6 percent increase in rural employment as agricultural
production grew by almost 5 percent, while nonwage employment
grew faster than wage employment (4.4 -percent as opposed to 3.9
percent). Whereas the increase in public sector employment between
1979 and 1981 occurred mostly in the central govemment, the
decrease between July 1981 and July 1982 took place primarily in
autonomous and semi-autonomous enterprises (in transportation,
commerce, construction, and utilities).
About the time that Monge (June-July 1982) instituted his LOO-
day plan to stabilize and revive confidence in the economy, real public
sector salaries and real minimum wages rose because they were
indexed to the inflation rate of the past period. With inflation slowing
down, the indexing mechanism pushed nominal salaries up faster than
prices.- In addition, more frequent adjustment of these institutionally
set salaries worked in the same direction. The increased real wage
probably had a dampening effect on the participation of secondary
workers: the share of nonhousehold heads in the labor force dropped
to its precrisis levels. Another large increase in public sector
employment between July 1982 and July 1983 accouunted for more
than the total increase in employment The Monge government thus
mitigated the negative welfare effects of the stabilization package
through increases in public sector employment, public sector wages,
and the minmum wage helped by the sharp increase in U.S. aid in
1983.
From 1983 to 1987, labor incomes continued to rise. Real wages
and real earnings regained their 1980 levels by 1986. The proportion
of younger and older people in the labor force resumed its earlier
downward trend, while the proportions of women and of
nonhousehold heads, after dropping from their 1981-82 peaks,
resumed their previous upward trends (table 6.4). Unemployment fell
to the low levels of the late 1970s, as did the average duration of
unemployment spells. In 1985 rural employment experienced a
temporary decline, related to a 5.7 percent decrease in agricultural
232 T. H. Gindting andAlber Berry
production probably due to the elimination of subsidies to rice and
sorghum producers (a structural adjustment policy), and to the
withdrawal of Standard Fruit from the Pacific coast. Employment in
the public sector remained high through 1986 and beyond.
Labor Market Institutions in the Adjustment Process
Addressing the economic crisis in Costa Rica required a sufficient
reduction of aggregate demand to narrow the gap between absorption
and production, and a sufficient resource shift toward tradables to
rectify the balance of payments disequilibrium. The key components
of many stabilization plans are expenditure reduction and devaluation.
Since absorption must fall, incomes must probably also fall- The
objective is to limit that fall to the level essential to achieve m-e
adjustment. Inflexible labor costs may mean that expenditure
reduction wil result in unemployment of workers attached to the rigid
wage sector or in underemployment or misallocation of labor among
sectors; oiutcomes that might be avoided if real wages are flexible. For
output composition to change some labor shift between sectors may
be important, which may require a change in relative wages, a lack of
barriers to movement, or both.
Labor Market Insituzuions and Real Wages.
Many of the labor market institutions that one might ext ect to
protect workers from falling real wages are firmly implanted in Costa
Rica- Relatively well enforced minimum wages are generally
considered to be the key institution influencing real wages in the
prvate sector.2 Public sector wages, clearly import'ant given the
quantitative significance of public sector employment, are also likely
to be resistant to sharp declines.
In examining the impact of labor market institutions on short-run
stabilization, we are particularly interested in two periods: the
aftermath of the drastic devaluation of the colon (December 1980 to
mid-1982), and the period of government expenditure reduction, tax
increases, and appreciation of the colon (mid-1982 to 1984). The
2. Minimum wages are legislated for over 130 different industrial classifications,
each with up to 9 occupational categories. Over 500 separate mminmum wages can be
legisIated
Costa Rica 233
proximate determinant of real wage movements during these periods
was the rate of inflation together with mechanistic wage setting rules.
The 34 percent decline in real wages during 1980-82 occurred
because accelerating inflation eroded the protective effect of these
rules. Both minimum wages (a key to private sector wage setting) and
public sector salaries are indexed.to the change in price level since
their last adjustment. Under such an arrangement, real wages fall in
times of accelerating inflation, .stabilize below pre-inflation levels if
inflation stops accelerating but stays high, and rise when inflation
slows. The fiscal crisis experienced by the government probably
accounted for the upswing i real public sector salaries being
somewhat slower than in the private sector. An assessment of the role
of devaluation and exchange rate policy as determiants of real wages
is made difficult because their effects occur primarly through their
impacts on inflation and may easily be disguised by and confused
with the intermediating role of inflation and the mechanisms it sets in
motion (see Gindling and Berry 1991 for a more detailed discussion
of these issues).
The influence of unions in Costa Rica has traditionally been limited
mainly to the public sector, private sector unionization has been
significant only among banana workers. Union influence reached its
peak at the end of the 1970s, since when it has declined dramatically.
A turning point was a 72-day strike against Standard Fruits banana
operation on the Pacific Coast in 1984, which relied heavily on 48
sympathy strikes by other private sector unions, and the prestige of the
most powerful private sector unio-, leaders rested on a successful
conclusion. When Standard Fruit closed its operations and. abandoned
Costa Rica, the power of the remaining private sector unions declined
dramatically and the union movement divided (Donato and Rojas
1987). Meanwhile, the government's fiscaI problems overwhelmed the
public sector unions. Monge's mandate to "do something" created a
sense of purpose in the- government that precluded unified action
against it, while the public sector -union movement was by then
suffering from disunity and confusion spawned by the economic
crisis. With the 1987 accord with the IME, in which the Arias
administration agreed to avoid increases in the average public sector
wage, union influence waned further.
234 T. H. Gindling andAlbert Berry
In summary, the sharp decline in real wages between 1980 and
1982 and a direct assessment of the major labor market institutions
suggest that these institutions did not significantly limit downward
flexibility of real wages during the crisis. They may actually have
pushed wvages below the levels that might have resulted under many
other institutional arrangements. In a sense, real wages fell because
these institutions failed to perform their planned function. By: this
failure, they may have provided the government with an easy,
apparently neutral, mechanism whereby real wages could be pushed
down when the state -of the economy most required it.
Sectoral Employment Shifts
We distnguish here four important sectors. Importables are those
tradable goods for which the likely alternative sources are imports and
domestic production, the latter often protected from foreign
competition by policies of import-substituting industrialization. The
approximation used here includes all private sector production of
manufactured goods plus basic grai (rice, sorghum, maize, and
beans), all of which have been heavily protected?3 Exportables are
those tradables exported to non-CACM countries or consumed at
home, including agricultural products and services (with the exception
of basic gris) plus mining. Private nontradables include the
actvities of construction, basic services, commerce, and services (UIIC
two-digit classifications 40 to 96)O The public sector is defined here to
include the central and mumcipal governments and autonomous and
semi-autonomous enterprises (parastatal enterprises). Although
primarily composed of nontradable services such as public utilities,
education, insurance, banking, pensions, and medical services, it also
includes production of some import substitutes (such as cement and
fertilizers) and even the processing of some exports (sugar, coffee, and
3. A dilemma in the distinction between the export and import categories arises
from the country's exports of significant amounts of manufactured goods to other
CACM countries, although these.items would otherwise be, and since.the demise of.
the CACM have become, importables. The main items in a group we refer to here as
the CACM importables are processed foods, textiles, petrochemical products, and
electrical machinery. Unfortnmately, because by the late 1980s some processed foods
and textiles are nontraditional exports, treating CACM importables systematically as
a separate category is not possible.
Costa Rica 235
so on). The presence of these latter activities means that this is not a
purely nontradables category.
The total current dollar value of merchandise exports increased
with the coffee boom from 1975 to 1977, eased up during the rest of
the decade, fell sharply in 1982, and then grew back to a new high in
1989, 40 percent above the earlier 1981 peak (table 6.5). This growth
path of the export quantum was rather different, however (table 6.1),
with good growth through 1978 and very little (only 10 percent) from
then through 1986, followed by a 30 percent burst in the next two
years. By 1983485, the colon price of exports (abstracting from
export taxes and subsidies) was back down to its precrisis 1978-79
level and well below that of 1975-77, so the stagnation of export
quantum was not surprising given the absence of a maintained
increase in the relative price of exports.
In the late 1970s, current dollar exports in each of traditional
exports, exports to the CACMt, and other nontraditional exports grew
fairly rapidly. During the heart of the crisis, the big loss of export
revenues occurred in the CACM category; the downward trend
continued through the rest of the decade.. Rising revenues during
1983-87 came from both traditional exports, whose prices were now
moving up, and from nontraditional exports, which leapt from 166
million in 1983-84 to 372 million in 1987. By 1987, the
nontraditional category provided fully one-third of total export
revenues. If the post-1983 adjustment policies deserve credit for this
impressive performance, they might be considered an overall success.4
While the relative price of exports as a whole continued to fall from its
peak in 1980-81, the price incentive for these new exports, with the
tax relief for nontraditional exports and draw-back incentive systems
in place, may have been strong. In any case, as the policies of short-
term stabilization gave way during 1984-1987 to those of longer-term
structural adjustment, the composition of exports did change in a
manner consistent with the goals of -die Costa Rican government
While export and import trends determine the achievement of
external payments balance, trends in the production of exportables,
4. Given the high import content of these ecports (as much as 80 to 90 percent of
material inputs used), the benefits from this growth of nontraditional exports are
somewhat less than they appear
Table 6.5 Merchandise Exports by Category and Sectoral Price Deflators, 1975-87
Category 1975 1976 1977 1978 1979 180 1981 1982 1983 1984 1985 1986 1987
GD)P (millions
of 16 collones) 7,472.5.- 7,885.1 ct,586.9 9,125.1 9,575.8 9,647,1 9,429.6 1,742.6 8,992.9 9,714.5 9,7R4.6 10,326.3 10,817.6
Total exports 493.6 592.9 128.0 865.0 942.1 1,100.9 1,008.6 1869.8 852.5 997,5 939.1 1,084.8 1,113.5
Ttadilional exports
(UradtiSa mlos 345.0 391.4 560,0 5H.9 624.1 #51.0 599.9 545.0 532.6 604.6 599.9 694.3 643.9
Exports to the Central
American Common
Market (CACM)
(USS millions) n,a. 130.6 173,8 178,7 175.4 270.3 238,0 167.2 198.2 192.9 143.5 98.9 98.0
Other exports
(US$ millions) n.a. 70,9 94.2 100,4 142.6 149.6 170.7 157.6 121.7 200.0 195.7 291.6 371.6
Nontraditional exports
* to ihued matkets
(VISt millions) . n.a, n.n. n.. n.m. n.a. n.a. n.a. 104.2 92.4 137.4 163.7 240,2 312.9
Maquila expotls (value added)
(VSS millions) n.a. n.a. n.. n,a. n,a. n.a. n.a. Il0.8 16.9 26.1 34.5 34.2 42.9
GDP price dellators (1966 = 100)
Agriculturc 215.5 264.4 358.8 355.0 366.7 424.6 720.4 1,373.6 1,573.1 1,737.0 1,985.R 2,664.4 2,553.4
Manufacluring 21S,9 242,5 264.1 276.'1 301.1 363.3 512.9 1,061.1 1,485.8 1,746,0 2,040.6 2,292.5 2,479.5
Scrvlcs 233.8 271.2 310.1 348.4 387.1 460.6 601.4 1,063.9 1,375.8 1,643.1 2,055.4 2,404,3 2,791.5
GDP 224.9 262.2 306,6 330,9 361,2 429.2 605.6 1,1 15,3 1,438.0 1,678.0 2,022.8 2,401.3 2,639.1
Prices relative to tlPP
Agriculture 0.958 1.008 1.1?0 1.073 1,015 0.989 1.190 1.232 1.094 1.035 0,9N2 1.110 0.968
Manufacluring 0.,960 0,92$ 0-6116 0.835 0.834 0,846 0.847 0.951 1.033 1.041 1.009 0.955 0.940
Services * 1.040 1,034 1.014 1.053 1,072 i.073 0.993 0.954 0.957 0,979 1.016 1.001 1.058
n.a. = not available
Note: The distinction between Iraditional and nontraditional exports to third markets (not thc CACM) is based on the laws defining nonlraditional cxports to
these markets. Prior to 1982, the data necessary to distinguish these categories are not available.
Sources- World Bank (1977, 1988); In additIon, some figures on CACM exports are from SliECA, Evadlslicas Anallicas del Comnerclo
Iniracentiroatnericano (various issues).
Costa Rica 237
importables, and nontradables are of equal interest in assessing
stabilization and adjustment policies, as they largely determine trade
flows in the longer rmn, when shorter-term influences have averaged
out. If one focuses on production of exportables and on the
increasing number of products with export competitiveness, Costa Rica
has managed a reasonably successful adjustment since 1980. If one
focuses on tradables, this is not the case: after an increase in 1980-32
associated with the fall of GDP, their share in output fell (table 6.6).
The public sector share of value added rose smoothly (from 9.9 to
11.2 percent) during 1976-82, then gradually declined to 9-7 percent
with the stabilization and structural adjustment programs of 1983-87.
An upward trend in the share of private nontradables was broken only
by a sharp fall in 1981.
In terms of labor immobility among sectors, the biggest probiem
for successful adjustment would have been barriers to movement into
the exportables sector, and in particular, -into the production of
nontraditional exports. Employment trends by trade-related sector are
somewhat different from those for value added (table 6.7), but there
are no strong hints of such immobility. Employment in exportables
decreased as a share of total employment from 1976 to 1980,
increased significantly in the crisis years, fell with the appreciation of
the colon in 1983, and stayed about constant thereafter what was
probably a downward secular trend up to about 1980 (recall that this
sector includes much of agricultLire) has been at least temporarly
stoppecL The employment share of the importables sector trended
down until 1981, rose through 1983, and then held about constant.
The share of private nontradables fluctuated only slightly during
1976-88, while that of the public sector continued its earlier upward
trend until about 1980, then leveled off (with a dip in 1982).
In general, these data suggest a reasonable degree of mobility
between sectors. Whether it was due to the devaluation or not,
employment in exportables rose significantly during 1980-82, while
that in the public sector fell. Such observed changes in employment
structure could have occurred without worker mobility, however. Thus
in the 1981-82 post-devaluation period, the dramatic increase in
exportables sector employment could have been due to new entrants
joining that sector. Much of the increase in exportables sector
Table 6.6 Value Added by Trade-Related Sector, 1975-87
Sccror 1975 1976 197? 1978 1979 )980 1981 1982 1 083 1984 1985 1986 1987
In millions of 1966 colouws
Exporlables 1,460 1,458.0 1,502.0 1,603.0 1,599.2 1,591.9 1,671.6 1,627.3 1,637.9 1,791.2 1,697.1 1,788.5 1,894.9
Imponrables 1,558 1,545.8 1,827.3 1,996.8 2,047.8 2,013.1 2,097,9 1,761.5 1,812,6 2,052.6 2,081.4 1,864.7 2,024.8
Private nonlradables 4,459.6 4,795,5 4,825.4 4,511.7 4,179.4 4,341.9 4,669.6 4,802.6 5,078.6 5,407.8
Public seclor 551.1 932.7 966,7 984.3 9S5.8 940.5 954.6 959.4 978.6 1,003.0
Piocessed foods 725.3 833.1 927.1 972.3 922.3 952.5 876.1 927,7 1,048.6 1,081.3 1,172,3
Textilees 276.3 295.5 293.6 282,3 284.6 282.7 316.9 328.1 320.1 316.3 302.1
Petrochemical products 279.5 327.7 364f5 368.7 397.2 406.2 307.4 300.1 341.3 339.4 381.8
Electrical machinery 114.5 140.8 156.9 3i66.6 191.5 155.6 108.3 80.2 89.8 90,0 103.0
As a percentagc of ltot value added
Exporiables 17.9 17.1 16.9 18.0 19.1 38.8 18.9 17.8 18.4 18.3
Importables 22,3 21.8 21.4 22Z6 20.7 20.8 21.7 21.8 19.2 19.6
Private nontradables 49.9 51.2 51,3 48.7 49.0 49.7 49.3 50.3 52.3 52.3
Public sector 9.9 9.9 10.3 0.5S 11.2 10.8 10.1 1.t,1 10.1 9.7
Processed food 30.4 10.4 9.8 10.3 10.3 10.6 11.1 11.3 12.1
Textiles - 3.3 3.0 3.0 3,1 3,7 3,8 3.4 3.3 3.1
Petrochemical producis 4.1 3.9 4.2 4.4 . 3.6 3.4 3.6 3.6 3.9
Electlcal machinery 1,8 1.8 2.0 1.7 1.3 0.9 0.9 0.9 1.,
Note: Exports arc defined to Include agriculturo and mining minus basic grains (rice, beans, sorghum). Importables arc defined as
manufactUTing and basic grains. Disaggregated data on manufaclured goads arc available only in current colones; these figures were
converted into 1966 colones using the GDP price deflator. Valuc added In agricultural goods is reported in 1966 colones. Private
nontiadables are defined as elasticity, construction, transportation, finance, durablos, plu$s other services. The public sector is defined
as the central govcrnmciM, municipal governments, and state-owned enterprises. The public scctor and private nontradables value
added aro reported In 1966 colones.
Souirces: World Bank (1977, 1988); Central Bank of Costa Rica, Cuetatas Nacioniales dte Costa Rica (varIous Issues).
Table 6.7 Employment by Trade-Related Sectors, 1976-83, 1985-88 (July)
Sector 1976 1977 1978 1979 1980 1981 1982 1983 1985 1986 1987 1988
Nuimber of warkers
Exportabtes 178,601 179,076 173,306 180,472 175,542. 380,771 197,525 181,378 188.633 199,672 213,250 223,067
Imporables 125,306 138,826 138,780 136,734 139,545 130,052 145,537 160,691 167,502 180,822 206,632 202,113
Private nonruadables 208,252 220,273 246,050( 259,270 267,914 269,809 278,944 278,993 309,334 323,093 348,921 352,927
Publicsector 102,666 114,917 128,856 130,448 142,271 141,692 334,281 145,254 157,792 183,738 150,513 167,501
Processed foods 22,606 25,656 26,850 31,387 31,185 29,862 29,807 28,142 34,878 3R,002 37,081 36,126
Texitles 26,086 32,883 32,143 28,867 311249 27,574 35,053 39,226 38,326 52,822 54,600 55,867
Peirochemlcal products 8,765 9,186 7,450 11,146 13,019 12,285 10,774 15,065 12,640 17,107 16,037 14,675
Electrical machinery 9,159 11,483 9,802 12,765 13,337 10,405 9,776 13,333 11,562 16,768 15,890 13,956
As a percentage of total cmployment
xUxpoaables 30,9 29.1 26,6 25.5 24.2 25.0 26.1 23,7 22,9 22.5 23.2 23.6
Importables 21.7 22.5 21.3 19.3 19.2 18.0 19.2 21.0 20.3 20.4 22,5 21.4
Private nontradables 36,0 35.7 37.8 36.7 36.9 37.4 36.9 36.4 37.6 36.4 38,0 37.3
Public sector 17.8 18.6 19.8 18H5 19.6 19.6 17.8 19.0 19.2 20.7 16,4 17.7
Puocessed rood 3.9 4.2 4.1 4.4 4.3 4.1 3.9 3.7 4.2 4.3 4.0 3.8
Textiles 4.5 5.3 4.9 4.1 4.3 3.8 4.6 S.1 4.7 6.0 5.9 5.9
Pelrochemical products 1.5 1.5 1.1 1.6 1.5 1.7 1.4 2.0 3.5 1.9 1.7 1.6
Eleclrlcalmachinery 1.6 1.9 1.5 1.8 1.8 1,4 1.3 1.7 1.4 1.9 1.7 1.5
Notes: For the definition of the sectors see table 6.6. Figures for public sector employment in 1987 and 1988 are probably biased
downward relative to those of earlier years, as the definition of the public sector was narrowed between the household surveys of 1986
and 1987. Accordingly, one or more of the other categories would be upwardly biased, The main candidate is private nontradablcs,
since most government employment coiisists of the provision of services (Oindling and Berry 1991). Public sector construction also
fell sharply between 1986 and 1987, probably due to this reclassification. Since construction also is a nontradable, it scems unlikely
that the figurcs for private tradabics riscs spuriously betwccn 1986 and 1987 due to the change in the definitlon of the pubilc sector.
Soutrce: Calculations based on the household surveys of employment and unemploymcnt for July of each year, For 1976-78,
employment hi agriculture was not available separately for basic grains and other products. We have estimated cmployment in basic
grains for those years by assuming tho same proportion of all agricultural workers werC so engaged as in 3979.
240 T. H. Gindling ndAlbertBerry
employment may have been of relatively lower-skilled workers, whose
behavior might indicate the degree of mobility of the higher skilled
workers needed in the nontraditional export sectors. It is difficult to
know how much or what sort of labor reallocation would be required
for a given process of structural adjustment to be effective.
Accordingly, one must consider direct evidence on impediments to
mobility, including sector-specific wage rigidity and public policies
that could increase the costs of mobility.
Sectoral Wages
The evolution of real wages in Costa Rica during the crisis penod
leaves no room for doubt that considerable downward flexibility exists
in the short run, both on average and for each category that can be
singled out. However, as real wages recovered rather quickly and
mechanically, one could argue that long-run downward rigidity does
exist Efficient reallocation of labor is more likely to be impeded by
rigidity of relative wages between sectors, so testing for such rigidity is
important, if possible in both the short and the longer run. In assessing
potential relative wage rigidity across trade-related sectors, the details
of wage setting processes are of obvious relevance, together with
statistical evidence on wage and earnings trends for workers in those
categories. Evidence on formal-informal sector earnings gaps and how
they have changed over time may be relevant to this question.
Gindling (1991) and Pollack and Uthoff (1986b) found wage
differentials of 20 to 30 percent between the private formal and
informal sectors and of 10 to 20 percent between, the public formal
and private formal sectors, after controlling for human capital
characteristics and selectivity bias. The gap between private formal and
informal sector wages for observationally identical individuals
increased during the crisis when real wages were falling sharply, and
then decreased (though perhaps not monotonically) between 1982
and 1985, when real wages were rising. Since informal sector wages.
fell more sharply during 1980-82 than formal sector wages, one
could argue that the latter, however sharp their descent, were more
aprotectedn from those of the informal sector at the height of the
crisis than previously or subsequently. As the share of workers who
fall in the formal category, is higher in the nontradables (public and
Costa Rica 241.
private together) than in the tradables sector (Gindling and Berry
1991), one might anticipate more rigidity in nontradables than in
tradables wages, but such evidence permits only tentative hypotheses.
Public sector wages/salaries have, on average, been higher than
those in the private formal sector for individuals with comparable
human capital, and it would not have been surprising if they had fallen
less during the crisis. The public sector is the only highly unionized
one, and the government is in any case sensitive to the charge that it is
not paying its workers a "just wage," a charge most likely to come
from the politically powerflul college educated group, most of whom
work in the public sector. Unfortunately, the behavior of public sector
wages during and after the crisis is unclear due to inconsistent pieces
of evidence (Gindling and Berry 1991). It is clear that with job
security high in the public sector, good working conditions, and other
nonwage benefits, people would tend to think twice before leaving the
public employ.
Meanwhile, the minimum wage system could affect the relative
wages of the exportable, importable, and nontradable sectors either
through differential adjustment or differential coverage, but Pollack
and Uthoff (1986a) found no evidence, of the former for 1976-82.
Enforcement is at best partial among small firms of the informal
sector, as reflected in the high share of salaried workers in agriculture
(from which most exportables come) earning below the minimum
wage.
While the background information on labor market institutions just
reviewed provides some grounds for worry that sectoral wage rigidities
.could hamper adjustment, it needs to be complemented by direct
evidence on wage patterns and trends. Household survey data reveal
that average wageslearnings ate lowest in the exportables sector, in the
middle range for the importables and nontradables sectors, and
highest in the public sector 'Gindling and Berry 1991). The lower
exportables sector wages could be due partly to their being less
protected than those in the other sectors and/or ic a. lower average level
of human capital. When one controls for human capital (see Gindiing
and Berry 1991) average pay. remains lowest in the exportables sector,
and although pay is, on average, higher in the public sector than in the
importables and priv-ate nontradables sectors, it is not systematically- or
242 T. H. Gindling andAlbertBerry
substantially so (tables 6.8 and 6.9). Although most of the difference
in average pay between the public sector and the importables and
private nontradables sectors is due to the higher average levels of
experience and education of the former group, the gap between the
exportables sector and the other three is not fully explained by such
human capital differences. It must reflect either market segmentation
or compensating nonmonetary benefits between the exportables sector
(many of whose workers live in rural areas) and the rest of the
economy.
Prior to 1980, real wages/eamings rose in all four of these sectors;
whether one adjusts for changes in human capital or not. Between
1980 and 1982, wages (unadjusted for human capital) fell sharply in
all sectors, but less in exportables and the public sector than elsewhere.
Thus, paid wages in the importables and nontradables sectors were not
at all inflexible either in absolute terms or relative to those in the
exportables sector. Although due to data problems there is some
ambiguity as to the trends in public sector pay, there is certainly no
clear suggestion of downward rigidity.
In the period of wage increases since 1982, all sectors have gained
significantly: all reached the 1980 level in either 1986 or 1987. The
increases were fastest for importables and private nontradables, the two
biggest losers during 1980-82. By 1987-88, the average wage
differentials across the four sectors were almost identical to what they
had been in 1979-80. Adjusting for changes in human capital all
three of the other sectors had gained somewhat on exportables since
1982, and the 1979-80 differentials had been approximately restored
here too.
Income Distribution, Poverty, and Unemployment
Income distnbution in Costa Rica has traditionally been unequal,
but substantially less so than in such countries as Brazil. Estimates of
the Gini coefficient of household income (with households ranked by
income, not per capita income) have typically fallen in the range
Table 6.8 Average Real Earnings and Real Wages by Trade-Related Sector, Controlling for Human Capital
Characteristics, 1976-83, 1985, 1987-88 (July figures)
Sector 1976 1977 1978 1979 1980 1981 1982 1983 1985 1987 1988
Wages (1975 colones per
hour) for all workers, Including
self-employed and owncrs
Exportables 6.0! 4.81 3.95 5,05 5,82 6.24 5.54
Imporlables 6.47 5.8 1 4.40 5,39 6,34 7.65 6.72
Private nontradobles 7.23 5.94 4.93 5.65 6.95 7.88 7,45
Public 6.68 6.09 4.65 5.99 6.66 7.63 6.87
Wages of all workers as a proportIon
of those in exportables
Exporlables 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Imponables 1.08 1.20 1.12 1.07 1.09 1.23 1.21
Pnmvate nontradables 1,20 1.23 1.25 1,12 1,19 1.26 1.34
Public 1 '1 1.26 1.18 1.19 1.14 1.22 1,24
Weges (1975 colones per hour)
of paid workers
Exportables 4.58 5.45 5.07 4.38 5,39 .439 3,73 4,92 5.49 5.51 5,14
Imponables 4.87 5.92 6.39 6.78 6.29 5.65 4.04 5.32 6.29 7.09 6,62
Private nontradables 4.99 5.88 5.86 6.47 6.53 5.32 4.12 5.07 6.08 7.36 6.62
Public 5.98 7.65 6.95 7.42 6.71 6.02 4.63 6.00 6.58 7.56 6.68
Wages of paid workers as a
proportion of those in
exportables
Exportables 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Impotables 1.06 1.09 1.26 1.55 1.17 1.29 1.08 1.08 1.15.I 1.29 1.29
Private nontradables 1.09 1.08 1,16 1,48 1,21 1,21 1.11 1,03 1.11 1.33 1.29
Public 1.30 1,40 1.37 1.69 1.24 1.37 1.24 1.22 1.20 - 1.37 - 1.30
(Table continues on the following page.)
Table 6.8 (continued)
Sector 1976 /977 1978 1979 1980 1981 1982 1983 1985 1987 1988
Eanmings (1975 colones per
month) of all workers,
including self-employed
and owners
Exportables 1,121.2 916.1 720.5 936.2 1,096.6 1,186.0 1,282.7
Impaunbles 1,275.1 1,080,9 813,5 1,021.3 1,217.7 1,425.8 1,390.2
Private nontradables 1,388.8 1,119.5 879.8 1,054.8 1,324.6 1,481.2 1,368.6
Public 1,331.1 1,207,5 915.3 1,156.5 1,295.0 1,478.8 1,356.2
Earnings of all workers as a
proportion of those in
exportables
Exporiables 1,00 1.00 1.00 1.00 1.00 1.00 1.00
Importables 1,14 1,18 1.13 1.09 1.11 1.20 1,08
Prvate nontradables 1.24' 1.22 1.22 1.13 1.21 1.25 1.07
Public 1.19 1.32 1.27 1.24 1.18 1.25 1.06
Eamings (1975 colones per
month) of paid workers
Exporables 838.2 1,003.8 988.7 804.1 1,064.5 868.1 667.0 938.5 1,072.4 1,099.5 1,072.9
Importablcs 1,006.6 1,247,3 1,330.1 1,388.3 1,279.6 1,104.9 801.7 1,044.0 1,239.4 1,441.9 1,329.0
Private nontradables 986.7 1,164.8 1,160.1 1,252.8 1.269.4 . 1,017.0 750.3 976.3 1,095.5 1,431.8 1,279.3
Public . 159,0 1,473.8 1,366,9 1,421.1 1,329.7 1,186.9 902,1 1,148.4 1,273.4 1,453.6
1,322,9
Eamings of paid workers ns
a proportion of those In
exports
Exporlables 1l00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Imporiables 1.20 1.24 1.35 1.73 1,20 t.27 - 1.20 1.11 1.16 1.31 1.24
Private nontradables 1.18 1.16 1.17 1.56 i.39 1.17 1.12 1.04 1.02 1.30 1.19
Public 1,38 1.47 1,38 1.77 1.25 1,37 1.35 1.22 1.19 1.32 1.23
Wages (1975 colones per
hour) of all workers,
Including self-employed
and owners, as a proporton
of nverage wages
Exporiables 0.90 0.85 0.87 0.91 0.89 0.84 0.82
Inmportables 0.97 1.02 0,97 0.98 0.97 1.03 1,00
Private nontradables 1,08 1,05 1,09 1,02 1,07 1.06 1.11
Public 1.00 1.07 1.03 1.08 1.02. 1.03 1.02
Wages of all workers as a
proportion of average wages
Exportables 0.86 0,.5 0,S0 0,71 0.86 0.83 0.91 0.93 0.90 . 0.80 0.82
Imporablcs 0.91 . 0.92 1.01 1.10 1.01 1.07 0.99 1.01 1.03 1.03 1.05
Private nontradables 0,94 0.91 0.93 1.05 1.05 1.01 1.01 0.96 1,00 1.07 1.05
Public 1.12 3.19 1.10 1.20 1.08 1.14 1.13 1.14 1.08 1.09 1.06
Earnings (1975 colones per
month) of paid workers as a
proportion of average earnings
Exportables 0.87 0.85 0.87 . 0.90 0.88 0.85 0.95
Impartables 0.99 1.00 0.98 0.98 0.98 1,02 1.03
Private nontmdabiles 1.08 1.04 1,06 1.01 1.06 1.06 1.01
Public 1t03 1.12 1.10 1.11 . 1.04 1.06 1.00
Earnings of paid workers as n
proportion of average earnings
Exporiables 0.81 0.79 0.79 0.67 0.86 0,84 0.87 0,92 0.93 0.81 0.86
Jmponables 0.97 0.99 1.06 1.16 1.04 1.07 I,05 1,03 1,07 1.06 1.06
PrlvaIenontradables 0.95 0.92 0.92 1.05 1.03 0.99 0,98 0.96 0.95 1.05 1.02
Public 1.12 1.17 1.09 1,19 1,08 i,ls 3.18 1.13 1.10 1.07 1.06
Source: Calculated from wage and earnings equations estimated for cach sector using the average values of education, experience, and
log of hours worked for each year. Nominal real wages and earnings (excluding fringe benefits) are deflited by the July San Jos6 cost
of living index for lower- and middle-income families.
Table 6.9 Wage and Earnings Equations, 1976-83, 1985, 1987-88
Sector 1976 1977 1978 1979 1980 1981 1982 1983 1985 1987 1988
Hourly wages as
dependent vadable,
paid workers only
Intcrcept .0.1136 0.001252 *0.11565 0.09484 0.2871 0.3671 0.7327 1,3817 1.915 2,284 2.5026
Edacation 0,114 0,1052 0.14S 0,1426 0.1348 0.1382 0.1341 0,1151 0.1146 0.1147 0.123
Experience 0.03572 0.05485 0.05637 0.05111 0.0505 0,05362 0.05327 0.04881 0.14297 0,042B 0.03911
Experience squared .0.00036 .0.00073 .0.00070 0.00062 .0,00063 .0,00065 .0.00064 .0.000S9 .0.000S0 -0.00051 -0.00045
Sex 0.2847 0,3275 0,2886 0.353 0.3318 0.3101 0.3416 0.3407 0,25855 0.2696 0.1945
R-Squared 0.28 0,43 0,46 0.462 0.472 0.459 0.421 0.379 0.4055 0.2632 0.356
Monthly earnings as
dependent vulable,
paid workers only
Intercept 4.1944 3.0631 2.8634 2.9932 3.1971 3.4728 3L9097 4.413 5.0388 5.321 5.306
Edacation 0,1321 0.1054 0.1463 0.1408 0,13325. 0,1334 0.13 0.1116 0.1098 0.116 0.1148
Experience 0,04332 0,05928 0,05949 0,05403 0.048835 0,0504 0.05498 0.0485 0,045 0,04651 0.04252
Experience squared -0.00050 O,00080 -0.00076 .0,00067 *0.00059 .0,00060 .0.00065 .0,00060 .0,00054 .0.00058 .0.00052
Sex 0,2998 0.3486 0.3165. 0.3713 0,3627 0.347 0.3974 0.3951 0,2846 0.3284 0,2691
Log of Hours 0.2319 0.5646 0.S937 0,6136 0,6173 0.5743 0.5391 0.5828 0.5623 0.5601 O.bI99
R-Squared 0.327 0.499 0.515. 0.526 0,494 0.497 0.4S 0.426 0,42 0.3674 0.372
Hourly wages as
dependent variable,
ail workers
Intercept 0.3437 0.4845 0,8232 1.474 1.989 2.171 2.503
Education 0.1325 0.1314 0.1277 0.1079 0.1091 0.121 0,113
Experience 0.04643 0.04756 0.05 0.04579 0.04183 0.0464 0.03911
Experierce squared *0.00055 *0.00055 -0.00056 -0.00055 .0.00048 -0.00055 -0.00045
Sex 0,3217 0.2966 0.3308 0.3474 0.2397 0,3133 0.1945
R-Squared 0.4021 0.386 . 0.332 0.299 0.319 0.3701 0.251
Mfonthly eawings as
dependent variable.
all workers
Intercept 3,1686 3.307 3.598 4.245 4.9812 4,978 5.0879
Education 0.1341 0,1383 0.1352 0.1171 0.11402 0.1212 0.1188
Experience 0.05271 0.05641 0.0577 0.05175 0.04586 0.04874 0.04547
Experience squared .0.00067 .0,00070 -0.00071 .0.00064 *0.00055 -0,00057 .0,00055
Sex 0.3488 0.3356 0.3626 0.3663 0.2734 0.3395 0.2806
Log of Hours 0.6159 0,5973 0,6092 0.6121 0.5677 0.6337 0.6644
R-Squared 0,531 0.534 0.519 0.482 0.481 0.4612 0.448
Note: In alt cases the natural logarithm of the dependent variable is used, All coefficients are significant at I percent,
SoUrce: Household surveys of employment and unemployment.
248 T. H. GLndling andAlbertBerry
0.43--0.50.5 Some authors suspect that income distribution was
worsening during the crisis years of the early 1980s (for example,
Trejos and EIizalde 1986), but no reliable data are available to clarify
the changes in household distribution at this time. The distnbution of
income among earners (substantially less unequal than that among
households, w*ith Gini's usually in the range 0.35-0140), can be traced
more successfully. They suggest little change during 1976-80,
possibly a mild worsening during 1980-82, and then a rather marked
improvement in the next two years (CEPAL 1987, table 6.1). As for
the key period of macroeconomic crisis, although the earner data
indicate some worsening, some less solid household data suggest the
opposite, so the case cannot be considered closed on this point 6 The
marked increase in nonhousehold heads as a share 'of employed
workers would by itself produce some worsening in the earner
distribution, but might simultaneously improve household
distnrbution. (Our figures, however, -show a higher share of household
labor income coming from household heads in 1982 and 1983 than
in earlier or later years. The reason for this apparent anomaly is not
clear-) The sharp drop in real wages in the fornal and public sectors
during the crisis would be expected to lower labor income most
sharply for the deciles toward the middle and the top of the
distnbution. As those incomes rebounded in later years, the shares
move back up again Since none of the trends suggested by the data
can be accepted with great confidence, one cannot nule out the
possibility that household distribution worsened during 1980-82. In
particular, the capital share would have risen at this time, implying an
increasing concentration that would go unrecognized in data that
either does not include at all or seriously undersates capital incomes.
All sources that have measured poverty trends during the crisis
show the expected sharp increase during 1980-82, with recovery in
5. Most surveys exclude nonlabor income, and thus understate income inequality,
but two (1971 national level, and 1974 urban level), which aimed at inclusion of all
incomes and whose underreporting seems to have been relatively small, produced Gini
coefficients of around 0.45 (CEPAL 1987, tables 4, 5.1, and 5.2).
6. Altimir (1984) reports a decline in the Gini coefficient among households
(ranked by per capita income) from 0376 to 0.346 between July 1979 and June 1982,
with significant share increases for each of the bottom deciles (from 20 to 2.6 for the
lowest deciles).
Costa Rica 249
later years back to or near the turn of decade figures. It is of interest
to know which households defended themselves best at this time, and
how the safety nets in place in Costa Rica performed. With a relatively
high poverty line (188 1980 colones per capita per month in urban
areas and 166 in rural areas), the incidence of poverty among
households rose from 48 percent in 1980 to 78 percent in 1982, then
fell back to about 45 percent by 1987-88. The increase was very
similar for female- and for male-headed households, with poverty
incidence typically 10 to 15 percentage points higher for the former.
Incidence rose much less than average for households headed by self-
employed people, whereas in 1980, the incidence was much higher
than for households headed by paid workers (56 percent to 40
percent). At the peak of the crisis in 1982, the figures were identical at
72 percent. Thereafter a gap reappeared, though smaller (6 to 8
percentage points in 1986-88). Possibly the incomes of self-
employed heads of households fell less than those of paid workers; we
know that the latter fell very sharply.
Prior to the crisis, unemployment was low, averaging under- 5
percent nationally each year during 1976-79. It was heavily
concentrated among younger workers. Persons with a high school
education had the highest unemployment rate, while those with no
education or a college education had the lowest rates. The female rate
exceeded that for men, and nonhousehold heads suffered a higher rate
than heads of households. The crisis raised unemployment rates for all
groups, but most notably for household heads and for men
(substantially overlapping groups). As noted earlier, the crisis saw an
increase in the labor force share of secondary family workers
(nonhousehold heads), making it striking that this group's share of
unemployment fell. These various trends reversed -themselves with the
recovery and structural adjustment The underemployment rate took a
year or so longer than the (open) unemployment rate to return to
precrisis levels (table 6.4).
Public Expenditure on Education and Other Social
Services
Public spending on education fell sharply in real terms during the
economic crisis (35 percent between 1980 and 1982 when curent
250 T. H. Gindling andAlbertBeey
price spending is deflated by the GDP deflator). With the upturn
spending has recovered somewhat, though the 1987/88 average was
still 15 percent below the 1980 level. The drop in spending on
education was not translated into a fall in education services, at least as
far as the available data indicate. The number of educatiohal
institutions and the staff at the Ministry of Public Education continued
to grow, albeit more slowly than before 1980. Real public spending on
education fell initially because salary levels fell during the economic
crisis, reflecting the fall in real salaries in the economy as a whole.
Public spending on materials, machinery, buildings, and capital have
also fallen dramatically since 1980, explaining the continued fall in
total expenditures during the period of economic recovery. If not
reversed, this trend will sooner or later have negative consequences on,
thc quality of education in Costa Rica (Gindling and Berry 1991).
Stabilization and structural adjustment influence the effective
demamnd for education as well as the supply. During the cr:isis years
from 1980 to 1983, total enrollment in formal schooling decreased
instead of recording the expected increase (Gindling and Berry 1991).
The most striking decrease, 24 percent in secondary enrollment
between 1980 and 1985, was probably due in part to former or
potential secondary and technical school students entering the labor
force as secondary workers to help maintain family incomes-
The reduction in overall government spending brought a decrease
in spending on most social programs, both in absolute ternis and as a
proportion of the GNP? Spending on health declined sharply from its
1980 high. As with education, the decrease was not reflected in a large
immediate decline in the level of health services, although there were
declines in some services (days in hospital, number of consultations)
and in the number of health care establishments in 1982.7 However,
investment in equipment and buildings fell in the face of the need to
meet day-to-day costs, so the major effects of the decline in overall
health spending will probabl) be felt only after a lag.
7. Days spent in hospital may have fallen for exogenous reasons. The drop in the
number of health care establishments was primarily due to a drop in temporary health
centers (mobile units and puestos .de salud)
Casta Rica 251
Women and the Crisis
Women accounted for 'about 25 percent of the Costa Rican work
force at the beginning of the 1980s and probably closer to 30 percent.
by 1988 (table 6.4). Most of the rise was probably the result of secular
forces, but the crisis years did seem to produce a slightly faster than
normal increase. The male/female earnings differential increased with
the economic crisis and then fell with the recovery. However, Gindling
(forthcoming) argues that this decrease in women's relative earnings
was due primarily to the lower educational level (and correspondingly
lower earings) of new entrants than of women already in the work
force. After the recession, the earnings differential fell back to its
previous level, and appeared to be continuing its downward secular
trend.
By entering the labor force during the recession, women helped to
shore up falling family incomes. New labor force entrants seemed to
find jobs at least as easily as men, judging from the fact that the
female share of unemployment fell even as their share of the labor
force was rising (Gindling forthcoming)-
Conciusion
Costa Rica's recovery from the sharp downturn of 1980-82 has
been relatively good, with output, wages, and unemployment
recovering to precrisis levels faster than in many other countries.. The
net resource inflow to Costa Rica (imports of goods and nonfactor
services minus exports of goods and nonfactor services) fell from over
10 percent of GNP in 1978-80 to about 1 percent during 1983-88
taken as a whole. The fiscal situation was greatly improved and
inflation brought down to about 15 percent or below from 1984 on.
The gross fixed investment ratio, after falling from its very high level
of about 25 percent in 1980 (constant 1966 prices) to 15 percent in
1982, gradually climbed back up to average a litde over 20 percent
during 1986-88 (table 6.3). All this suggests that the Costa Rican
model for stabilization, adjustment, and recovery was a good one.
Success, if that is what it is, has, of course, not come without a price.
All wages fell sharply, in most cases by 25 to 40 percent during the
crisis years 1980-82. Tradables sector wages fell by about the same
252 T. H. Gindhlng andAlbertBerry-
amount as those in the nontradables sector. Unemployment and
poverty incidence rose for all groups, but less for female-headed
households and for households headed by self-employed workers
than for other groups. Secondary workers entered the labor force in
significant numbers, and did not appear to have unusual difficulties in
getting jobs. Enrollment in secondary schools fell markedly. However,
the crisis was not drawn out Real wages recovered quickly to precrisis
highs by 1987, unemployment and underemployment returned to
precrisis lows, and primary and university enrollments resumed their
upward trends. Secondary enrollment, however, remained far below
the 1980 level.
For some time the main doubt was whether enough structural
adjustment had taken place to permit sustained growth. The initial
balance of payments adjustment came mainly via the sharp
curtailment of imports, whose 198586 average was still 16 percent
below that of the peak years 1978-80. By 1987-88 they exceeded the
1978-80 level by 18 percent. The level of exports (of goods and
nonfactor services) rose by under 10 percent between local peaks in
1979 and 1986, but a boom in 1987-88 put the 1988 figure 42
percent above that of 1979. The brightest spot on the export front has
been the rapid growth during the recovery of nontraditional, non-
CACM exports, a growth that more than offset the sharp reduction in
the latter category, at least in terms of gross foreign exchange
revenues. By the late 1980s, therefore, there were grounds for
optimism that the trade imbalance was on its way to being brought
under control.
Note that taldng the postcrisis period as a whole, adjustment of the
balance of payments was not sought by the broad pnrce incentive of a
sustained high exchange rate. Though the 1980 devaluation was sharp,
the real exchange rate eroded fairly quickly, so that by the latter half
of the 1980s it was at about the same level as in the mid-1970s. A
main push was provided by incentives to nontraditional exports, and as
noted earlier, these have become the dynamic component of exports.
Given the record of low national savings, the generation of high
domestic investment without the large foreign savings inflows of the
1970s was destined to be a serious challenge. Here too the outcome
thus far is encouraging. The nation savings rate, which averaged just
Costa Rica 253
13 percent during the 1970s, exceeded 20 percent during 1986-88
(table 6.3).
An interesting question is whether the' rapid return to relatively
high wages discouraged overall labor absorption. The evidence from
the unemployment rate suggests not. Indeed, it may be that the quick
recovery of real wages contributed to the smooth macroeconomic
recovery by keeping aggregate demand up. On the question of
whether wage setting procedures discouraged intersectoral mobility,
taking an adequate reading is harder. Public sector wage policies do
appear to have pushed wages in that sector above those of the private
sector (for comparable workers) and differential enforcement of
minimum wage laws does presumably contribute to the wage gap
between the formal private sector and the informal sector. However,
these factors certainly did not prevent.wages in the protected sectors
from falling sharply in response to the devaluation/stabilization events.
of 1980-82, nor greatly widen the wage gap in relation to the
relatively unprotected exportables sector workers. The quick and
complete wage recovery is probably due mainly to these institutional
mechanisms. The fact that recovery was as complete. in the. less
protected as in the more protected sectors does not mean that the
"protective" wage institutions were unimportant. In Costa Rica, social
resistance to low wages appears to be greater than' in many other
countries, so informal sector wage movements might-follow those of
the formal sector more than in most countries..
We have found no evidence that wages in the importables,
nontradables, or public sectors are inflexible relative to those in the
exportables sector, nor that labor mobility across sectors was seriously
limited. Some data suggest that the flexibility of relative wages and the
mobility of workers in the aftermath of the devaluation was
particularly marked during 1980-82, though given the lack of direct
information, there is inevitable ambiguity on these points.
It is ironic that the institutions designed to limit downward wage
flexibility in Costa Rica not only were ineffective during 1980-82
because the mechanisms were not designed to deal with rapidly
accelerating inflation, but that they may have facilitated the wage
declines'by making them an automatic result of those impersonal
mechanisms. Any surmise as to the effect of the institutions depends,
254 T. H. Gindfing andAlbertBerry
however, on one's interpretation of how decisionmakers use the
existing mechanisms, and what political pressures they are subjected
to.
If Costa Rica continues on a satisfactory growth path for a few
more years, its crisis management and adjustment strategies will have
been judged successful. Several factors have probably contributed
Perhaps the most general has been a relatively satisfactory sharing of
the burden of crisis/adjustment such that no groups remained so
disgruntled as to pursue highly disruptive tactics. Strong political
leadership at a crucial time (1982-86) by an individual with a
prolabor history but a recognition of the need for belt-tightening may
have been important here. Also of note was the decline of union
influence since 1980, both in the private sector, where it was never
strong, and in the public sector. That this decline created no more of a
furor than it did may be associated with the rapid recovery of real
wages, and it is not clear in the context of the Costa Rican institutions
that produced that wage recovery that the unions' weakness was an
important contributor to adjustment. The rapid real wage recovery
may have prevented some potential negative multiplier effects that
might otherwise have prolonged or deepened the crisis.
Another special feature was the country's small size. This may have
made it easier to enter new foreign markets, and crtainly facilitated
access to foreign aid. Perhaps the over 25 percent fall in per capita
absorption between 1980 and 1982, comng off a peak only achieved
for a few years, eased the trauma. Total absorption in 1982 was the
same as in 1976, and absorption per capita was only 16 percent lower
and private consumption per capita only 11 percent lower.
On the trade side, the absence of a worsening of the terms of trade
during the crisis/adjustment period was a bit of luck, given the sharp
fluctuations experienced in the 1970s. In terms of policy, the use of
special incentives to chosen categories of exports may have been both
an efficient way to stimulate exports and an approach to minize
social costs, if a more devalued exchange rate would have either
promoted. a higher rate of inflation or tended to push real wages
lower.
The broad question that emerges from the Costa Rican experience
is whether it provides a valuable model to be pursued by other
Costa Rica 255
countries in comparable situations. Certainly the outcome was better
than most. It is not yet clear which among the relatively good access to
foreign finance, the considerable social consensus, the sharp decline
but sharp recovery of real wages, the targeted support for new types of
exports, or the quick action on the fiscal problem after 1982 were
central to the relative success achieved. When the effect of each of
these factors has been better sorted out, the policy implications of this
case will hopefully come into focus.
References
Altimir, Oscar. 1984. "Poverty, Income Distribution and Child
Welfare in Latin America: A Comparison of Pre- and Post-
Recession Data." World Development 12(3):261-282.
Bourguignon, Frangois. 1986. "Income Distribution and Extemal
Trade: The Case of Costa Rica" Paris. Processed.
Carillo C., Mario Alberta. 1988. "Las formas de financiamiento
estatal en el desarrollo economico reciente de Costa Rica."
Working Document No. 117. Instituto de investigaciones en
Ciencia Economica
Celis, Ri, and E. Lizano. 1990. 'Development in Costa Rica: The
Key Role of.Agriculture." Paper prepared for the IFRRI
Conference of Agriculture and the Road to Industrialization,
Taipei.
CENPRO (Centro de la Provici6n de Exportaciones). 1986. "Analisis
de] Potential de Exportaciones no Tradicionales de Costa
Rica.' San Jos6. Processed.
CEPAL (Comision Econ6mica par America Latina). 1987.
Antecedentes Estadisticos de la Distribucidn del Ingreso en
Costa Rica: 1958-1982. Santiago, Chile: United Nations.
Cespedes, Victor Hugo, Runolfo Jiminez, and Alberta DiMare. 1985.
Costa Rica: Recuperacidn sin Reactivacie5n San Jose:
Academia de Centroamerica.
Donato, M. Elisa, and B. Manuel Rojas. 1987. Sindicatos Politica y
Economia: 1972-1976. San Jose: Editorial Alma Mater.
256 r. H. Gindling andAlberrBerry
ECLA. Various years. Statistical Yearbook for Latin America and the
Caribbean.
Gindling, T. H. 1989a. "Crisis economica y segmentaci6n en cl
mercado de trabajo urbano de Costa Rica." Revista de
Ciencias Economica v (University of Costa Rica) 9(1): 77-93..
. 1989b. "Women, Earnings and Economic Crisis in
Costa Rica." Paper presented at the International Congress of
the Latin American Studies Association, Miani, December.
. 1991. "An Investigation into Labor Market
Segmentation: The Case of San Jt., Costa Rica." Economic
Development and Cultural Change 39(3): 585-606.
. Forthcoming. "Women, Earnings and Economic
Crisis in Costa Rica." Economic Development and Cultural
Change.
Gindling, T. H., and Albert Berry. 1991. The Labor Market in
Successful Adjustment: Costa Rica. Processed.
Gonzalez-Vega, Claudio. 1984. "Fear of Adjusting: The Social Costs
of Economic Policies in Costa Rica in the 1970's." In Donald
Schulz and Douglas Graham, eds., Revolution and
Counterrevolution in Central America and The Caribbean, pp.
351-384. Boulder, Colorado: Westview Press.
Latin American and Caribbean Contemporary Record. 1983. New
York: Holmes and Meier.
Nelson, Joan. 1989. "Crisis Management, Economic Reform, and
Costa Rican Democracy." In Barbara Stallings and Robert
Kaufiman, eds., Debt and Democracy in Latin America, pp.
143-162. Boulder, Colorado: Westview Press_
Pollack, M*, A. Uthoff. 1986a. "Wages and Price Dynarics in Costa
Rica: 1976-1983." Monograph on Employment No. 51,
Santiago, Chile: PREALC/ECIEL.
- 1986b. "lInfacion, salario minimo y salarios.
nominales 1976-1983." Revista de Ciencias Economicas
(University of Costa Rica) 6(1): 57-78.
Costa Rica 257
Rivera Urrutia, Eugenio, Ana Sojo, and Jos6 Roberto Lopez. 1986.
Centroamerica: Politica Economica y Crisis. San Jos6:
Editorial DEI.
Rodriguez V., Adrian. 1987. "La deuda puiblica externa de Costa
Rica: Cresimiento, moratoria y renegociaci6n." Revista de
Ciencias Economicas (University of Costa Rica) 7(2):12-35.
Seligson, Mitchell, and Edward Muller. 1987. "Democratic Stability
and Economic Crisis: Costa Rica, 1977-1983." International
Studies Quarterly 31: 301-326.
Trejos, Juan Diego, and Maria Laura Elizalde. 1986. "Ingreso,
desigualdad y empleo: Evidencias recientes sobre las
caracteristicas y evolucion del perfil distributivo en Costa
Rica." Revista de Ciencias Economicas 6(2): 87-104.
World Bank. 1977. "Economic Positions and Prospects of Costa
Rica." Report No. 1666-CR. Washington, D.C.
. 1988. "Costa Rica: Country Economic Memoran-
dumr.' Report No. 7481-CR. Washington, D.C.
. Various years. World Debt Tables. Washington, D.C.
'COi D'IVOIRE
Richard Blundell
Christopher Heady
Rohinton Medhora
The Government of C6te d'Ivoire introduced a structural
adjustment program in 1981 in response to the growing balance of
payments and budget deficits. By 1980, the balance of payments
deficit on current account had reached 17.4 percent of GDP and the
budget deficit had reached 11.9 percent of GDP. This situation
represented a severe deterioration from five years earlier, when the
corresponding figures were 8.2 percent and 2.2 percent, and was
accompanied by a rapid increase in external indebtedness.
A major cause of these deficits was an increase in public
expenditure, particularly investment, that was stimulated by the
revenues generated in the coffee and cocoa boom of 1975-77, but
was financed by foreign borrowing after the boom* ended. The
balance of payments situation was aggravated by inflation that was
more rapid than in Cote d'Ivoire's main trading partners. This was
particularly serous because Cote d'Ivoire is a member of the Western
Africa Monetary Union (UMOA), and so cannot independently
devalue its currency.
The aims and instruments of the structural adjustment program are
set out in the statements of development policy that the government
produced in the context of each of the three World Bank structural
adjustment loans (1981, 1983, and 1986). The overall strategy was
The authors would lie to thank Rob Alessie, Paul Baker, Paul Glewwe, Costas
Meghir, John Newman, Valerie Kozel, and K. Yao and the editors for many helpful
comments, and lindara Addabbo for excellent research assistance.
259
260 Richard Blundell7 Christopher Heady, and Rohinton Medhora
first to eliminate the ilarge deficits and then to restructure the economy
with a shift toward the traded sector.
The elimination of the deficitsthe stabilization phase-was
achieved mainly by a dramatic cut in public investment, fTom 11.6
percent of GDP in 1980 to 3.2 percent in 1985, and a freeze of public
and minimum wages. By 1985, both deficits had been turned into
small surpluses, but at the cost of a severe recession: from 1980 to.
1984, per capita GDP fell by 26.2 percent and per capita private
consumption fell by 22.6 percent (Glewwe and de Tray 1988).
The stabilization phase took place in the context of favorable
intemational economic developments. The U.S. dollar and several
other currencies appreciated relative to the franc, allowing firms to
increase exports at a time of depressed domestic demand. The main
goals of the restructuring program were to reduce government support
for inefficient sectors of the economy, increase aie relative pnces of
traded goods, and shift the internal terms of trade in favor of the rural
sector. Unfortunately, the international economic environment had
changed by 1985, and the CFA franc increased in value against the
currencies of its main trading partners. This counteracted some of the
policies that-were intended to enlarge the economy's traded sector.
Another difficulty that inhibited restructuring was the severe
shortage of credit that developed. Lorch (1989) reports that medium-
and long-term loans to modem manufacturing enterprises declined by
42 percent between 1984 and 1987. This was compounded by the
state's policy of paying suppliers only when it wanted to place the
next order.
In the light of these adverse circumstances, it is not surprising that
the restructuring phase was not as successful as the stabilization phase.
However, some evidence indicates that the losses of public enterpnses
are being brought under control (World Bank 1987). The government
has also introduced some policies to. alter incentives. The measures
used to increase the relative prices of traded goods were changes in
tariffs and an export subsidy scheme aimed at producing a uniform
rate of effective protection (about 40 percent) for manufactured
goods. This was essentially an attempt to mimic a devaluation, but it
was partly offset from 1985 to 1988 by the appreciation of the CFA
franc.
Cd6e d'lvoire 261
The government planned to redress the urban-rural imbalance by
holding down wages in the urban areas while increasing agricultural
prices. Some evidence suggests that urban real wages did decline
around 20 percent, and although many of the increases in agricultural
prices did no more than keep up with inflation, there is also evidence
of a 10 percent increase in the relative -price of traded goods
(Berthelemy and Bourgignon 1989, p. 424).
Most of the analysis in this chapter deals separately with the urban
and rural areas of Cote d'lvoire. There are two main reasons for this
separate analysis. Frst, the structual adjustment program had a much
greater direct impact on the urban areas, which include virtually all the
manufacturing sector. Second, the structure of the labor markets, the
retums to education and experience, and data availability in the two
sectors are very different.
The Macroeconomics of Cote d'Ivoire
As a member of the franc zone, Coite d'Ivoire's recent
macroeconomic history and experience with adjustment make it an
interesting country to study. By belonging to the UMOA, one of the
two monetary unions in francophone Africa, Cote d'Ivoire enjoys
certain privileges and operates under certain institutional and historical
constraints that most developing countries do not.
The UtMOA is a complete monetary union that has evolved, since
independence in the early 1960s, from France's colonial governing
institutions in its African territories. Membership in the lUMOA is
fluid. Today, Benin, Burkina Faso, C8te d'Ivoire, Mali, Niger, Senegal,
and Togo belong. Mauritania left the union in 1973. Mali joined in
1984 and the possibility now exists that other, nonfrancophone
countries might join. "Completeness" involves having a common
.currency (the CFA franc) issued by a common central bank (Banque
Centrale des Etats de l'Afrique de l'Ouest [BCEAO]), and partially
pooled international reserves. The link with France is in the -form of a
fixed exchange rate with the French franc (CFAF 50 = F 1 since
1948), a guarantee of convertibility of the CFA franc by the French
Treasury, and French representation in the policy decisions and
operations of the BCEAO. (For a more detailed and analytical account
262 Richard BWundel;t Christopher Heady, and Rohinton Medhora
of the UMOA see Bhatia 1985; Guillaumont 1984; Medhora 1989;
Neurrisse 1987; Vinay 1980.)
Thus, other things (such as export price shocks and world interest
rate changes) being equal, COte d'Ivoire's membership in the UMOA
may give it certain advantages in dealing with crises. By not having
complete control over the common central bank, monetary policy
may be less subject to political pressure. By having access to its
partners' international reserves, country-specific short-term balance of
payments problems can be smoothed over. The foreign exchange
constraint is made even less onerous by the French guarantee of
convertibility: in practice a promise by the French Treasury to
augment, at a small cost and without explicitly stated conditions, the
union's pooled reserves maintained in an operations account in Paris.
However, none of these factors were enough to prevent C6te
d'Ivoire from sharing in the developing countries' economic malaise
of the late 1970s, leading to the use of World Bank and LM[F resources,
debt rescheduling, and being classified by the World Bank as a
severely indebted middle-income country.l We argue that the
UMOA's structure was not equipped to deal with the types of shocks
that its largest member faced, and that membership limited the policy
options available to implement structural adjustment programs.
By African standards, C6te d'Ivoire's postindependence economic
history has been enviable, with rising standards of liig and low-
inflation rates (table 71). An open trade regime, risk free foreign
investment - (due to the unchanged parity and guarantee of
co nvertibility), and free movement of factors of production within the
union made the C6te d'Ivoire paradigm receive much favorable
attention in textbooks on planning72
Through most of the 1970s, the level of -economic activity was
driven by growing public outays in the form of investment and
1. A severely indebted middle-income county is defined as one with a debt:GNP
ratio above 50 percent, a debt:exports ratio above 275 percent, an accrued debt
service:exports ratio above 30 percent, and an accrued interestexports ratio above 20
percent.
2. Devarajan and de Melo (1987) find thai. for the period 1960-82, and especially
for the subperiod 1973-82, the franc zone countries had GNP growth rates that
compare favorably to those of other countries, especially other SubSaharan African
countries. They attribute this to membership in the franc zone.
C6Ce d'Ivoire 263
Table 7.1 Basic Economic Indicators, 1970-87
GNP Public Real GDP Industial Reat
per capita investment Inflaiana (CFAF bilons, produucionb ijwaonefc
Year (cwrem USS) (CFAFbillions) (%) 1980 prices) (1985=100) (1986=100)
1970 270 n.a. 9.4 n.a. 24.0 n.a.
1975 500 6.0 11.4 1,491.7 47.8 58.4
1976 -580O 11.0 12.1 1,670.7 59.7 100.1
1977 670 64.4 27.4 1,749.3 69.5 207.6
1978 820 108.8 13.0 1,922.5 80.0 320.6
1979 980 78.5 16.6 2,022.4 82.3 259.7
1980 1,170 32.7 14.7 2,149.9 94.8 143.3
1981 1,130 23.1 8.8 2,179.9 95.7 113.8
1982 960 16.4 7.3 2,245.4 105.1 76.4
1983 760 15.4 5.9 2,045.6 89.5 72.3
1984 660 24.7 4.3 n.a. 97.0 78.8
1985 630 24.4 1.8 n.a. 100.0 76.8
1986 700 25.7 7.3 n.a. 108.1 100.0
1987 -750 13.3 0.4 n_a 108.1 78.6
n.a. = not available
a. The inflation rate is for the consumer price index rate of change.
b. The industrial production index includes the mining, manufaturing, and energy
sectors.
c. The reaI investment index excludes the petroleum refining sector.
Sources: GNP: World Bank (various years); investment: Lorch (1989); inflation and
industrial producdon; IMIF (various years).
government consumption and buoyant prices for the country's
principal, exports, coffee and cocoa. During the decade, government
consumption expenditures grew sixfold (table 7.2), and public
mvestment grew from CFAF 5.1 billion in 1974 to CFAF 78.5 billion
in 1979. Total real investment quadrupled between 1975-and 1979,-
industrial production quadrupled between 1970 and 1981, while real
GDP rose 40 percent between 1975 and 1981.
During the 1970s, C8te d'Ivoire shared in the boom. brought about
by high coffee and cocoa prices. Although some of the dollar gains in
these prices were tempered by the strength of the French franc against
the U.S. dollar, even in CFAF terms, export unit values had more than
264 Richard Blundel, Christopher Heady, and Rohinton Medhora
Table 7.2 Investment, Government Expenditure, and Trade, 1970-87
Gross fixed
capital Government Terms of Current
formation consumption trade account
Year (CFAF billions) (CFAF billions) (1985 = 100) (US$ millions)
1970 83.9 64.9 n.a. -37.9
1975 199.4 - 141.8 83.6 -379.0
1976 247.2 180.3 120.1 - -249.3
1977 397.7 209.7 151.8 -177.3
1978 529.0 290.4 126.4 -839.3
1979 526.7 353.8 123.4 -1,383.3
1980 523.6 362.4 104.3 -1,826.5
1981 558.4 403.6 88.1 -1,411.4
1982 539.S - 449.4 90.3 -1,017.3
1983 527.7 463.6 95.5 -931.2
1984 352.6 448.6 104.0 -58.2
1985 359.3 437.1 100.0 63.6
1986 386.2 491.7 108.0 -138.4
1987 n.a. n.a. 95.4 n.a.
na. = not available
Sources. Same as table 7.1.
tripled between 1970 and the peak year, 1977. To be sure, the oil
shocks of the 1970s had contributed to a rise in the price of imports,
but in 1977, CMte d'Ivoire's terms of trade stood at their highest
historic value (table 7.2).
The crunch began in 1978 with a collapse in the price of coffee,
followed, three years later, by a fall in cocoa prices. There was no
similar let-up in import prices or volume, so that the current account,
which had always been negative (but manageable), worsened
considerably between 1977 and 1980. By the early 1980s, the
international reserves accumulated during the boom years had fallen
sharply.
It is during times like these that membership in the UMOA is
supposed to ease the strain, via access to pooled reserves, and
ultimately, the French Treasury. However, given the circumstances,
C6te d'Ivoire 265
these arrangements could not have insulated C8te d'lvoire from
extemal events. The collapse in raw materials prices was universal, and
also affected the other members of the union, notably Senegal. So was
the rise in oil prices. With most members (and certainly the two
largest) in distress at the same time, reserve pooling, by its very nature,
will not work. Instead, in 1980 the overdraft facility with the French
Treasury was used for the first time. What had previously been a
guarantee on paper in the statutes goveming the BCEAO and its
relations with the French authorities was now being tested.
In principle, a cnsis management scheme is supposed to be
implemented when the BCEAO's external reserves fall for three
consecutive months to less than 20 percent of its short-term liabilities..
That threshold was crossed in mid-1980, at which point the statutes
require the governor of the BCEAO to meet with the Council of
Administration, review the situation, and take all appropriate measures;
however, what these measures should be is not spelled out Whether the
BCEAO has all the powers needed to act in a situation such as this is
not clear. Although the BCEAO's discussions with the French
authorities are usually confidential, there is little doubt that the lack of
explicitly stated conditionality on the use of the overdraft facility led
the French to advise GBte d'Ivoire to do what every other country in
its circumstance did: approach the World Bank and the IMF
Thus, membership in the UMOA can smooth out short-run,
isolated, country-specific external shocks. But in the face of a shock
common to all members, coupled with largely autonomous fiscal
policies, membership confers no special privileges on the countries.
However, membership does affect the implementation of a structural
adjustment program, as discussed next
Structural Adjustment
As stated earlier, World Banlk/IF-supervised strctural adjustment
in C8te d'Ivoire has seen roughly two phases. The first started in
1981, and the priority here was stabilization. The second evolved
during 1985, is ongoing, and has the longer-term goal of
restructuring, and ultimately, sustainable growth.
Stabilization meant curtailing public spending and depreciating the
real exchange rate, but without the aid of a nominal devaluation. This
266 Richard Blundell Christopher Heady, and Rohinson Medhora
makes the literature on devaluation-especially the debate on whether
a devaluation will be contradictory or not-a moot point (for a recent
review of the literature see the contributions of Axida and Taylor and
Edwards and van Wijnbergen to Chenery and Srinivasan 1989).
Rather, the discussion jumps to how else to stabilize. One possibility is
the "classical" adjustment process, where aggregate expenditure is
curtailed by sharp cuts in public spending. Another is to use fiscal
proxies-such as tariffs and subsidies-to mimic a devaluation. Cote
d'Ivoire has used both, with mixed results.
The monetary union and the free flow of capital within it precludes
interest rate increases in one country only. Instead, a credit ceiling was
imposed. Public gross fixed investment fell from 11.6 percent of GDP
in 1980 to 3.2 percent in 1985, and investment in paastatals returned
to preboom levels (Lorch 1989). Public sector and minimum wages
were also frozen.
Crucial help came exogenously in the form of the strength of the
U.S. dollar against the French franc between 1980 and 1984, as shown
in table 7.3. The net result was a real effective exchange rate
depreciation and an improved export performance. Between 1980 and
1984, the French franc halved in value with respect to the U.S. dollar,
which accounted for about three-quarters of the 40 percent fall in the
real effective exchange rate index.
The mild recovery in exports and production was reversed starting
in 1986. Cocoa prices fell once more, as did the U.S. dollar, the
-currency appreciated again, and the economy was thrown into another
recession.
Another feature of most structural adjustment programs is a
nominal devaluation. While much of the structuralist literature comes
out in favor of a tariff-subsidy scheme over nominal devaluation, this
system has certain practicaI limitations?
3. See Taylor (1981, 1983) for the potential effects of a contradictory
devaluation. Islam (1984) shows that a tariff-subsidy scheme is preferable to a
devaluation. Pegatienan Hicy's (1987) simulations show that a heterodox package
that includes import quotas, tariffs, and easy money is less deflationary than an
orthodox one that featnres absorption-reducing policies and tariff liberalization.
Laker (1981) reviews the literature on fiscal proxies for devaluation in a historical
context
s
Cdre d'Ivoire 267
First, tariffs and subsidies are almost never uniform and universal,
as would be necessary to mimic a devaluation perfectly. Capital
account transactions will be exempt, and a dual or more exchange rate
system is the more likely result. The danger of allocative
inefficiencies, loopholes, and corruption then anrses. Administering the
system can introduce further corruption and delays in payments to
-exporters. Finally, export subsidies during structural adjustment may
require a budgetary outlay at precisely the time when the government
budget is under pressure.
If despite this such a system is put in place and run efficiently,
there is always the risk that an exogenous event (such as French or
American policies that affect their bilateral exchange rate) will wipe
out-quickly and completely-the effects of taiffs and subsidies.
Table 7.3 Exchange Rates and Foreign Debt, 1970-87
Real effective Nominal effective
exchange exchange CFAF:$
Year rate index rate index index DebtlX
1970 n.a. n-a. 161.2 n.a.
1975 n.a. n.a. 208.2 n.a.
1976 a.a. 85.2 186.7 n.a.
1977 n.a. 83.8 181.4 n.a.
1978 123.3 86.8 197.8 n.a-
1979 135.7 92.4 209.6 . n.a.
1980 138.5 95.6 211.2 159.4
1981 118.7 88.3 164.9 226.7
1982 108.2 84.2 136.3 275.0
1983 104.1 87.7 117.5 307.4
1984 99.7 91.4 102.3 259.4
1985 100 100 100 306.1
1986 116.9 118.5 1289 301.9
1987 128.0 135.4 148.4 360.9
1988 127.2 147.6 149.9 415.8
n.a. = not available
Notes: Debt/X is the tatio of total external debt to exports of goods and services. For
first, second, third, and fourth columns, an increase implies appreciation.
Sources: Second, third, and fourth columns: lIMF (various years); debtcl World Bank
(various years).
268 Richard Blunddell Christopher Heady, and Rohinton MedJora
Since the introduction of the new trade regime in 1986, Cote
d'Ivoire seems to have undergone most of these problems. For
political and administrative reasons, as well as the country's
commitments to its regional trading partners, the government has not
reached its goal of a uniform tariff structure with 40 percent effective
protection for all industries. Some evidence of import fraud,
smuggling, and ad hoc exemptions for favored enterprises has
surfaced.
The export subsidy has not worked. To stay within the budget
deficit limits set during structural adjustment, pay-outs have been
limited to tariff revenues. As a result, ten firms received 75 percent of
the payments, and Lorch (1989) reports that their production,
investment, and export perform-ance have been worse than that of
nonrecipients. Moreover, the subsidy is not large enough to offset
high start-up costs, payment delays, and ultimately, the strength of the
CFA franc thanks to the weakness of the dollar. The average subsidy
of 13 percent of the value of exports in 1986 was more than offset by
the 20 percent depreciation of the U.S. dollar with respect to the
French fanc in that same year.
In sum, the new trade regimne has yet to prove itself, and so long as
exogenous exchange rate movements keep occurring, it may never get
a chance to do so. Nominal devaluation may be faster to work than a
tariff-subsidy scheme, but is unlikely under current conditions.
The government might consider other options as yet unused. A
foreign exchange "tax" may deter imports by driving up the cost of
foreign exchange for traders. However, it would not solve any of the
problems that export subsidy recipients face. In any case, how the rest
of the UMOA would react to such a move is unclear. A group
devaluation would necessarily be too much for some members and too
little for others.
The Urban Labor Market
Cote d'Lvoire is relatively highly urbanized for a country at its level
of per capita GNP (US$730 in 1986), with 45 percent of the
population in urban areas, of whom 34 percent live in the capital,
Abidjan. Despite this, 65 percent of the total labor force were engaged
7re d'lvoire 269
in agriculture in 1980, while 8 percent worked in industry and 27
percent in services (World Bank 1988).
The structure of the labor market in urban areas is fairly complex.
Fields (1989) found that in 1986, 38.6 percent were wage employees,
31.1 percent were self-employed in business, and 25.1 percent were
self-employed in agriculture.4 Of the employees, 44.6 percent were
employed i. the public sector, 28.8 percent were employed in the
formal private sector, and 45.5 percent were employed in the informal
private sector.5 Fields defines all self-employed workers as being in
the informal sector, while employees are included in the formal sector
if one of the following apply: a union is present in the work place;
wages are subject to minimum wage legislation, employees have a
formal employment contract; employees are entitled to paid holidays,
paid sick leave, a retirement pension scheme, or free or subsidized
medical care; or employees have access to social security benefits.
However, the term informal is often used in a less precise manner to
indicate activities that escape some or all taxes and government
regulation.
This complexity of the labor market is reflected in the need to use
different data sets to analyze different aspects of the market. On the
one hand, the Banque des Donnees Financieres has collected data on
modem manufacturing firms (formal sector manufacturing firms that
use modem technology) for a number of years. This data set has the
advantage of providing detailed data for a period that includes the
boom of the 1970s and the periods of stabilization and restructuring,
but its-disadvantage is that it excludes informal and nonmodern
manufacturing, services, and agriculture. On the other hand, data from
the Living Standards Measurement Survey (LSMS) includes all types
of employmient, but has only been collected since 1985, and was,
unfortunately, discontinued in 1989, just before the staggering 50 -
percent reduction in the procurement price of coffee. This is even
4. Some workers had more than one status, thus the percentage of the urban work
force who were in employment in the seven days before the survey is slightly less
tan the sum of these percentages. The remainder were not working.
5. These percentages add up to more than 100 because some people worked in
more than one sector.
270 Richard Blundell, Christopher Heady, and Rohinton Medihora
more critical for our discussion of rural labor markets and distribution
that follows later.
The Modern Sector
The structural adjustment program has affected the urban sector
the most. As most government expenditure and employment occurs in
urban areas, they have been most severely affected by the program's
stabilization phase. In addition, {he change in tariff structures and the
export subsidy scheme were designed to stimulate manufacturing
industries producing tradable goods, and these are concentrated in
urban areas.
Table 7.4 presents figures on employment in modern
manufacturing by industry group, excluding petroleum. The table
shows how total employment in modem manufacturing grew during-
the investment boom and Lhen fell dunng the structural adjustment
period. This fall in emp.oyment continued during the restructuring
despite the government's efforts to increase incentives for the
production of traded goods, most of which are produced within
manufacturing. This might have been due to C6te d'lvoire's
decreasing international competitiveness produced by the high value
of the CFA franc, or by firms' slow response to changed incentives.
Although the changes in incentives failed to raise employment in
manufacturing, one can obtain some idea of industry's responsiveness
to changed incentives by looking at the changes in each industry's
share of employment.
Two policies changed incentives within the manufacturing sector.
the restructuring of. protection and the export subsidy scheme. The
restructuring of protection started in 1985 and was aimed at
producing a uniform rate of effective protection. Lorch (1989)
reports that uniformity was not fully achieved, but a general
movement in that direction occurred. Food and textiles were both
heavily protected before 1985, and so suffered a reduction in
protection. Table 7.4 shows that both these industries experienced a
fall in their employment share after 1985. Chemical and rubber
products and mechanical and electrical goods were both receiving
average protection before 1985, and so were little affected by the
changes: the employment share of chemical and rubber producsa rose,
COte d'Ivoire 271
Table 7.4 Employment in Modern Manufacturing, 1974-87
(number of workers)
Sector 1974-75 1976-78 1979-81 1982-84 1985-86 1987
Food processing 22,278 30,957 40,840 40,632 38,451 35,384
(41.1%) (40.8%) (45.8%) (47.4%) (45.9%) (44.8%)
Textiles, garnents, etc. 8,041 13,901 14,866 14,359 13,897 12,709
t(14.%) (18.3%) (16.7%) (16.8%) (16.6%) (16.1%)
Wood procesing 8.632 9,958 8,984 7,314- 7,701 7,158
(15.9%) (13A1%) (10.1%) (85%) (9.2%) (9.1%)
Chemicals and rubber 7,246 9,104 11,692 12,783 14,275 15,058
(13A%) (11.9%) (13.1%) (14.9%) (17.0%) (19.1%)
Mechanical, electrical 5,662 8,533 9,030 .7,272 6,178 5.435
(105%) (11.2%) (10.1%) (8.5%) (7.4%) (6.9%)
Others -2,323 3,588 .3,793 3,339 3,335 3,188
(4.3%) (4.7%) (7.2%) (6.9%o) (7.0%) (7.9%)
Total 54,182 75,952 89,206 85,699 83,836 78,932
Notes: Employment in .petroleum is excluded.. Figures in parentieses are percentages
of total.
Source: Lorch (1989, table 9). Calculated from Banque des Donn6es Financi&res data.
while that of mechanical and electrical goods fell slightly- Finally,
wood processing and construction materials (an important part of
other industries) were b1oth lightly protected before 1985. They would
have benefited from increased protection, and table 7.4 shows that
both industries experienced an increase in employment share. These
findings are consistent with the view that firms responded to changes
in incentives, and that the labor market was sufficiently flexible to
reallocate labor.
However, thesc; changes in relative employment could have been
the result of other events. For example, the governmenW Thtroduced an
.export subsidy scheme in 1986 that lasted until 1988. This scheme
was intended to apply to all,exports to offset the anti-export bias of
import duties. The idea was to use import and export taxes to simulate
a devaluation in the face or the fixed exchange rates in the fianc zone-
However, Lorch (1989) reports that the subsidy was only implemented
272 Richard Blunde!, Christopher Heady, and Rohinton Medhora
selectively, and almost all the recipient firms were in four industries:
food, textiles, wood, and chemical and rubber products.
Table 7.4 also shows that employment share fell in both food and
textiles. These were both industries that had suffered from reduced
protection from imports. Wood processing experienced an increase in
employment share, but that could equally well have been due to the
improved protection from imports. However, chemical and rubber
products provide an example of an industry that did not benefit
greatly from tariff reform, but that did expand employment share.
Perhaps the export subsidy is part of the explanation for this.
This discussion has shown that the changes in employment share
are consistent with firms responding to changes in incentives. Another
piece of evidence supporting that view is provided by a more detailed
analysis of the Banque des Donnees Financieres data. Lorch (1989)
examined the data at the four-digit level and allocated each product
category into a high tradability group or low tradability gTOUp
depending on whether imports plus exports exceeded half domestic
production. He then compared the performance of the two groups.
His results are presented in table 7.5.
Table 7.5 shows that the high tradability group improved its
relative performance in every indicator after 1985,. despite the high
international value of the CFA franc This suggests that the efforts to
increase the resources devoted to the production of tradable goods
had some effect. However, data on the relative employment growth of
the two groups is not easily availab!e.
As mentioned above, the shortcoming of the Banque des Donndes
Financieres data is its limited coverage, which might produce
misleading results. For example, the reduction in modern
manufacturing employment might not imply a reduction in total
manufacturing employment if informal manufacturing enterprises
expanded rapidly. Thus, to examine overall employment in the urban
sector more fuly, we use the LSMS data. However, these data only
relate to the restructuring period, and the sample size is too small to
provide a detailed analysis of employment by industry.
The LSMS started in 1985 and is designed on a semipanel basis, so
that half the households interviewed one year are reinterviewed the
next. Ainsworth and Mufioz (1986) describe the survey's design,
C6:e d'Ivoire 273
'T'able 7.5 Relative Performance of High and Low Tradability Groups,
1979-88
(percent)
Growth rate 1979-81 1982-84 1985-88
Growth rate of number of firms
Low tradability 6.9 2.5 -4.5
High tradability 4.9 -0.7 -2.1
Nominal investment growth rate
Low tradability -38.0 -8.5 8.4
High tradability 5.0 0.8 19.2
Nominal value added growth rate
Low tradability 13.4 11.8 -4.7
High tradability 15.0 6.6 1.3
Nominal production growth rate-
Low tradability 14.7 10.1 1.3
High tradability 10.8 8.5 8.9
Real production growth rate
Low tradability 7.7 -1.6 0.8
High tradability 3.1 -1.6 5.7
Source: Lnrch (1989, table 18). Calculated from Banque des Donnmes Fmanci&res data
Newman (1987) presents the main labor ifiarket data for 1985, and
Fields (1989) provides a descriptive analysis of the data for 1985 and
1986. This analysis will concentrate on the people in the LSMS panel
for 1985186 (those people who were interviewed in both 1985 and
1986). After the removal of people under seven years of age and data
with reporting errors, we are left with 1,373 individuals. In 1985, 39
percent of these people worked in the 12 months before the survey.
Between 1985 and 1986, 21 percent of the labor force stopped
working and 17 percent joined the labor force, a net reduction of 4
percent If we look within the work force at the number of employees,
24 percent left and 17 percent joined. This reduction in employment
is consistent with the analysis of the -Banque des Donnees Financieres
data.
274 Richard Blundell, Christopher Heady, andRohinton Medhora
Table 7.6 shows the distribution of the labor force between wage
employment and the two types of self-employment for individuals
who worked in 1985 and 1986 (stayers), those who did not work in
1985 but did in 1986 (oiners), and those who worked in 1985 but not
in 1986 (leavers). The first two columns show a sliglht move from both
wage employment and self-employment in business toward self-
employment in agriculture for those who worked in both years.
However, the shift toward self-employment in agriculture is much
more obvious in the joiners (who disproportionately joined
agriculture) and the leavers (who disproportionately left business).
To examine interindustry movements between 1985 and 1986-
movements that may in part have been due to structural adjustment-
we have analyzed the industry of each person's main employment as
indicated in the LSMS- We have defined the main employment as the
main employment during the last 7 days, or for those not employed
during the last 7 days, the main employment during the past 12
months. This analysis is based on people in the panel.
The LSMS identifies 30 different industries. However, as, we are
interested mainly in analyzing structural adjustment and the
movement between the traded and nontraded sectors, we have
aggregated these industries into four groups: (a) agriculture, forestry,
and fishing; (b) manufacturing (presumed mainly tradable); (c)
services, commerce, and utilities (presumed mainly nontradable); and
Table 7.6 Employment Status of Stayers, Joiners, and Leavers
(percentage of labor force)
Stayers Joiners Leavers
Type of employment 1985 1986 1985-86 1985-86
Wage employment 39 .37 29 35
Self-employment in agriculture 25 29 39 21
Self-employment in business 36 34 32 44
Source: Living Standards Measurement Survey.
Cdied'lvoire 275
(d) other industry. The reasons for keeping other industry separate is
that we are not sure what this category includes as it expanded
enormously between 1985 and 1986.
The modem sector might have reacted differently to structural
adjustment than the informal sector- Thus, table 7.7 presents data on
the industrial distribution of employees and self-employed workers. It
shows a large apparent movement from manufacturing and services
into other industry among employees, the group that includes modem
sector workers. This could well be the result of coding changes rather
than real changes. The figures for the self-employed are not
contaminated by this possible error, and here we see a slight increase
in manufacturing employment. However, the numbers here are small.
Although the LSMS data relate to a relatively small number of
individuals, they provide an interesting addition to the Banque des
Donnmes Financieres data. First, they confrm the general reduction in
employment, particularly wage employment Second, the evidence of
growincr self-employment in manufacturing shows that the Banque des
Donnees Flnancieres data is becoming less representative of the urban
labor market as a whole.
Table 7.7 Industrial Composition by Employment Status, 1985 and
1986
(percent)
Employees Self-employed
Industry 1985 1986 1985 1986
Agriculture 3.8 1.0 45.1 45.1
Manufacturing 19.0 9.2 5.8 7.9
Services 74.9 60.0 47.9 46.7
Other 2.4 29.7 1.2 0.3
Source: Living Standards Measurement Survey.
276 Richard BlundeU, Christopher Heady, and Rohinton Medhora
Informalization, Self-Employment, and Nonagricultural Family
Enterprises
The LSMS data showed an increase in self-employment in
manufacturing despite a reduction in the number of wage employees
in the formal manufacturing sector. Lorch (1989) also observed this
move toward informal employmentL He used the Banque des Donnees
Financieres data to suggest a movement of firms from the formal to
the informal sector- He also proposes a number of reasons for this
growth in informal manufacturing. First, he suggests that informal
enterprses were less affected by the economic situation of the 1980s:
they were less likely to produce internationally traded goods, and so
suffered less from the period of overvaluation of the CFA franc; they
were never able to obtain cheap credit, and therefore did not suffer
from the reductic in credit availability; they had been less reliant on
pubHc investment projects for the sale of their products; and their
informal labor.contracts meant that they could cut wages instead of
employment Second, he suggests that many formal firms had an
incentive to become informal- Many of the advantages of formality,
such as cheap credit, had disappeared, while the advantages of
informality, particularly lower labor costs and avoidance of taxes and
regulations, remained. This incentive became particularly persuasive
for firms that were suffering financial distress as a result of the
structural adjustment program.
Much less data are available for informal firms than for formal
firms. However, the LSMS data include information on
nonagricultural, urban self-employment, and Vijverberg (1988) has
analyzed these data. In Abidjan, 2(0.7 percent of males and 62.7
percent of females in the labor force report nonagricultural self-
employment. The corresponding figures for other urban areas are
24.8 percent and 49.7 percent, respectively.
In Abidjan, these enterprises employ some paid workers, an average
of 1.59, but far fewer in other urban areas (only 0.19 on average).
The distribution, of enterprises between industry groups is very similar
in Abidjan and other urban areas: approximately 14 percent are in
manufacturing, 14 percent in services, 45 percent in food commerce,
and 27 percent in nonfood commerce. An important aspect of
C&te dlvoire 277
nonagricultural self-employment is that it provides substantial
employment for non-Ivorians. In Abidjan, 31.4 percent of Ivorian
workers were self-employed in nonagriculture, but the figure was 65.3
percent for non-Ivorians, including 31.1 percent of workers from
Burkino Faso and 75.8 percent of workers from Mali. The figures for
other urban areas are similar.
- -Wage Determination and Real Wage Growth
The available evidence indicates significant returns to training,
education, and experience within the urban sector during the period of
structural adjustment both for public and private sector workers. Table
7.8 documents the data from the 1985 LSMS. The traditional
schooling system in Cte d'Ivoire includes six years of elementary
school, four years of junior high school, three years of senior high
school, and a university program Table 7.8 presents the average years
of schooling for each level of education. Of the six or more years of
total schooling, only a very small part is university training, while most
of it is elementary education. The LSMS data allow differentiaton of
total experience into experience related to the current occupation and
other general experience. Occupation-specific experience is broader
than tenure on the current job, as it includes work experience in
previously held jobs that have the same job description-as the current
one.
Table 7.8 zveals that public sector employees are on average better
educated, with an average of 9.2 years of education, versus 5.3 years
in the private sector. In addition, more public sector employees hold
school diplomas. There are no non-Ivorians in the public sector, and
2.5.9 percent of the public sector labor force is female, compared to
14.9 percent in the private sector. Total experience (the sum of
general and specific experience), measured as age minus years spent
in school and technical training minus five (the age of which they
normally start school), averages about 20 years in both sectors.
Occupation-specific experience, however, is much lower in the private
than public sector.
Table 7.9 presents some estimates of retums. Note that while real
wages in the private sector have risen, the reverse is true for the public
sector. Non-Ivorians earn a higher wage in the private sector, as do
278 Richard Blundel4 Christopher Heady. wad Rohinion Medhora
women. However, the returns to basic education are higher in the
public sector. Occupational experience and higher levels of education
appear to yield a higher return in the private sector.
Table 7.8 Summary Statistics on Wages and Human Capital
Private sector Public sector
('N =301) (N 212)
Standard Standard
Symbol. Category Mean deviation Mean deviation
LNW Log of hourly wage rate (CPAs)a 5.557 1.29 6.577 0.99
General background
AGE Age in years 32.554 10.16 :35-565 8.70
GEXPIR General work experience 13.135 9.36 9.705 8.85
EUPOCC Occupation-specific experience 7.399 758 11.116 823
YRS-APP Years apprenticeship 1.166 2.20 0.241 1.08
YRS-TEC Yeas technical training 0.734 1.58 1.462 11.61
MR Reading, writing, and
arithmeticsdllsb . 1.973 137 2.637 094
NAT Non-lvorian % 0.275 0.44 0.000 0.00
FEMALE Female-% 0149 035 0.259 OA4
Years of schoo-ing
TRSCIHL Toal years of schooling 5.269 4.94 9.179 526
YRS-EL Years of elementary schooling 3.561 2.84 5.132 2.08
YRSn- Yearsjunior high school 1215 L69 2.472 1-80
YRS-H2 Years senior high school - 0.322 0.89 0.859 1.29
YRSUN Years university 0169 0.83 0.717 1.87
Diplomas obtained
DIP-El Elenternary school diploma % 0.478 0.50 0.830 038
DIP-Hl Junior high school diploma % 0.182 038 0.491 0.50
DIP-OPP Higher diploma - 0.089 0.28 0.236 0.43
DIP-TEC Technical diploma 16 0202 0.40 0.472 050
a. Wages are measured in CFAs; 50 CFA= 1 FE; I OFF = USS1 in 1985. The averages
reflect CFA 595 in the private sector and CFA 1,173 in the public sector.
'b. This index is zero for the completely illiterate and increases by 1 for cvery skill
aoqnired
Source: van der Gaag and Vijverberg (1989).
C6t,e d'Ivoire 279
Table 7.9 The Determination of Cross-Sector Wages
(a) Real wages
(1979 = 100) Public Private
1979 .100.0 100.0
1984 95.6 113.4
(b) Coefficient from log wage regressions
CONS]' 2.841 (7.87) 3.452 (14.65)
NAT (1 if non-Ivorian) n.a . 0.285 (2.20)
SEX ( 1 if female) -0.125 (-1.04) 0.141 (0.97)
DIP-EL . 0.801 (2.50) 0.395 (1.92)
DIP-H 0.424 (2.14) 0.617 (2.40)
DIP-UPP 0.621 (2.10) 0.221 (0.45)
DIP-TEC 0.002 (0.02) 0.031L (0.17)
YRS-II 0.205 (0.408) 0.012 (0.21)
YRS-UNI 0.206 (5.66) 0.300 (421)
YRS-TEC 0.036 (1.34) . 0.098 (2.42)
YRS-APP 0.067 (1.85) 0.008 (-0.31)
EXP-OCC . 0.087 (4.79) 0.116 (7.32)
n-a = not available
'Note: The figures in parenthesis are asynptoi- t-statistics. Symbols are those used in
table 7.8.
Source: van der Gaag and Vijverberg (1989).
Table 7.10 shows that during the stabilization period, when a
recession occurred, real wages were flexible downwari This simple
analysis at the occupational level, first identified by Lavy and Newman
(1989), points out how misleading aggregate data can be, since
although average real wages increased over this period, once they
adjusted for occupation, they showred that real wages fell substantially.
Moreover, Lavy and Newman (1989) found additional evidence that
new hires were paid very much less i.an. retained workers, although
even retained workers saw a shift down in their experience-wage
profile during this period.
280 RichardBlundell, ChristopherHeady; andRohintonMediora
Rural Labor Mark-ets
In contrast to the urban sector, the rural sector was not initially
subject to any direct effects of structural adjustment.-For example,
until very recently the procurement prices for coffee and cocoa had
hardly changed in real terms. The prices of other crops, particularly
subsistence crops, did vary, but as they are not under direct
government control, ascertaining whether these changes were a result
of the structural adjustment programa is difficult. Nevertheless, the
response to these price changes should indicate rural households'
overall ability to adapt to price reforms induced by structural
adjustment
More thain half the labor force works in rural areas, and agriculture
produces almost all the country's exports. Moreover, the recent
change in coffee prices suggests that the adjustment program will soon
lead to a closer Linkage between procurement prices and world prices..
Therefore, considering ths effects of price changes on labor supply in
the rural sector, and in agriculture in particular, is a worthwhile
exercise. We therefore present some results obtained by Alessie and
Table 7.10 Wage Flexbility During Stabilization, 1979 and 1984
I1979 . 1984
Average Average wage
Percentage wage Percentage deflated
Professional sharc of (CFAF 1,000 share of (CFAF 1,000
category employment per month) employment per month)
Director 1.8 345 2.4 384
Upper mnagement 3.7 341 5.4 305
Middle management 2.8 181 3.6 161
Technical employee 6.4 110 1 1.8 97
Skilled worker' 16.6 56 24.6 52
Unskilled worker 68.9 . 31. 56.0 28
Apprentice 0.2 20 0.2 17
Sourcec Lavy and Newman (1989).
COte d'lvoire 281
others (1990) on the price responsiveness of participation. in
agricultural work by young people. Since for young people the
choices are essentially between work, schooling, and migration, we
investigate the interplay between work and schooling.
Agricultural Enterprises and Household Production
Fields (1989) used LSMS data to show that in 1986, 86.2 percent
of rural workers were self-employed in agriculture, and as few as 1.4
percent were wage employees.6 The remainder of the workers were
self-employed in nonagriculture. Vijverberg (1988) demonstrated that
approximately 90 percent of these enterprises are in the same
household as an agricultural enterprise. The rural sector can therefore
be seen as consisting predominantly of household farms, frequently
combined with other small businesses, using very little outside labor.
In rural areas, the . small, nonagricultural enterprises provide
employment for around 10 percent of the women and 6 percent of the
men. Looking at it by country of origin, they provide employment for
some 6 percent of Ivorians and 23 percent of non-Ivorians (including
10 percent of those from Burkino Faso and 49 percent of those from
Mali). Of rural, nonagricultural enterprises, more than half are
engaged in food comnmerce and another quarter are engaged in other
commerce. Fewer than 15 percent are engaged in manufacturing.
Typically these enterprises do not employ any paid workers (the
average number of paid workers is 0Q12). They therefore share the
urban small enterprises' role of providing employment for non-
Ivorians Their industrial composition is also similar to the urban small
enterprises; howev-er, they provide employment for fewer people.
To investigate pnrce responsiveness and working pattems, we need'
to look closely at the activities &f those engaged in agricultmre.
Glewwe and de Tray (1988) assessed the proportions of farmers who
grow each of the main traded crops: 34.4 percent grow cocoa, 37.5
percent grow coffee, 31.6 percent grow rice, 16.0 percent grow oil
6. There are reasons to belicvc that the figure of 1A percent understates the
pmportion of outside workers. First, the sampling framei rural areas might well
underrepresent both wage workers and sharecroppers as these group often live
outside the viflages. Second, the questions in the LSMS do not always allow the
identification of sharecroppers.
282 Richard Blundell, Christoper Heady, and Rohinton Medkora
palm, 12.6 percent grow pineapples, 8.7 percent grow rubber, 2S2
percent grow coconut, and 2.2 percent grow sugar. Cotton is grown
almost exclusively in the northern savannah region, while the tree
crops are predominantly grown in the east and west forest regions. In
addition to growing cash crops, most farmers produce subsistence
crops such as cassava, yams, sweet potatoes, maize, millet, and
sorghum. Thus, separating out farmers who produce traded crops
from those who produce nontraded crops is impossible. The LSMS
provides data on the output composition of each farm, but the lags
-ivolved in changing the output mix, particularly when tree crops are
inv6lved, mean that changes cannot be observed in two years of data.
Inputs will respond much more rapidly than outputs, but the LSMS
does not ask how much time is devoted to the cultivation of each crop.
This means that we cannot determine whether resources. have been
reallocated toward the production of traded goods within the rural
sector. We can only ask whether structural adjustment resulted in an
increase in labor supply to agriculture.
Schooling and Work in Agricultural Household EntepjLrises
Table 7.11 presents, for each year of LSMS panel data, the
proportion of individuals who report having worked in Ihe preceding
12 months by age group. The general picture is that between the ages
of 24 and 60, roughly 95 percent of individuals repor't some labor
market activity. During the two years, overall reportedi labor market
activity increased. This- increase was conceutrated amo?ng those aget
groups with the lowest percentage of workers, namely) among those
aged under 12 and over 60. A somewhat cunrous feature is that we
observe a slight fall in work activity for the 12 to 18 age; group.
The above description of the data has concentrated -ctn individuals'
rates of participation. However, to gauge price responsiveness we are
interested in movements of individuals between the states of working
and not working. Table 7.12 shows the numbers and proportion of
individuals according to their working status in each year. The general
pattern- of labor market transitions shows that between" the ages of 19
and 60 the numbers of individuals leaving or entering!the work force
are very small. The vast majority of changes in working status are
concentrated among those individuals aged 18 and under. Since work
CO1e d'Ivoire 283
Table 7.11 Rural Labor Force Participation, 1985 and 1986
(percentage of panel sample)
Percentage of panel sample
Age group Number of
(1985) 1985 1986 observations
7-11 22.5 1.3 7
12-18 71.9 69.8
19-23 8i6.2 90.0 130
24-30 94.2 96-7 154
31-40 96.8. 96.7 217
41-50 97.4 96.1 230
51-60 95S1 94.5 182
Over 60 66.5 73.2- 209
Total 71.4 76.0 1,930
Source: Living Standards Measurement Survey.
Table 7.12 Work Status Transitions, by Age, 1985 and 1986
(number)
Age group Do not work Worked 1985 Worked 1986 Worked both
(1985) either year only only years
7-11 202 43 -121 51
(48-4) (10.3) (29_0) (12.2)
12-18 53 65 57 216
(13.6) (16.6) (14.6) (55.2)
19-23 9 4 9 108
(6.9) (3.1) (6.9) (83.1)
.24-30 3 2 6 143
(2.0) (2.3) (3.9) (92.9)
31-40 4 4 -3 206
-(1.8) (1.8) (1.4) (94.9)
41-50 3 6 3 218
(1.3) (2.6) (13) (94.8)
51-60 5 5 4 168
(2.8) (2.8) (2.2) (92.3)
Over 60 45 11 25 128
(21.5) (5.3) (12.0) (61.2)
Total 324 140 228 1,238
(16.8) . (7.3) (11.8) (64.2)
Note: Figures in parauthesis are row percentages.
'Source: Living Standards Measurement Survey.
284 Riciard BRlndell, Christopher Heady, and Rohinton Medhora
activity for this group is likely to result in reduced schooling, we turn-
briefly to discussion of the schooling system in Cote d'lvoire.
.The school system, inherited under colonial rule, follows the
French system.7 Six years of elementary education lead to the
Certificat d'Etudes Primaires (CEPE), awarded on the basis of a
nationwide examination. The CEPE is a prerequisite for entrance to
secondary school, although because secorndary school places are
scarce, the score required to gain entrance is often higher than that
required to obtain the CEPE certificate. Four years of lower secondary
education lead to the Brevet d'Etudes du Premier Cycle (BEPC),
which, if successfully completed, allows the student to enter three years
of uppe.r secondary education leading to the Baccalaureate.
Alternatively, those students who have a CEPE or who successfully
complete some or all of their lower secondary education can enter
various training programs. One feature of the Ivorian education
system is that many students at all levels repeat grades, thus the'
number of years for which students are enrolled will on average
exceed the numbers outlined above.
Table 713 shows for our sample the percentage of individuals who
attended school in 1985 and 1986 broken down by age and sex.
Overall, the table shows a steady fal in attendance between the ages of
12 and 16, with a distinct drop for those aged 17 or over in 1985. As
we might have expected, school attendance falls for all age groups
from 1985 to 1986. Looking at the pattern of attendance across the
sexes, attendance rates are significantly lower for females than for
males. Note that of our sample, no -18-year-old females attended
school.
A variety of factors may influence households' decisions on
whether to send children to schooL One such factor is the distance that
students must travel, since time and travel costs will increase the further
students have to travel to attend school. However, pnmary schools in
the Cte d'dIvoire tend to be neighborhood schools. Some 85 percent
of our sample have primary schools located close by, while the
remainder are all within eight kilomete.s of the nearest primary
7. We are grateful to Paul Glewwe for providing information on Cate d'Ivoire's
education system.
Cdle d'lvoire 285
Table 7.13 School Attendance, 1985.and 1986
Both sexes Boys Girls
(percentage of (percentage of (percentage of
panel sample) panel sample) panel sample) Number of observations
Age
(1985) 1985. 1986 1985 1986 1985 1986 Totawi BOys Girls
12 68.4 64.5 73.1 67.3 58.3 58.3 76 52 24.
-13 57.7 53.9 66.7 64.4 45.5 39.4. 78 45 33
14 - 40.0 36.5 5.0 44.4 18.8 18.8 52 36 16
15 29.1 25.5 35.7 32.1 22.2 18.5 55 28 27
16 29.3 21i.9 36.0 28.0 18.8 12 5 41 25 16
17: 12.8 1 0.3 16.7 12.5 6.7 6.7 39 24 15
18 12.0 10.0 20.0 16.7 0.0 0.0 50 30 20
: Total 40.2 36.3 47.9 43.3 27.8 25.2 391. 240 151
school. This is not the case for secondary schools,.which are located in
urban areas. Of our sample only 10 percent live within 10 kilometers
of a. secondary school and some 60 percent live more than 20
kilometers from a secondary school, with='a mean distance from a
secondary school of 27 kilometers. One effect of the distance to
secondary school is the phenomenon of "child fostering" found- in-
West Africa. Children commonly live with friends or relations, so that
they can attend school (Ainsworth1989).8
.,' + The vast majori,ty of. students in Cote d'Ivoire attend public
schools, which do not charge tuition.; Some private schools are
available, but most are located in urban areas.. Table 7.14 breaks' down
school attendance by type of school.. The table shows that within our
sample? some 20 percent of individuals aged 13 to 16 r'emained in
education after obtaining their CEPE certificate in 1985.
Turning our attention to .work activity, table 7.15. shows that the
- slight fall in work activity is'-concentrated among those individuals
aged 13 to 16 in. 1985. In general, a .higher percentage of females,
than males are engaged in work activity. -
8. Such "foslered" children are probably not included in the rral panel sample.-
286 Rcicard Blundell, Christopher Heady, and RoJinton Medhtora
Table 7.14 Type of School Attended, 1985 and 1986
Percentage of panel sample
Primary Lower. Higlher Number of
Yeartage None school secondary secondary observationsX
1985
12 31.6 59.2 9.2 - 76
13 42.3 44.9 12.8 - 78
14 59.6 21.2- 19.2 - 52
1i 70.9 9.1 20.0 - 55
16. 70.7 -7.3 22.0 - 41
17 87.2 2.6 5.1 5.1 39
18 - 88.0 - 2.0 6.0 4.0 5 0
1986-
12 35.5 44.7 19.7 - 76
13 46.2 37.2 16.7 - 78
14 63.5 15.4 21.2 - 52
15 . 74.6 5.5 20.0 . 55
;16' .78.1 . 2.4 19.5 - 41
17 . 89.7 2.6 - 2.6 5.1 39
18 90.0 2.0 . 4.0 4.0 50
=- not applicable..
Note: The survey does not ask what type of school panel individuals attend. However,
since obtaining a CEPE diploma is a prerequisite for attending lower secondary school-
and a BEPC diploma is a prerequisite for higher secondary school, we assume that
individuals holding diplomas attend the appropriate establishment.
Source: Living Standards Measurement Survey.
So far we have only considered our sample's rates of participation
in work and school. We are, however, primarily concerned with
examining movements of individuals between the states of working
and not,working and attending,school and not attending school. In
particular' we- wish to examine the relatiornship between the two
decisions since, to the extent that -work activity may reduce school
attendance, the two are closely related. Table 7.16 cross-tabulates the
number of individuals in -our sample according to their- work and
education status in. both years. Within the- sample, nearly 60 percent
did not attend school in either year. In terms of work status -transitions,
COte d'Jvoire 287
Table 7.15 Individuals 'Working, 1985 and 1986
Percentage of pane! sample
Bollh sexes Males- Females Number of observations
Age
(1 985) 1985 .1986, 1985 1986 1985 -1986 Tota! Mates Females
12 46.1 48.7 46.1 51.9 .59.3 41.7 76' 52 24
13 64.1 59.0 .64.4 66.7 63.6 48.5 78 45 33
14 73.1 59.6 66.7 52.8 87.5 75.0 52 3 6 :16
1 5 87.2 83.6 89.3 85.7 85.2 81.5 55 2 8 27.
1 6 82.9 .80.5 84.0 80.0 81.3 81.3 41 25 16
17 84.6 92.5 83.3 87.5 '86.7 100.0 39 24 -15.
18 86.0 88.0 86.7- 80.0 .85.0 100.0 50 3 0 20
Total 7119 69.8 69.2 68.7 76.2 71.5 391 .240 151
Table 7.16 Cross-Tabulation of Work and'School. Status of Sample,
1985 and 1986
Work Work- -Work~ Not work
Status ~~~~both years 1985 ontly 1986 only either year -Total
School bath years 20_ 728413
School 1985aonly 6 0 89 4 i8
School1986aonly. 1 2; 0 0 3
Not attend either year 189 16 21 5 231
Total 216 65 57 53 391
15 percent entering the work force ina 1986 were recorded as:not
working n 1985. ore itrsing, nearly 20 percent of.the sample
left work in 1986. Ofthose individuals wh'o changed their wOrking
status, the majority did so'while attending school in both years.
288 Richard BMundell, Christopher Heady, and Rohlinton Medhora
Labor Supply and Agricultural Prices-
Most households are -engaged in agricultural production, therefore,
analyzing labor supply decisions in the context of agricultural
household models is appropriate (see Singh and others 1986). These
models represent behavior as the result of collective household
decisions about production, labor supply, and -consumption. In
general, these decisions are interdependent and.. must be modeled
simultaneously. However, under the assumrption of complete
competitive factor and product markets, the decisions have a separable -
or recursive structure. Production- decisions are made on the basis of
profit maximization, as a result of which the household receives profit
income. The household labor supply and consumption decisions are
then* made to maximize collective utility, given' consumer prices,
wages, and profit income (appendix A provides a formal basis for the
following discussion).
In the "separable" model, agricultural producer prices will only
affect household labor supply through their effect on profit income.
Increased output prices will raise profits, and thus (assuming 'that.
leisure is a normal go6d) reduce household labor supply. This
reduction in household labor supply is consistent with the production
decision to increase output and labor input in response to increased
prices: the difference between labor input and household labor supply
is covered by labor market transactions.
Possible 'reasonus for allowing a more general specification than the
separable model will be discussed later, but first we' must consider how
the separable model would allow analysis of the incentive effects of
structural adjustment. The discussion in the previous paragraph
suggests' that increases in agricultural prices will reduce 'household
labor supply, but this was based on the implicit assumption that wages
were constant.- In reality, one would 'expect the increased demand for
labor that follows from increased agricultural prices to result in
increased wages. These increased wages will have both income effects
and -substitution effects. The substitution effects will increase labor
supply, while the direction of the income effect will, depend on
whether the household is-a net buyer or net seller of labor. In either
case, the income effect wil,be-proportional to the extent of net labor
CCte d'lvoire 289
purchases (or sales), and so will be small for the typical Ivorian farm
household, which buys or sells relatively small quantities of labor.
Thus, for farm households the substitution effect of wage increases
can be expected to dominate their income effect and provide the main
route through which increased agricultural prices might stimulate
labor supply.
This:argument suggests that the effect of output price changes on-
labor supply cannot be captured by simply looling at a model that
takes the wage rate as an exogenous variable. This cannot capture the-
effect of product prices on wages, and so omits thef only way in which
increased product prices can increase labor supply. The analysis5-
requires the addition of a relationship between producer prices and the
wage rate. This relationship should include the' producer pnces of all
crops produced by labor from the local labor market. in order to
reflect the role of piices in determining the. demand for labor.
However, as many of the crops have long periods of production, the
demand for current labor will depend in part on. expected future
prices for the product. Ideally, therefore, the expectations of future
prices should also be included.
Although "structural"- estimation would re'veal- more detailed
information about household behavior and labor market; response, the
estimation of- a reduced form, which combines the wage and labor-
supply relationships, can provide the answer to our main ;question:
does labor supply respond to structural adjustment? This can also be
estimated without using data on market wage rates; a considerable
advantage in Cote d'Ivoire as few data on rural wage rates are'
available, andAthose available are of doubtful quality. Finally, as shown
below, the distinctions in the structural model become blurred as soon
as the assumption of complete (and perfect) markets is relaxed.
The separable model is based on the. assumption of complete
perfect markets. The labor market might violate this assumption in two
main ways. First, the labor market may be very thin, or. even
nonexistent, so that households may have difficulty in either buying
or-selling labor. Second, hired labor may not be a perfect substitute
for household labor because of greater supervision-needs.
With regard to the first violation, our data -suggest that very little
labor market activity takes place: landless laborers frequently obtain
290 Richard Blundell, Christopher Heady, and Rotintan Med/ora
access to land through sharecropping rather than wage labor. On the
second issue, obtaining empirical evidence about the substitutability
between household labor and hired labor is much more difficult, but
the possibility of the two types of labor being less than perifect
substitutes has some plausibility.
The implication of both possible violations is that household labor.
supply does not depend on the market wage, but on a shadow wage
that reflects the marginal value of extra household labor. This shadow
wage will be influenced by the market wage, if a labor market exists,
but will also depend on'factors that influence the supply and demand
for labor within the household. On the demand side, output prices and
the amount of . available land' would be Xexpected to increase the-
shadow wage. On the supply side, -an increase in 'the :number of
household members who are prepared to work will reduce the shadow
wage.
Note that fairm profits in these circumstances are no longer well
defined because of the difficulty of valuing household labor. One can
define a shadow profit, based on the shadow wage, and its value will
depend on the variables already included..Thus, there is no need to
include a separate term for profits in the reduced form equation and
we shall simply incorporate nonfarm income.
The producer price index should now represent the prices (ideally
including expected future prices) of -products produced by. this
particular household, because it.is those prces that determine the
demand for- labor by this particular household. This suggests that each
household should have its own producer price index with weights that
reflect the relative importance of the. different crops that it produces.
There, is, therefore, the possibility of considerable variation in.
incentive changes between households. Our research makes use of this
variability by constructing household-specific Divisia price indices
(see Alessie and others, 1990) that reflect the relative importance of
different crops in each household's total output.
We derive separate composite 'price indices for *both .'gross
production and the production of cash crops. Our app'roach to the
construction of both indices is as follows. First, we .calculate; for each
COte d'lvoire 291
year average regional prices for each crop.9 These average prices are
then weighted according to the share of each crop in the individual
household's production in 1985, thus forming household-specific
composite indices of price changes based on first period weights. in,
calculating. these price variables we look only at households located in
rural areas, which are in turn divided into five regions.10
Having calculated average regional prices, we derive composite
Divisia price indices for each agricultural household 'h' using the-
following formula:
P --p wk Aen pi
where w?, = first period value share of curmmodity i in total gross
(cash or gross) crop production for household h.
As equation (1) shows, whereas the value shares are household
specific, the price ratios vary only with region. As mentioned above,
we calculate two price indices thitt vary according to the value shares
in production by whichi they are, weighted. For the gross production'
index (FRODIND) the value of production is defined as the amount
of the harvest sold, plus the value of replacement capital such as seeds,
9. The five regions are: (a) NORTH, north of Kassou Lake; (b) SW1, between Buyo
Reservoir and the Guinean and Liberian borders; (c) SW2 cenitral southern area to the:
southwest of Kassou Lake; (d) SEI, north of Abengourou; (e) SE2, Abidjan hinterland
to the south of Abengourou.
10..The LSMS collects information on the quantities sold and the prce received
for a total of 22 crops. In the case of nine of these crops it is not possible to construct
crop price indices either because the value share of the crop ;n total production was
rather small, or because insufficient price information was available because the price
of the crop was generally reported in nornmetric units. These crops are rubber, coconut
,palm, wood, tobacco, pineapple, sugarcane, taro, sweet potato, and millet. The
remaining,crops used in the construction of our-price indices are as folloaws:
1. Cocoa 4; Plantain 7. Cotton 10. Yam 13. Vegetables
2. Coffee 5. Fruit trees 8. Peanut 11. Maize
.3. Oil palm 6. Cola nut 9. Cassava 12. Rice
For the above crops we calculate average regional prices for,five regions. In
calculating average prices for each-crop we exclude those. observations where the
quantities sold were reported in nonmetric units, and where it was therefore not
possible,. to calculate metric unit prices. Also excluded were some outlier
observations. Selection of outliers is based upon a 95 percent confidence interval
rule. Note that we were unable to take account ofany seasonal variation..in prices
because we do not know at what time of year crops were harvested and sold.
.292 Richard Blundell, Chtristopher Heady, and Rob;inton Medhora
the value of any harvest given away, and the value of horne
production. For the cash crop index (CASHIND), the value of net cash
crop production equals the value of cash crops, where cash crops are
defined as coffee, cocoa, cotton, and cola nut.11 Constructing this-
second index was necessary because of the wide dispersion in prices of
subsistence crops observed within- regions (this dispersion is due, in
part, to the fact that we have few observations for some subsistence.
crops). For cash crops, however, prices are generally controlled,:and-
hence more uniform, thus yielding a more reliable crop price index.
Note that in calculating these idices we include only the value of
primary production. We are unable to include the value of processed
itenms as insufficient price information is available, and similarly we do
not-include animal products within the indices...
Table 7.17 shows the regional mean values of the price change: ..
indices. It indicates clearly that the prices of cash crops do6 not change
by as much as the prices~ for gross production. With the exception of
the north region, the gross production index shows a larger increase
than the cash crop index, indicating an increase in the relative prices-
Table 7.17 Regional Mean Price Indices of Production
Region
Category: Total North SWI SW2 SEl SE2
Gross production PRODIND 0.29 -0.17 0.23 0.45 0.76 0.27'-
- -- (0.42) (0.27) (0.19) (0.27) (0.39) (0.30)
CashproductionCASHIND 0.03 0.01 0.04 0.03 0.03 0.03
(0.02) (0.01) (0.02) (0.02) (0.03) - (0.01)-
Note: Standard deviations in parentheses.
11. Details of the LSMS questions used in the construction of the price weights are
as follows: the value of net production is taken to be the value of crops sold (Sec 9B
Q4,5); the value of gross production is the sum of net production plus the value of any
seeds retained (Sec 9b Q7); tie value of any crops exchangtA. in retlrn for labor (for ali
goods except coffee and osotwa Sec 9B 09); and the value of arny crops used for home
coisumption (Sec 128 03-5).
C6te d'voire 293
of subsistence crops. This means that incentive changes, in agriculture
have favored nontraded goods, the opposite of what is required for.
structural adjustment. The difference in the cash -crop -index observed
in the north region is to an extent explained by the fact that -the
government raised the procurement price for cocoa and coffee not
grown within the region.
Labor Force Transitions and Agricultural Price Changes
In this section -we estimate the determinants of transitions in and out:
of the labor market using the model and price indices just descrbed.
Empirical studies of individual participation in work are usually
restricted to the use of cross-sectional data due to the lack of panel
data in developing countries. Alessie. and others' (1990) study was
therefore in the almost- unique position of being able to exploit the
repeated observations on eacht individual across pairs of years in the -
Cote d'Ivoire panel survey. This enabled us to assess directIy the.
degree of history. dependence in work behavior, as. well as the benefits.
of panel data over cross-sectional data in-investigating such behavior.
As many decisions are based on ionger horizons than two years, the
models we consider, based on those discussed above, attempt to exploit
the richness of the cross-sectional. data to. capture long-term
influences, : leaving the panel data to provide information on short-run.
transitions and history dependence.
The cross-sectional data in the' Cte d'ivoire survey provide ideal:
instmments for. measuring longer-mn influences. For example, we can
use the consumption expeniditure records as a measure of life-cycle
income or wealth of the household, unit. Moreover,' the effect of local
constraints on behavior can be- captured via the. extensive information
on infrastructure variables. Equally, the short-run effects -of -price
changes, a central' focus in this. research, can be identified through the
cash crop (CASHIND) and total product (PRODIND) price indices.
constructed for each household as described earlier. The empirical
models presented-in this section will.therefore attempt to exploit all the
cross-sectional information that 'may explain the work status of young
people living in rural- households: Through the use of the panel, we
can also assess the importance of price changes not easily identified in
cross-sectional analysis.
294 Richard Blundeli, Christopher iHeady, and Rohinton Med/iora
For the individuals in the sample, the initial period's work status
can take* three easily recognizable states that relate directly to 'the
sample split described in table 7.16. First, we may consider the
likelihood of an individual bein'g in work conditional on having
worked in the first period. In table 7.16, this would refer to the sum of
the first two columns (216 + 65), which shows that of the 281 young
people in our panel who worked in 1985, .216 were engaged in some
work in 1986. From these we niay split 'out-the 73 (20+47+6) who
were engaged in some education in the first period. A comparison of
these tWo groups will enable us to assess whether being involved in-
some educatior in the initial period influences the decision to keep-
working in the next period. This is an example of the type of history
dependence that is critical to the issues under'anhlysis. In particular,
we will wish to know whether the reactions to price incentives across
these two groups differ. Finally, we may consider those individuals
who move into work from full-time education in the first period. In
summary, by conditioning on past period behavior, we can assess the
irportance of history dependence in individual work behavior. This
modeling approach is detailed in appendix B (see also. Hckman
1981;' Nakamura and Nakamura 1985).-
Our results in this section describe a binary model of current work
status conditiona on past work and schooling behavior.12 This should
be interpreted as a reduced form transition' model since we, do not
estimate the. direct effects of current period schooling behavior on
work status or the effects of current period decisions over work and
schooling 'by other household members.
For the cash crop, price change variable CASHIND, we might
expect a strong positive effect on the probability of work in the
second period if incentives are having a strong effect on work.
activities. However, given the reduced form nature of this equation, the
12. There are; in principle, two primary measures of work status 'available in the
panel survey. These relate to working behavior in the past week or during the past
year. In our studies we use both measures, but in this report we present results for the
latter definition alone as these were not only similar to those for the' past week
definition, but are generally more precisely determined This is encouraging since it
suggests our results are not unduly affectd by seasoaal 'work patterns. The results.
presented here also use a similar past-twelve-month measure for sicooling behavior.
C6te d'lvoire 295
cash crop price variable CASHIND will also capture'the income effects
of an output price change.
*Folowing the arguments detailed earlier, we attempt to capture a
longer-run or life-cycle measure of other incomne and wealth with the
use of consumption expenditures. Again, this is likely to be a current
period endogenous varable, so to allow for such endogeneity in this
reduced form model, we use a consumption measure from -the first
survey, FOOD(Q). In turn, this is restricted-to cover food expenditures
only, since they appear to not only make up the largest share of a
household's budget, but. also appear to be the best measured
consumption item. These expenditures;include the value of home.
production.- In the conditional probability models of state dependence
used here, these income and other household decision variables in the
first period will only'be important insofar-as they capture longer-run
effects not captured by the initial -period work status variables.-
The results for the subsample of individuals who ate working
during the first period are presented in table 7.18 The descriptive
statistics relatingAto each variable' and further details of variables used
are provided in 'Alessie and others (1990). The precision and sign of
the CASHIND variable are comforting.0 In al 'the results set down here,
we have tried to present a reasonably parsimonious parameterization.
of each model "spbcification. Indeed, a general feature of our results
was the dominating importance of -the cash. crop price' effects
.measured through CASHIND over the general index PRODIND. This
probably reflects 'possible measurement - problems in the prices of
noncash crops.'
Turning to.the other'factors; we started with household variables
that might be important in' determining this conditional probability. In'
particular, we found the individual's age (AGE) and the number of
other household woakers (OTHWK)'were of some imnportance. Since
'we. are deliberately not attempting to model the simultaneous' work
status decisions of- all 'family members at this stage, we use the number
of other household workers as recorded in the first survey denoted by
(1). A variable measuring land size was also included in preliminary
models, but this was never found'to play a role once regional effects
were allowed for. As :mentioned earlier,= initial period food
*.expenditures FOOD(l) were added to capture any longer-run income
296 RkctIard Blundell, Clristopher Heady, and Roh intoi Medhtora
Table 7.18 State Depenldent Work. Behavior
Standard
Variable Coeffident error' T-statistic P-value
CONSTANT -2.579169 0.971435 -2.655008 0.007931
CASHIND 15.965741 5.782105 2.761233 0.005758
AGE 0.354416 0.059647 5.941882 0.000000
OTHWKQI) -0.085246 -0.048298 -1.765000 0.077564
FOOD(l) -0,450348 0.255106 -1.765341 0.077506.
RELIG 0.428714 0377991 1.134193 0.256714
REGDSWI -1.999087 0.591587 -3379194 0.000727
REGDSW2,. -2.353115 0.516375 -4.556985 :,000005
REGDSEI -1.406769 0.568329 -2.475273 - 0.013313
REGDSE2 -1.256331 : 0596177 -2.107313 0.035090
Notes: obs = 281, logL = -103.235.
effects not reflected .in the discrete state conditioning' variable. The'.
negative coefficient.confirms. such an effect., In addition to. these
factors we felt the individual's religion may be important, and In table
7.18 we have retained the coefficient on RELIG, a dummy variable
that is unity if the individual, is:Muslim.
Location variables could play an important role in both shaping
preferences and identifying constraints. The regional dummies
(REGDSWI-REG3DSE2) are clearly, important,' although some further-'
grouping looks possible. Finally, we tried a number of infrastructure
variables and a variable indicating the sex of the -individual. These-
turned out to be of little significance for this transition.
We argued above that it was of some interest to analyze the
behavior, of the subset of 73 individuals from this group who had-not
only been working in the first period, but who had also -been in
education during the fist.year. Table 7.19 reports the results for this
subgroup. These show a similar overall pattern, but an even stronger
cash crop price effect. Note that in this. smaller sample, we were able to
group the dummies for SW regions with the base region. We.also
considered.the group of individuals who were wholly engaged in
MCOe d'Ilpotre 297
Table 7.19 Working Belhavior for Those- Engaged in Some Education
- - - . ~~~~~~~~Staetait vIo
Vtarifble C'aeJjlatI . - error T-s:aitLuIk P-vale
CONSTANT -2.95884B 2,099278 -1,409460 0.163459
CASI1IND 38,182585 11,462775 3+331007 0.001422
AGE1 01.83416 0.142478 1.287326 0.202540
OTHWK(I) -0.165548 0,079879 -2.072476 0.042184
FOOD(I) -1.026199 0.618155 -1.660099 0.101705
RELIG 1.362558 0.622039 2.187657 0,032289
REGDSEI 0.039072 0.526423 0.074222: 0,941061
REOCDSI2 1.220068 0.513474 2,376103 0,020445
Notes: obs = 73, -logL = 33.395.
Table 7.20 The Exit From Full-time Education to Work
Standard
Variable Coefficient errr r-stistic P-vatue
CONSTANT 1.714820 1.629952 1.052067 0.296187
CASHIND - 14.462155. 9.713291 -1.488904 0.140756
AGE -0.016953 0.102740 -0.165006 -- 0.869388
OTH4WK(I) -0.186352 : 0.081931 -2.274508 0.025827
FOOD(I) 0.915047 0.652827 1-401668 0.165189
RELIG -0.533052 0.458154 -1.163477 0.248368
REGDSWI -0.130967 - 0.913481 -0.143371 0.886386
REGDSW2 -0.942830 0.862473 -1.093171 0.277859
REGDSEI -0.645450 0.894543 -0.721542 0472846
REGDSE2 -1.884299 0.936895 -2.011217 0.047938
Notes: obs = 84, -logL = 40.655.-
education in the first period. Results for this group are presented in
table 7.20, and contrast distinctly with those discussed so far. The cash
-crop, price effect is negative, suggesting a more dominant income
298 Ricaiard Blundtoll, Chiristopiher Handy, and Rolisiton! MAedhora
effect of price movements for this group of Individuals. These results
show that labor force participation'is generally. responsive to price,
Incentives. However, people with different work and, education
histories respond differently.
Distribution, the Role of Women, and Migration
This section consideis the.effects that structural adjustment has had
on the distribution.of income, the role of women, and ,the migration
between urban 'and rur'al'areas.
Income DistributiOn and Poverty
The collection of the first year of LSMS data in 1985 .constituted
'the first attempt. to provide a comprehensive measure of inconie'
distribution and poverty-: in' C6te d'Ivoire.l Berthelemy and
*Bourgignon (1989, p. 20) 'mention some estimates based on partial
evidence'in.the 1970s, but they are not comparable with the Gini
coefficient of 0.44 for consumption expenditures per adult equivalent.
that Glewwe (1987) calculated from the. LSMS data. Thus, arriving at
.any definite conclusions about the effects of structural adjustment on
income distribution and poverty. is impossible. The' an.dlysis is.
necessarily speculative, 'using data from 1985 and later to infer what.
the effects of structural adjustment are likely to have been.
Glewwe-and de Tray (1988) and Kanbur (1989) analyze ihe likely
effects of structural adjustment on poverty.. Glewwe and de Tray show
that 92.3 percent of the poorest decile and 87.5 percent.of'the. poorest
three deciles are in households whose head is engaged in agriculture.
Thus, most poverty is in agriculture, where-there has been'little change
in real procurement prices for cash 'crops. As shown in the- previous
section, the' prices of subsistence crops in areas other than the north
increased between. 1985 and 1986; however, whether these changes
were part of. a price trend or a.'result of the 'structural adjustment
policies is not clear. Much of the analysis is therefore in' terms of the:
consequences of possible future policies rather than the 'effects 'of
actual policies that have been implemented.
Regional factors are very. important in Cote d'Ivoire, particularly
for poverty. The most important distinction is between the east and
west forest 'areas 'on the one. hand, and: the northern' savannah on the
Coie d'Ivoire 299,
other. The forest regions grow cocoa and coffee as their major export
crops, but these cannot be grown in' the north. Cotton is' the only
export. crop grown in the savannah.
Largely as a result of these differences, the north is. particularly
poor, with 56.8 percent of the poorest.decile of the population and
40.1 percent of the poorest three deciles, although only 18.9 percent
of the population live there. An increase in export crop prices wiLl
only benefit poor northern farmers if it includes cotton. However
Glewwe and de Tray (1988) show. that higher producer'princes for
cotton would benefit 28 percent of.the poorest'decile and nearly 20
percent of the- poorest three. deciles. If this 'was accompanied by
increased prices for coffee, cocoa, and oil palm, the benefit would
spread to include 66 percent of the poorest decile 'and nearly 70
percent of the poorest three deciles. Such price 'increases would also
benefit richer farmers and might ev'en increase rural inequality.
However, as rural incomes are on average lower than urban incomes,
they could well reduce overall inequality and would certainly reduce
poverty.
- This means that a well-designed policy of price increases could
benefit a high proportion of the poor, although any reduction in-
fertilizer subsidies to cotton farmers would 'have. an opposite effect
The urban sector experiences relatively little poverty, but has
experienced considerable changes as a result of structural' adjustment.
One possible cause of poverty in urban areas is' unemployment, and
the reductions in employment in formal 'manufacturing suggest that
unemployment could have increased, and this is suppored by the data
on unemployment. Newman (1987) reports that the LSMS data for
1985 revealed a national unemployment rate of 2.94 percent, which is
somewhat higher than the' figure of 2.5 percent' for 1975. reported by
the World Bank (1987). Also, Berthelemy and 13ourgignon (1989, p
95) report an increase in modem sector unemployment from 59,100
in 1983 to 86,400 in 1985.
The unemployment rate is virtually zero in ruiral hreas, was about
11 percent in urban areas in 1985, and is highest in Abidjan (the .area
with. the least poverty), at about 20 percent. Most 'of'the unemployed
are under 30 years old and may well be members of households' that
can support them without falling below the poverty line. This is
300 Richard Blundedl, Christopher Heady, and Rohinton Medisora
supported by the fact that fewer than 1 percent of the heads of the
poorest 10 percent of households are unemployed, while almost .75
-percent of them are self-employed (Glewwe and de Tray 1988,- table
6). Thus, urban poverty results from the poor earnings of the self-
employed, rather than unemploynment. However, both of these can be
the result of low demand in the urban labor market.
Unfortunately, no data on the effects of structural adjustment on
the earnings of the self-employed are available, but some more
detailed idea of the effects of structural adjustment on urban
unemployment can be obtained by looking at the people in the LSMS
panel who stopped wbrking. Of the people: who left the urban labor
force between 1985 and 1986, around 36 percent were aged 18 or
less, 37 percent were women between the ages of 19 and -59, and 8
percent were aged 60 or over. Many of the young people might he
returning to full-time education, many of the women may be leaving.
because of household responsibilities, anid the people- over 60 may be
retiring. The remaining 19 percent: are men aged 19 to 59. Of those,
the majority had been employees in 1985 and were either looking for
work or waiting to start a new job. It is therefore likely that most of
these adult men were either dismissed or laid off temporarily. :
This confirms that the reduction in formal employment has indeed
caused unemployment. The importance of this unemployment for
poverty depends, in part, on the speed with which people can find new
jobs. We can get some idea of this by looking at people who were
unemployed.
-If we look at the 1986 employment status of those unemployed in
1985, 81 percent of them were still without a job, although 42 percent
had stopped looking for work. Of the people who found work, 50
percent were employees (although some of them were also self-
employed in business or agrculture). This is higher than the
proportion of the overall urban work1force wli6 were employees-
(about 37 percent),. and is interesting because it shows that self-
employment cannot be regarded as the single major route into the
labor market.
This shows that the unemployed have considerable difficulty in-
finding employment, and therefore that Job losses: can cause:
considerable hardship, although:one could argue that these are the
C6te d'Jvoire 301
people who can afford not to enter self-employment. It is also
evidence of a generally weak level of demand in the urban labor
market, whichi can be expected to limit the ability of the self-employed
(the main sufferers from poverty) to improve their standard of living..
Fertility and the Role of Women
Women's labor force participation has a different pattern from that
*of men.. Their participation rate in rural areas is almost as high as
* men's, but -it is substantially lower in -urban areas, particulary in
Abidjan. Also, the form of participation is different: women are much
less. likely to be employees and are mainly engaged. in self-.
employment in business or agriculture. These- differences mean that
they are likely to be differently, afected by structural adjustment that
affects the balance between sectors.
Turning first to the distinction between male- and female wage rates,
table 7.21 provides the results of wage determination by sex. The
differences in. nationality and. junior .high. school -coefficients are
particularly noticeable, with non-Ivorian women attaining a higher
wage,. other things being equal. The nature of returns to" education and
experience also differ by gender. However, it is comforting to find
that there are significant returns for. both men and women. The sample
size confirms the much lower participation rates among urban women
in wage labor (the employee sector). Much of. this is due to their
-interrupted work patterns. caused mainly by child rearing. The
experience coefficients in table 7.21 point clearly to the effects, on
femnale wage: rates of lost work experience. If we couple this with the;
earlier drop out of women. from education, the implications for the
*relative economic position of women are -critical. This would be
especially so if under structural adjustment the returns to human- and
job-specific capital increase.
Fertility has a major effect on participation for women, and in turn,
'a major effect, through experience, on their returns from the labor
market. Table 7.22 presents some estimates of the impact of.certain
critical factors on the number of children ever born. The difference of
income and' schooling effects across urban and rural sectors provides
more support for. our decision to split. the analysis along ithese lines.
Descriptive statistics for these women are provided in appendix. C,
302 Richard Blundeli Chlristopher Heady, and Rohinton Medhora
Table 7.21 Wage Equations for Meii and Women
Symbol Men 'Women
Sample Size 414 -100
NAT -0.214 (1.96) - 0.558 (2.03)
YRS-EL 0.041 (0.99) -0-113 (0.81)
YRS-HI 0.079 (1.67) 0.164 (1.72)
YRS-H2 0.074 (0.64) -0.260 (1.20)
YRS-UN 0.211w (5.64) 0.936 (1.41)
BASICED 0.109 (1.58). 0.109 (0.46) ;
DIP-EL 0.528 (3.07) 0.683 (1.52)
DIP-HI. 0.523 (2.75) 0.615 (1.88)
DIP-UPP 0.223 (0.68) 1.489 (2.08)
DIP-TEC 0.059 (0.47) -0.173 (0.76)
YRS-TEC- 0.073 (2.32) 0.102 (1.61)
YRS-APP 0.004 (0.19) 0.128 (1.03)
Exp. 0 O.117 (8.51) 0.093 (2.79)
x- .( '7-9.
Exp2 -0.002 (4.27) -0.002 (1.77)
Const 3.537 (17.94) 4.554 (11.36)
R2 - . - 0.673 0.655
Note: See tatle 7.8 for defiuitions of symbols. :
Source: van der Gaag and Vijverberg (1989).?
tables 7.C1(a) and 7.Cl (b). The mos't impressive aspect of tabl'e -7.22
is the strong impact of education on family size among urban women.
.How eve for rural women the income effect appears to dominate. This
partly reflects tht lower education levels of women in the rural sector,
but may also Su gest a different allocation-of work actiities in the two
sectors.
CMte d'Jvoire .303
Table 7.22 Children Ever Born
Variable Urban women Rural women
Age 0.4891 0.4106
(0.0308) (0.0285)
Ages -0.0046 -0.0036
(0.0004) (0.0003)
. . School years -0.0990 -0.0633
(0.0168) (0-0391)
* -: Log (incone) -0.0842 0.5827
- (0.1286) (0-1444)
- - Cdonstant - -;:- -5.2096 -12.0359
(1.588) (1.810)
R2 0.528 0.363
N 597 847'
Notes: Income refers to permanent income per adult; asymptotic standard errors in
parentheses.
Source: Ainsworth (1989).
Migration and Economic Incentives
Evidence on the extent anrd.determinants of migration is scarce.
The LSMS panel .does provide some distinction between those who.
remain.in their local area and those who leave, including some
information on reasons for leaving. However, it does not provide
employment and income. measures for those. who leave (table 7.23).
As a result, measuring the effect of -economic incentives on migration
seems a difficult task. However, for those in employment before
migration Vijverberg (1989) provides a rather innovative analysis that
lends some strong support -to the hypothesis of clear economic
incentives to migrate. Although these results'are preliminary, the
importance of this transition for countries -undergoing adjustment
places a high value on any-information about the nature of economic
incentives on migration (table 7.24). IL table 7.24, the wage-gap and
prof-gap variables refer to the estimated gap in wage and -profits,
304 Richard Blundeli, Christopher Heady, and Rohinton Medliora
respectively, between the actual and expected returns to working. The
returns to labor in household enterprises is measured by its
contribution to enterprise profits. The expected return is calculated
Table 7.23 Characteristics of LSMS Panel Members, Aged 12 to 65,
by Participation and Migration Status
Participants Manpanit fpanns
Other Other
Category Abidjait urban Rural Abidian urban Rural
A. Per capita consumption
expenditures (CFA 1,000)
Migrants 687 359 262 514 314 215
(781) (249) (153) (469) (166) (163)
Nonmigrants 702 416 257 633 395 .212
(722) (350) (200) (633) (302) (179)
B. Reason for migration
among migants (percent)
Work (self/faniily) 28.6 41.2 26.7 50.0 40.2 30.3
Marriage 143 88 -20.0 115. 6.5 28.3
School 0.0 5.9 0.0 1L5 26.2 13.1
Other/unknown 57.2 44.1 53.3 26.9 27.1 28.3
Total 100.0 100.0 10X.0 100.0 100.0 100.0
C. Destination of migrants
(percent)
City 42.9 5050 53.3 34.6 54.2 54.2
Town . 0.0 26.5 13.3 26.9 28.0 -11.0
Large village 0.0 14.7 6.7 7.7 9.4 17.9
Small village 14.2 2.9 . 20.0 7.7 6.5 .55
Camp 0.0 2.9 0.0 7.7 .9 11.7
Other/unknown 42.9Y 2.9 6.7 15.4 .9 1A
Total 100.0 100.0 100.0 100.0 100.0 100.0
Number of observations
Migrants . 7 34 15 .26 107 145
Nonmigrants 246 221 205 389 440 1,720
* Annual, modified by equivalence scale (Olewwe 1987). -
Source: Vijverberg (1989).
COte dI'voire - 305
Table 7.24 A Probit Model of the Determinants of Migration
Variable Coefficient T-sta tistic
Constant -1.190 (2.76)
Education (years) 0.033 (1.11)
Age -0.009 (1.11)
Female -0.010 (0.05)
- Head of household 0.109 - (041)
Household size 0.006 (0.40)
Wage -0.070 (2.20)
Wage gap 0.346 (2.07)
Profession -0.113 (3.03)
Profession gap 0.110 (1.90)
Notes: Data as in table 7.23.
'Source:-Vijvcrberg (1989)..
using standard wage and unit profit regressions. The strong positive.
effect, especially for the'.'wage-gap variable, indicates the potential
importance of economic incentives in migration trends. If we add to
this the*,similarity, in characteristics, between migrants who were
participants (and ti erefore used in table 7.23) and those who weremnot
in work, we might expect migration to respond quite significantly to'
the relative returns to work across different areas in C6te d'Ivoire.
* ,'.Taken together with our results exported. earlier that: showed the
movement into agricultuial self-employment, in which -output prices
have maintained their real value, it suggests that labor markets in Cote.
d'Ivoire have responded-quite generally to relative returns available in
different labor market sectors and locations.
Conclusion'
We have argued that any analysis of structural adjustment in C6te
d'Ivoire must distinguish. between the rural and urban areas. Until
recently the rural. areas were hardly affected by structural adjustment.
This means that the analysis has had to be in terms of the likely
consequences of future agricultural price changes. Our analysis
306 -Richard Blundell, C1hristopher Heady, and Rohinton Medhora
showed that. labor supply in rural areas is responsive to such
incentives. We have also seen that price increases of four main crops
can help the majority of poor people. Thus, a price increase can
1increase output and reduce poverty. However, in the context of the
recently announced coffee procuremenut price reductions, our results
* suggest a reduction in labor supply and an increase in poverty.
By contrast, the urban sector has been affected by a combination of
* structural adjustment and external . shocks for longer. Formal.
employment has' fallen while informal employment and self-:
employment has risen. There is some evidence that those industries
favored by government: policy, such as those producing: goods that are
heavily traded intemationally, have performed better than average, and
: some have even increased their labor force. Thus, the labor market has
* succeeded in reallocating labor between industries, and the evidence
suggests that this has involved a reduction2 in the real wages of zmost
workers. The urban labor market continues to' suffer from a low level
of demand and the unemployed have difficulty finding work. Urban
poverty has therefore probably increased during the period 'of-
structural adjustment.-
Our analysis suggests that structural adjustment has resulted in a
move of the labor force into agriculture. Thereis also some evidence
of. a shift in urban' employment -from: services into manufacturing.
There is no particular reason to suppose that an increase in the
production of manufactuired tradable goods is not sustainable in the,
long run. There are, however, some doubts as to whether agricultural
exports can be expanded without serious reductions in world prices.
These. doubts center mainly on cocoa and coffee exports, whose
significance - has been greatly reinforced by the' recent reduction in
procurement prices. If agricultural exports were increased through.
diversification, the shift in resources into agriculture might be
beneficial in the-long run.
The other possible long-run problem of structural adjustment is the,
diversion of young people out1of education and into work at. too early:
a stage. The paper by Alessie and others (1990) suggests tat this may
not be a severe problem:: the young people who respond to incentives
are those. who are maintaining .s'.ne' contact with school, while the
decision to le'ave school completely does not seem Ito be:strongly
~~;. . ...
C&te d' voire 307
influenced by current prices. As reentry into full-time 'education is
reasonably easy for those who maintain contact, this suggests that
households may not be sacrificing their children's long-term interests
for the. sake of exploiting short-term price movements.
APPENDIXA
TIE LABOR SUPPLY MODEL
The separable model of labor supply can be expressed as:.'.
-Ls F(cp,w, I, z) (A.i)
where Ls is a measure of labor supply, cp is an; index of consumer
prices, w is the market wage-rate; I: is profit and other nonlabor
income, and z is a vector of household characteristics.
Market wages are given by:
w = G(pp, x) (A.2)
where pp is an index of product prices, and x is a vector of other
factors that affect wages.
This produces a reduced form:
=5 H(cp, pp I?, z, x) (A.3)
In the nonseparable modlel, labor supply is given by:
- f(cp, I, w, pp, LAND, z) (A.4)
where LAND is the quantity of land available to the household.
Once again one can substitute out for wages in the case where there
is a labor market, resulting in:'.
Ls h(cp, I, pp, LAND, z, x). (A.5)
The nonseparable model has additional terms in both the structural
equation (A.4) and the reduced form equation (A.S). A comparison.
of equations (A.4) and (A.i) shows the addition of producermprices-
308
CMe d'Ivoire 309
and land, while a comparison of equations (A.5) and (A.3) shows the
introduction of land.1
1. it is possible that the area of land.is also a choice, variable of the household. In
that case, a full analysis- of labor force participation would also requii'e the modeling-
'of land area decisions.
:APPENDIX B
TiE STATISTICAL MODEL OF TRANSITIONS
The most natural statistical model for describing such relationships'
takes the form of a state-dependent discrete transition model in which:
current* period work status is related directly to a vector of
characteristics "zi" fbr individual "i", conditional on last period's
work status.. If there, were only two initial states, work and nonwork,
described by whether an index St equals unity or zero, then the
probability of being observed in' work in the current period
(S? 1)would be given by:
Pr|S, l]= -t(z'p). Pr[S =1] + F(zy ).Pr[S? o] 0rny
.~ ~I!I . t
where F(.) describes each discrete state-dependent probability, and B
the parameters of staying in work conditional on working in the initial
period while Y the parameters of the transition probability into work.
-If ,ff and-Y, the parameters of the two possible transitions into work
in equation (BI),were equal, there would be .no state dependence.
Indeed, apart from variables that may change from one period to the
next, P4[S0=l] - Pr[S1=l]. In this case cross-sectional data alone could
be used to estimate the underlying parameters since Pr[S=lj = F(z!f ).
However, with state-dependence the repeated observations available on
each individual in panel data is required. Since we have stressed the
polic importance of state-dependence in our general model of work
and 'schooling this issue will become an important aspect of our
empirical results.
310
APPENDIX C
DESCRIPTIVE STATISTICS FOR THE FERTILITY
ANALYSIS -
- - . . . - : - . .da . .. er t i
Table 7.Cl(a) Sample Means and Standard Deviations for Fertility
Analysis, by.Location.
All Wonien Urban Rural
Variable Mean' SD Mean SD Mean SD
Children ever born 3.91 3.30 3.14 3.05 4.46 3.37
Age . 34.31 15.07; 30.5'18., 12.93 37.07 15.85
Years of Education 1.69 .3.43 - 3.40 4.47 0.48 1.59
-~~~~6 yer :.3 0. -. . -. .-
D Dummy, 1- years 0.136 0.343 0.194 0.396 0.094 0.292
Dummy, 1-2 years . 0.029 0.168.- 0.027, 0.162 0.031 0.173
Dummy, 3-6 years 0.107. 0.309 0.168 0.373 0.064 0.244
Dummy,7+years 0.107 0.309 0.243 0.429 0.011 0.103
Urban dummy 0.41 0.49 1.00 0.00 0.00 0.00
Ln permanent
income/adult 12.59 0.82 13.09 0.72 12.24 0.68
Ln current
income/adult 12.29 1.43 12.81 1.33 11.92 1.39
Ln nonlabor.
income/adult. 8.31 3.32 10.10 2.52 7.04 3.23
N 1,444 . 597 847
311
312 Richiard Blundell, Clhristopher Heady, and Rohintori Madhiora
Table 7.C1(b) Sample Means and Standard Deviations for Fertility
Analysis, by Age
Age 15-24 Age 25-34 Age 35+
Variable Mcan SD Mean SD Mean SD
Childreneverborn 1.13 1.28 3.90 2.27 6.06 3.29
Age 19.53 2.73 28.94 2.73 48.74 11.38
YcarsofBlducation 3.06 3,88 2.26 4.20 0.31 1.56
Urban dummy 0.53 0.50 0.45 - 0.50 0.30 0.46
Ln permanent
incomeladult I12.65 0.77 12.77 0.88 12.44 0.79
Ln current
income/adult 12.44 1.00 12.47 -1.52 12.04 1.62
Ln nonlabor
income/adult - 8.76 2.99 B.42 3.17. 7,89 3.59
N 473 355 616
COte d'Ivoire 313
References
Ainsworth, M. 1989. Socioeconomic Determinants of Fertility in Cote
dilvoire. Living Standards Measurement Study Working Paper
No. 53. W'shington, D.C.: World Bank.
Ainsworth, M., nd J. Munoz. 1986. The Cote d'ivoire Living
Standards Survey: Design* and Implementation. Living
Standards Measurement Study. Working Paper No. 26.
Washington, D.C.: World Bank.
Alessie., R., P. Baker, R. Blundell, C. Heady, and C. Meghir. 1990.
"The Working Behaviour of Young People in Rural Cote
d'Ivoire." UCL Discussion Paper: 90-01, January. London:
University College London.
Berthelemy, J. C., and F. Bourgignon. 1989. "Growth and Crisis in
CBte d'Ivoire." Padis: DELTA. Processed.
Bhatia, Rattan 1. 1985. The West African Monetary Union, An
Analytical Review. IMF Occasional Paper No. 35. Washington,
D.C.: IMF.
Chenery, Hollis, and T. N. Srinivasan. 1989. Handbook of
'Development Economics, vol. Il. Amsterdam: North Holland.!
* - Deaton7 A., and D. Benjamin. 1987. "Household Surveys and Policy:
Reform: Cocoa and Coffee in'the .Cote d'Ivoire." Research
--Program in Development Studies -Discussion Paper No. 134.
Princeton, New Jersey: Princeton University.
Devarajan, .Shantayana, and Jaime de Melo. 1987. "Evaluating
Participation- in African Monetary Unions: A Statistical
Analysis of the CFA Zones." World Development 15: 483-
496.
Fields, G., 1989. "Labor Market Policy and Structural Adijustment in
Cote d'Ivoire." Washington, D.C.: World Bank.
Glewwe, P. 1987. The Distribution of Welfare in the Republic of C6te
d'lvoire in 1985. Living Standards- Measurement Study
Working Paper No. 29. Washington, D.C.: World Bank.
314 Richard BDindell, C/lristopiser Heady, and Rolinton Med hora
Glewwe, P., and D. de Tray. 1988. Thze Poor'during Adjustment: A
Case Study of Cote d'Ivoire. Living Standards Measurement
Study Working Paper 47. Washington, D.C.: World Bank.
Heckman, J. J. 1981. ."Heterogeneity and State Dependence." In S.
Rosen, ed., Stutdies in Labor Markets, pp. 91-139. Chicago:
University of Chicago Press.
IMF. Various years. International Financial Statistics. Washington,
D.C.
Islam, S.- 1984. "Devaluation,. Stabilization Polici6s, and the
Developing Countries." Journal of Development Economics
14:- 37-60.
Xianbur, R. 1989. 'Poverty Alleviation Under Structural Adjustment:
A Conceptual Framework and Its Application to the Cote
d'Jvoire." Washington, D.C.: World Bank. Processed,
Laker, -John F. 1981. "Fiscal Proxies for Devaluation: A General
Review." IMF Staff Papers 28: 118'143.
Lavy, V., and J. Newman. 1989. 'Wage Rigidity: Micro and Macro--
Evidence on Labor Market Adjustment in the Modern
Sector."' The World Bank Economic Review 1(1): 97-117.
Lorub, K. 1989. "C6te d'Ivoir-: Industrial Competitiveness Durng
Economic Crisis and Adjustment."' Washington, D.C.: World
Bank. Processed.
Medhora, Rohinton. 1989. "The West African Monetary Union:
Institutional Arrangements and the Link With France."
Toronto: University of Toronto. Draft...
Nakamura, A., and M. Nakamura. 1985. "Dynamic Models of the
Labor Force Behaviour of Married Women That CaniBe
Estimated Using Limited Amounts- of Past Information."
Journal of Econometrics 27(2): 273-298.
Neurrisse,.Andre. 1987. Le Franc CFA. Librairie General de Droit et
de Jurisprudence;.
Newman, J. L. 1987. Labor Market Activity in Cote d'Ivoire and
Peru. Living Standards Measurement Study; Working Paper
36. Washington, D.C.: World Bank.
COle dklvoire 315
Newman, J. L., and P. Gertler. 1988. "Female Farm Work and Home
Work in Rural Cote d'Ivoire." Welfare and Human Resources
Division. Washington, D.C.:World Bank. Processed.
Pegatienan Hiey, Jacques. 1.987. Ivory Coast, Stabilization and
Adjustment Policies and. Programs. Country Study No. 16.
Helsinki, Finl and: World Institute For Development
Economics Research..
Singh, I., L.: Squire, and J. Strauss, eds. 1986. Agricultural Household
Models. Baltimore, Md.: Johns Hopkins University Press.
Taylor, Lance. 1981. "ISILM in the Tropics: *Diagrammratics of the
New Structuralist Macro Critique." In William R. Cline and
Sidney Weintraub, eds., Economic Stabilization in Developing
Countries. Washington, D.C.: Brookings Institution.
-__________ . 1983. Structuralist Macroeconomics, Applicable
Models For the Third World. New York: Basic Books.
van der Gaag, J., and W. Vijverberg. 1989. "Wage Determinants in
C6te d'Ivoire: Experience, Credentials and Human Capital."
Economic Development and Cultural Change 37(2): 371-
381.
Vijverberg, W. 1988. Nonagricultural Family Enterprises in Cote
d'fvoire: A Descriptive Analysis.- Living Standards'
Measurement Study Working Paper 46. Washington, D.C.:
World Bank.
.1989. Labor Market Peformance as a Determinant
of Migration. Living Standards Measurement Study Working
:Paper 59. Washington, D.C.: World Bank.
Vinay, Bernard. 1980. Zone Franc et Cooperation Monetaire. Paris:
Ministry of Cooperation.
World Bank. 1987. ."The C6te d'[voire in Transition: From
Structural -Adjustment to Self-Sustained Growth."
Washington, D.C.
-_________ .1988. .World Development Report 1988. New York:
Oxford Uniiversity Press.
- __ - . Various years. World Debt Tables. Washington, D.C.
S~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
EGYPT
Ragui Assaad
Simon Commander
What happens in the labor market-is critical for the aggregate
efficiency of adjustment.. However, common assumptions include not
only ex. ante full -employment, but also ielatively frictionless
adjustment mechanisms, particularly as regards labor -mobility
(Corden 1989). A standard .framework accommodates: some
adjustment costs, but the switching assumptions yield the same levels
of aggregate employment. Wage rigidity is assumed to be absent if the
ex ante employment level is to be undisturbed.
Some of the problems that arise with such a framework for applied
work include the following. First, with the normal two goods, four
factors model, the key price relationship-is that.of tradables/
nontradables or-the real exchange rate, but this assumes free trade, or
at least low to constant tariffi. Two independent, domestic, relative.
prices exist: the price of exportables relative-to importables, and the
price of importables relative. to nontradables where the former
depends on the terms of trade and tariffs, and the latter on the-
exchange rate and domestic money supply (Collier 1988). Second,
one normally assumes that capital is sector-specific in the. short-
medium- term, but that labor. is mobile. Wages are also held to be
flexible.- If wage rigidity obtains in the-covered or formal sector, this
would imply a sharper fall in real wages in the uncovered sector. If,
however, real wage flexibility exists alongside segmentation and hence
a lack of mobility, a real devaluation will require a correspondingly
higher level of real wage decline in both sectors and unemployment in
-; 317
Table 8.1 Macroeconomic Indicators, 1973-88
Andicator 1973 1974 1975 1976 1977 1978 1979 :1980 1981 1982 1983 1984- 1985 1986 1987 1988
Rate of growth
GDPa 0.8 2.7 9.1 15.3 13.5 5.9 6.2 10.3 3.8 10.1 7.6 6.2 6.7 2.7 Z.5 3,2
Tradables outpu- 0.6 -1.4 10.9 8.5 10.5 11.5 7.1 13.8- 4 5.8 5.9 8.5 1.5 n.a. n.t. n.a.
Nontradables outputa 20,1 16.1 9.3 20.5 9 5.9 12.4 15,9 1.1 6.4 4.2 4.7 2.3 n.e. n.n n.a.
CPI 4.2 12 11.6 11.2 10.8 13.6 7.4 23.4 14.7 15.7 22.1 10.3 14 23.7 20 18.9
Realinterestrate 0.6 -6.5 -6.2 -3.9 -8.6 -5.3 -0.7 -9.9 -3.5 -5.6 -5.6 0.6 -5.3 -10.2 -6.4 --53
Shares of GDP
Budget deficit -10.1 -20.5 -28.6 -18.8 --16.9 -23 -26.9 -15.1 -25,3 -19.1 -23.1 -22 -22.8 -17.1 -20.0 -19.8
Current account
deficit 0,2 -14.4 -18.2 -7.9 -6.8 . .5.4 -10.6 -6.4 -9.5 .-11.8 -6. -7 -0 -9,7 -6 -3.1
Reittances . 1.3 1.7 2.7 4.4 4.2 7 13;5 11.4 11.5 6.5 9.2 9.7 7.4 5.4 4.6 4.1
Gross fixed
investment. -21 16 25 22 22 27 30 27 29 27 24 22 -2 18 18.5 19
Tennsof tradeb -.96 96 83 .83 82 74 93. 100 102 93 93 89 8.8 76 57 n.n.
Real effecdive
exchangerat . 164 155 159 176 183 178 95 100 116 139 166 199 226 240 255 281
n.a. = not available
a. 1980 prices. . - .
b. 1980 = 100..
Saurces: Central Agency for Public Mobilization and Statistics (CAPMAS), Government of Egypt, and World Bank.
Egypt 319
the uncovered sector.1 The output. effect in the tradables sector will
consequently be constrained by the capital/labor slack. This. appears to
be recent Egyptiian experience. Further, if labor does not move into
the tradables sector, investment incentives will be weak given that the
marginal physical product of extant capital does not increase. In such
cases, the only stimulant to new investment is the relative price increase
for output over capital.. When-as, in the Egyptian case-the
adjustment path is marked by wage flexibility alongside constraints on
mobility and an inability to correct fundamental macroeconomic
-imbalances, the costs of that partial adjustment on both the.wage.and
employment sides become correspondingly higher.
Some Macroeconomic Features of the Egyptian Economy
Under the impetus of an oil windfall-in the early 1970s and higher
transfers into the Egyptian economy, growth accelerated significantly
over trend attaing around .10 percent per annum from .1973 through
1982 (table 8.1). Since then deceleration has occurred, which has been
associated with strong adverse terms of trade shocks' equal to 11;
percent. of GDP between 1982 and 1987. Since 1986- per capita
consumption growth has turned'negative, real-investmnent-has falle'
sharply, while the degree of fiscal adjustment- required of the economy
has been made more profound by the large external debt overhang.
By 1988 aggregate external debt exceeded 115- percent of GDP, with
debt service amounting to over 60 percent of exports. Current
projections suggest little likelihood of OG&growth surpassing 3.5
percent per annum over the medium ter'm. Moreover, the economy,
remains marked by profound domestic and external imbalainces, -with
the current account deficit exceeding 10 percent of GDP and the fiscal
deficit extending to over 20 percent of. GDP through the 1980s.
Conservatively estimated, domestic inflation rates have ranged from 15
to 25 percent per annum. Real wages bave fallen to roughly 80
percent .of: 1982 levels, 'and open unemployment, though consistently
rising from. 1960 onward, appears to have grown rapidly between
1976 and 1986. Census data indicate that the unemployment rate
1. Such a: result could be achieved through job security, other nonwage.
benefits and particular institutional features of the labor market.
320 Ragul Assaad and Simon Commander
more than doubled from 5.3 percent in 1976 to 12.4 percent in 1986
(CAPMAS 1976, 1986).-
The policy response to external shocks-declining export revenues,
falling remittances, and lower real transfers into the economy-have
been spasmodic and inconsistent. Agreement with both the. IMF and
the World Bank, particularly. on the pace of economic reform, has
been largely absent.'A managed adjustment path remains particularly
difficult given the scale of controls. and price -fixing -rules operating in
the economy.Such. controls.- include not only adminstered prices-and
forced deliveries for the major agricultural sector tradables-cotton
and rice-but also 'particular labor market interventions. 'These
interventions include centralized determination of the floor for wage
increments, constraints on hiing and firing for enterprises withi more
than ten employees, and the massive presence of the: consolidated
public sector in' the employment of Egyptians. Equally, the
widespread use of subsidies on:wage goods (with. total. subsidies
accounting for at least 5.5 percent of GDP in 1988), and the use of the
public food ration and subsidy- system for basic income support. raises
direct, short-run tradeoffs between fiscal pressures and a historical
commitment to low-cost wage goods.
Gradualism has been the dominant adjustment method. The -fiscal
deficit had been trimmed somewhat to some .20 percent of GDP by
1988, subsidies on both food 'and energy had been reduced through a
combination of price increases and falling import costs, while the
range of -goods attracting subsidies had been .cut.2: In all, subsidies
have been reduced by,over- 8 percent'of'GDP during the past six
years. As for the majority of developing- countries, public.expenditure.
reductions have fallen.mostly on capital outlays, with rapid growth in
interest payments marking the current. expenditure side. Gross.
domestic investment, fell by' about 1.3 percent per annum between
1982 and 1987, with drastic declines in .1986 and 1987. While real
wage decline and fiscal contraction have achieved- some, demand
dampening, -aggregate consumption has risen to over 90 percent of.,
GDP. from around 85 percent through the- 1980s.
2. The size of the wedge between domestic and barder prices for energy is,,.
such that over 300 percent upward adjustmernt of nominal prices has still left
petroleum producis at around 30 percent of border prces.-
Egypt 321
-A more -active exchange rate policy has resulted in greater
consolidation of transactions in the so-called free market or
commercial pool; which has depreciated by 20 percent since 1982. A
.weighted index for the real effective exchange rate indicates, however, '
:an. appreciation of over 40 percent between 1982 and 1987/88.
Further measures-such as foreign earnings retention-designed to
stimulate exports. have resulted- in some positive response for -nonoil
. exports,'but despite explicit policy measures to.raise agricultural prices
and limit the share of output taken in forced deliveries, output-of the
major exportable-cotton-has been held back by producer prices
ranging from 30 to 40 percent of international prices, as well as low
relative returns to both cotton' and rice, the other main exportable.
Although.-the size of the real wage decline .in the economy in recent
years suggests gairIS in external competitiveness (real wages being a
counterpart of the real exchange rate), the absence of appropriate
fiscal and monetary policy, combined with powerful market rigidities
and a reduction in. the size and short- to medium-term scope for
expansion of the market sector, has tended to dampen any shift into
exportables. As will become clear, labor market rigidities, in particular,
the consequences of an expanded public, and: government sector for
l labor. mobility and output mix,- remain key features:: yet. to -be
addressed in a sustained adjustment. program.
Windfall, Economic Liberalization, and the Labor Market
Economists have argued that the-post-1973 boom and the
government's economic liberalization measures (Infitah) provide- a
classical Dutch disease story (Dervis and 'others 1984). Under the
standard argument of a relative price effect, and hence appreciation of
the real exchange rate, the oil' and associated activities' windfall
effectively contracted. the nonbooming-tradables sector, bolstered
output. from the nontraded goods sector, and, finally, as' the windfall
petered out,; left the economy exposed to massive external and
domestic account imbalances. This outcome on the external account
* ' - side4was derived'from contraction in the nonoil exportables subsector:
and from rigidities that constrained the transfer. of resources
postwindfall into the nonoil tradaAes sector. Further, as the windfall
accrued primarily to government and the elasticity of expenditure to
322 RaguiAssaad and Simon Commander
permanent revenue exceeded unity with no consumption smoothing,
the level of fiscal* imbalance was likewise exaggerated, with public
expenditure cuts incommensurate to the subsequent decline in
income. The presence or absence. of such a Dutch disease effect is
*important not only for understanding the macroeconomic story (and
hence the required. adjustment), but also for the particular. labor
* market response that might be-expected.- .
In labor market terms a Dutch disease model can generate a
number of outcomes. In. the first place, the expenditure effect
combined with resource movement will generate shifts in sectoral
labor allocation (Corden and Neary 1982).-The most obvious would
occur as labor moves into the booming sector. If that sector-as is
habitually the case with oil-generated booms-is an enclave, then the
derived demand for labor and materials from the domestic economy
will be low given both the capital and import intensities of the enclave.
However, the .conventional' weak employment shifts that might be'.
expected need to be qualified somewhat. In the Egyptian- case, the
regional nature of the oil boom had significant implications for the
demand for labor. One component of Infitah was the liberalization of
rules regarding the rights of Egyptians-to work abroad. This resulted.
in a rapid growth in migration to other Arab. states, with remittances-
increasing more than 14-fold (in real terms) between 1973. and 1980.-
Thus, the counterpart of one major component of the growth in:
exogenous resources into the economy, was. an outflow of labor. In
this respect, the regional spillover effects resulted in labor being' bid
awa.y frm the nonoil sectors, particularly agriculture. This caused
upward pressure on the wage. level, and in the late 1970s/early 1980s
led to severe skill mismatching in segments of the labor market"
(Hansen and Radwan 1982)..
The effects of expenditure growth, public expenditure growth in
particular, complemented. the growth in regional derived demand for
labor' from the oil sector.. If, on standard assumptions, output' in the
* tradables and nontradables sectors is a function of the real product
wage, then a windfall will drive up the wage rate when measured in
traded goods, at the same time raising the relative price of nontraded
to. traded. goods. As output in the nonbooming tradables sector. is -
likely to be inversely related to the wage rate; this would result in a
Egypt 323
change in the composition of aggregate output, with a declining share
in that.sector. For many developing economies this is'likely to have an
impact on the agricultural sector, especially the exportables subsector.
If the booming. sector is indeed an enclave, this would 'result in
-deteriorating intersectoral terms of trade for agriculture-. With likely
widening- in the, rural-urban wage gap. Dynamically, one might expect
this to yield accelerated internal migration. However, if-as with
Egypt-the booming sector (regionally construed) :does bid away
labor and a significant share of that labor originates in:agriculture, the
effect on the' nonoil exportables sector will be similar, but the adverse
terms of trade shift against agriculture will be absent. The derived
labor demand from the enclave will raise the real wage, with -tbe
ultimate employment outcome- depending on the relative labor
demand schedules for the nontradables sector compared to nonoil
tradables. In the case of Egyptian"agriculture, disentangling. these'-
effects is complicated by' joint' production of tradables and
nontradables and the system of forced.deliveries. Even if we assume
weak resource movement,: the spending effect cof a windfall would
itself be a sufficient condition for raising the real wage.. 1-Towever, if
the share of the nontraded goods output in the consumption basket
exceeds its contribution to a weighted average of the supply elasticities
of the.-two sectors (that is, excess demand for the nontraded good),
then a windfall might result in a fall in the wage and unemployment.
When the reverse relationship holds, the outcome would be a rising
consumption wage and labor: scarcity (Neary and van Wijnbergen-
.1986).-
The -latter pattern appears to have characterized the Egyptian
experience Jin certain respects. Real 'consumption wages rose
significantly between 1973fl4 and 1982/83, and in certain sectors
significant labor shortages and skill mismatching, .-exaggerated by: the
structure of extemal migration, emerged. On a macro plane, 'however,
some expected results were absent. Contrary -to standard models, a
current account surplus- with parallel reserve inflows did not
materialize. This can be attributed to the strong pulling-in of imports
that. followed a penod of repressod demand and was financed through
own-exchange imports. Moreover, as citizens were allowed to.hold
foreign currency balances, the appreciation was associated with capital..
324 RaguiAssaad and Simon Commander
* outflow (Braga de Macedo 1982). Nevertheless, with a fixed exchange
rate, a strong parallel market for foreign exchange, and legalized
holdings of foreign money balances a number of Dutch disease
features resulted.. Nonoil exports fell in real terms, while a declining
exchange rate had predictable current account .consequences The .
distribution of gross investment likewise reflected not only a shift of
resources in the oil sector, but some decline in the share of nonoil
tradables:in aggregate investment.
In looking at the resource allocative effects of the windfall, it is:
* critical to accommodate explicitly the impact of tariffs and the trade
regime on the distribution of investment across exportables,
importables, and home goods subsectors.. Here the free trade
assumptions of the standard model lack relevance, so. that investment
can be hypothesized to respond not only to decline in the exchange
rate and the domestic real interest rate, but to their combination with
actual' tariff levels. In the Egyptian case, despite some liberalization
after 1973, tane government maintained a wide range of quantitative
restrictions and tariffs, with periodic notification. This resulted in a
somewhat different set of outcomes' than that generated from 'a
- standard Dutch disease model; Protection changed the impact on. the.
nonoil tradables sector significantly.
Prior to considering the dynamic results of: the boom, it would be
useful to indicate the scale of expenditure effect associated with the
windfall. In this context, the combination of oil revenues, tourism,
Suez Canal earnings, and remittances shifted ihe share of exogenous
resources from around 6 percent'of GDP in. 1974 to 45 percent by
1980/81 '(World Bank 1983).3 Oil and Suez 'Canal revenueis alone
accounted for a quarter of GDP growth in this period.- The principal
revenue source-oil-accounted for nearly a. fifth of. GDP by 1980
and was captured almost exclusively by the state. Exogenous revenues
had increased -from under 10 percent to 35 percent .of total
government revenues: by 1980, while real. government expenditure
expanded by around 10 percent of GDP. Among other developments,
- -. this ~was associated with a-powerful, growth in the share of the labor
'force employed -by the state with, until -1982, implicit wage
- 3. This excludes substantial concessional capital inflows.'
Egypt 325
indexation.4 In addition, commitments were maintained with regard to
guaranteed employment for graduates, the formal abrogation of which
has not yet occurred. Alonlgside the growth in public spending, the
period was also marked by a major increase in private investment,
which rose from just over 1 percent of GDP in the early 1970s to 5 to
6 percent during 1978-82. Figure 8.1 indicates aggregate investment
growth and the rising share of private investment.
Figure 8.1 Total Investment, 1973-86
(public, private and total, 1973 prices.)
1,600
1,200 S
1~~~~~~~~~~~~~~~~~~.. .... - .
1,.000: , ,
60 '- -
* 400
200- . . - -
--p
1973 1975 1977 1979 1981- 1983 1985
( cPublic v-Puivae -T - - - - - -otal
Source: CAPMAS.
4. Iridexation was of an . post type, namely. w? a,+-Pt.i:
326 RaguiAssaad and Simon Commander
Measuring the relative price effects of the windfall is rendered
complex by. the presence of administered prices and controls, which
are most pronounced for nontradables, such as-housing, electricity,
and transport, but also significant with regard to. key tradables prices;
such as wheat.5 Official price series thus appear to demonstrate a far
stronger rate* of increase: for tradables over nontradables prices
through the windfall and beyond.6 This reflects the weight of price
controls on nontradables. Calculations of accounting prices, based on
output values and hence measuring gross resource cost, indicate that
alongside very high variance, median values for nontradables
significantly exceeded .unity, while those for tradables fell slightly
below one (World Bank 1983).7 This might imply a tradables/
nontradables price relationship the inverse of that apparently signified
by official price:series. The range of varance over the accounting
ratios also points to difficulties in netting out these effects in the
absence of reliable consumption weights. Rather,, following Shafik.,
(1989), the resource allocative consequences of the windfill -are best
mappeJ ex post, where sectoral shifts in private investment and labor.
are interpreted (with some caveats) as a response to relative factor
ieturns in the three subsectors.
The oil windfall clearly provoked: an investment boom, with real
total investment trebling between 1973/74,- and peaking in 1982/83
(Jigure 8.1). In a standard Dutch disease framework, one would expect
rising investment in the booming sector and nontri dables, with each
round of investment further appreciating the exchange rate and
aggravating the current account position. This would be further'
exaggerated if the speed of adjustment to desired capital stock levels
5. Ratios of shadow to market prices midicate, for example, that in 1980 for
energy the mean ratio (unweighted) was around 7, with high subsidy urban
consumer goods-butagas, rice, vegetable oil, and wheat (unweighted mean:
4.75)-being offset by heavily taxed goods, such as tea, coffee, and tobacco
(unweighted mean: 0.5).
6. In 1973 terms, the index for tradables (drawn from official data) moved to
260 in 1980, 474 in .1985, and 792 by 1988; for nontradables 195, 324, and-497,
respectively.
7. The accounting ratio being defined as the ritio of the shadow price to its
market price, with shadow prices calculated using international prices for
traded inputs and shadow factor prices for primary resources. An accounting
ratio of one indicates nondistortion.
Egypt 327
was strongly associated -witb foreign exchange availability-and the
supply of domestic capital goods was constrained. In the Egyptian
case, moreover, one can assume that controlled price domestic capital
goods are consum.d almost entirely by the public sector. This implies
that private srTJtor demand for capital -goods was largely satisfied
through own-exchange imports.; An index for components of
aggregate investment shows a strong acceleration in-imported capital
goods prices as also -for the construction component (appendix figure
8.A1). Consequently,. the aggregate index grows much -more rapidly
than the GDP -deflator. There are good reasons, however, for .
supposing that the GDP deflator. has a downward bias. Note also that
despite, high levels of savings held in foreign currency with
correspondent banks abroad, the, banking sector was very liquid
through the 1970s. This promoted a relatively expansive path for net
credit to the private sector. Total domestic credit to the economy grew
by over 16.5 percent per annum in real termns between 1973 and 1981,
with private sector credit growing by over 18 percent per annum.
Investment was' thus buoyant for a number of reasons. First, the
elasticity:of capital and intermediate goods imports to -foreign
exchange availability was high. The windfall raised that. availability..r'.
Second, borrowing at significantly negative real interest rates (see table
8.1) from the domestic banking system to finance own-exchange
imports was feasible. Third, fwhile conservative -in relation to the
standing - of bofrowers, net credit expansion- was significant and
facilitated the investment boom.: Fourth, following a period of
stagnationf and collapse in private investment,'the liberalization after
1973 promoted a surge in private investment. The tAift in expectations
and the low base, from which 'private investment --expanded -are
significant'factors in explaining the rate of increment. 'Fifth, the
government enacted laws that eased' foreign investment rules, such as
Law' 43 of 1974, and later, Laws 159.and 230. In' reality, such rule
changes stimulated domestic investment 'more tha'n foreign direct.
investment. Sixth, with cost-plus pricing rules dominating Egyptian,
*industry and with protection, the weak competitive 'features of the
closed market raised'desired capital stock levels via profitability and
'aggregate demand effects in the economy. With the prevailing interest
: rate structure, this yielded a relatively capital-. and import-intensive
328 Ragui Assaad and Simon Commander
structure of investment. Private investment in. import substituting
industry expanded significantly, but tended to be' very. capital
-intensive, particularly for the limited number of Law 43 enterprises
that were established (Hansen and R,dwan 1982).8 The incremental
capital/labor ratio in these projects w t eover double the 'average for
the economy as a whole. Between 1978 and 1988, Law 43 enterprises
generated less than 10 percent of total employment creation in the
private sector, principally in manufacturing (Handoussa 1989).9
:Note that taTiff protection was.both higher and more significant for
- . the private sector than for the public sector. In the latter case, a
:substantial part of-public sector-industry-including cotton textiles-
faced negative effective protection rates by the early 1980s, which
demonstrates the perversity of-the tariff anci domestic pricing structure
(World Bank 1983). Decomposing the effects of price controls and
* tariffs shows that the former had the most powerful effect on the level
of (dis)protection. The reverse was true for the private sector, where
price controls have been less apparent. Public sector industry tended
not only to be marked by low efficiency levels (Hanidoussa 1983), but
also by incoherent pricing rules at both enterprise 'and economywide
levels. Anomalies emerged so that the public sector cotton textile
industry -faced negative protection, while the private textile sector
benefited from protection, largely via tariffs. C-onsequently,.investment
was not as skewed. toward nontradables- as- might be. expected. in a
Dutch disease framework. Figures 8.2 and 8.3 show that while
nontradables investment increased to a limited extent for the public;
. sector, this was not the case for the,private-sector. Apart from rapid
expansion of investment. in the. booming sector, that in nonoil
tradables experienced steady growth through- the 1970s.- Indeed,
private investment in: import-substituting industry doubled as a share
o: private investment from around 11 percent for 1972/74 to some 22
percent between 1978 and 1981. This was reversed somewhat after
1980 (Shafik 1989).
S. The largest share of Law 43 investments has been in the financial sector.
9. At least a third'of total investment under Law 43 projects has actually
'been mobilized from the public sector, either by way of direct equity stakes or
long-term financing from public sector financial institutions.
Egypt 329
Figure 8.2 Public Investment, 1973-81
(by subsector, 1973 prices)
70
60" S> > #
50
.40
20
10 - .... .
1973 1974 1975 1976 1977 1978 -1979 1980 1981s
Yeaws
-N:N-oo -- Non- -----Oil ----Import-
: trahdbles tradables substituting
industry
Source: CAPMAS.
The labor market consequences of the boom were marked by two
main, features, the expansion of public. employment and the relatively
weak employment multiplier from incremental investment in both the
domestic oil sector (as. expected) and the import-substituting import
sector. In the-latter, employment remained broadly constant (see table
8.2). The sharp decline- in the share of the nonoil tradables sector can
be largely attributed to developments in agriculture. In terms -of
-investment, output,. and employment, agriculture. was subject to the
greatest contraction. As significant; the composition of output in the
primary sector shifted in a major way against exportables. The
available: evidence indicates clearly the extent to .which the- major
tradables had negative effective protection rates through most of the
period 1973-85, whether using official or equilibrium exchange rates.
-'4'~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -~
330 Ragui Assaad and Simon Commander
Figure 8.3 Private Investment, 1973-81
(by subsector, 1973 prices)
90
80
70
10 -
E tmaabe tradble substituting/
30" ' - ~,
* o - r a e a a s o v a a b r
1973 1974 1975 1976 1977 1978' 1979 1980 1981
, , ~~~~Years
~~~~~Non- o i N OD --- i - mport-
- tradabIles trdbbles .substitudng
- . ~~~~~~~~~~~~~~indistry.
negative,over a range of 22 to 200 percent in this period. Using an
-'equilibrium exchange rate measure, effective. protection was generally
negative and in excess of 55 :percent. Similarly -high rate's of net
taxation held for rice' and wheat, With producers subj'ect to taxatio.n
though btdietad indirect.price interventions (Detir 1989). A
woighted output price ratio for the main tradables against the principal
nontradables shifted downward from 2.5 between 1965. and 1969, to
2.0 betwe'en.' 1975 and 1979, and 1.8 between' 1980 'and 1985.,
Producers of foddercos horticultural poueand te
For cotton, t er efctvprops, onrt: sn preoduicia'd exchaner
noncontrolled outputs (sicluding birseem and wheat straw, the main
types of livestock feed) benefited either from positive effective
Table 8.2 Structure of the Labor Force, 1973-86
Annual compound
fumber of people (millions) growth rate (percent) Share of Iota; (pcsrent) Share of increase (percen)
Category 1973 1982 1986 1973-82 -1982-86 1973 1982 1986 1973-82 1982-86
Economic sector
Agriculture 4.7 4.3 4.2 *1.1' -0.4. 50.5 36.8 32.7 -19.0 -5.0
Mining and quarrying -
(includes petroleum) 0.0 0,0 0.0 3,8 19.2 0.2 0.2 0.4 0.0 2.0
* . Afanufucsurinagt ' ' 1.3 1.7 1.4 . 2.9 . -3.7 13,9 14.4 11,2 17.0 -1 8.0
:s1b , (0.7) ().7) (0.7) 0.2 0.8 (7.4) (6.1) (5.6) (0.7) (1.8)
* -, ConstructIon' . 0.3 0.- ',U,6 - 0,8 ' 9.5 6.7. 2.8 5.3 6.2 15.0 15.0
Servicet 2.9 4.3 4.8 4,6 2.3 31.0 37.6 37.1 64.0 33.0
Transport (tA) (0,6) . (0.6) . 5.5 0,5 (411) (5.3) (4.9) (15) (1.0)
Toral employment . . 9.1 10.9 11.2 Z.0 0,8 98.$ 94.3 87.6 78.0 27.0
* Government . (1.S). (2.0) (2.4) 3.6 4,3 (1S.9) (17.7) (18.9) (25.1) (302)
Public enterprise (0.9) (1.2) (1.2) 3.9 0.3 (9.3) (10.6) (9.7) (15.9) (1.0)
Private (10 or more
employees) . (0-2) (0.3) (0.3) 4.5 7.7 (1.8) (2.2) (2.7) (3.6) (7.1)
Unemploymcn t 0.1 0.7 1,6 17.0 22.1 1.5 5.7 12.4 22.0 73.0
Total domestic labor force 9.3 11.6 . 12.8. 2.4 2.6 100.0 100.0 100.0 100.0 100.0
.Workers abroafd . .0.1 - 0.6 1.2 *. 23.7. 23.4 0.8 5.2 9.4
iVotes: Includes individuals 12 to 64 only. Labor force-proportions for 1973 and 1982 were calculated from the respective Labor Force Sample Surveys
and applied to labor farce estimates based on census results, to achieve comparability with 1986 census data.
a. The share of manufacturing seems to be understated and that of construction overstated in 1986.
b. 1SI (import-substitution industrialization) includes all public and private (10+ employees) manufaccuuing..
c. Services include utilities. commerce, fina,nce, insurance, real estate, transport, communications, community, and social and personal services.
. d. The estimate of the number. of Egyptian workers abroad in 1986 is the same as the figure Ferghany (1988) provides for early 1985. No estimate is.
* available for 1986.
Sources: CAPMAS, Labor Force Sample Surveys for 1973 and 1982, Population Census 1986, Wage and Hours of Work Bulletin, Permanent Employees in
the Government and the Public Sectoi for 1982 and 1986; other government employment data from Hansen and Radwan (1982, table.A.l); public
enterprise data from Public. Enterprisc Information.Center, Cairo;'migration data from Ferghany (1988),'.
332 RaguiAssaad and Simon Commander
protection and/or significantly lower taxation rates. This was
particularly -true for the livestock industry.
One consequence of the agricultural taxation policy and the
relative price structure thereby generated' was to induce particular
patterns of technical change. First, the demand for crop labor (given
higher demand, particularly for adult male.labor, in tradables relative
to nontradables) fell to some degree. Second, linked to the growth in
outmigration from the sector to other Arab countries, after.1980, the
government introduced an explicit set of policy measures-including
discriminatory interest rates and direct credit allocations-whose
objective was to induce capital/labor substitutions in production
(Commander 1987). Thus, within agriculture resources shifted away
from exportablks, with. labbr reallocations to the import-substituting-
livestock and home. goods sectors. The net effect of protection for the
livestock subsector, with the. growth in. nonfarm employment and
income as well as expanded external migration, offset domestically-
up to 1984-the shift out of the exportables subsector. This allowed
for. a strong positive real wage effect, even as the move out of tradables
had adverse implications for the trade balance.
. As expected, the general investment boom was translated"into rapid
growth in the construction sector, whose share of the t6tal labor force
nearly doubled from .2.8 percent to 5.2 percent between. 1973 and
1982. Moreover, the employment elasticity of output was high relative
to other subsectors (table' 8.3). However, the. most important factor
explaining the relative growth in nontradables employment can be
attributed to the expansion of government and public employment,
which grew by more than 3.5 percent per annum between 1973 and.
1982, a full percentage point above the rate.of growth' in the total
labor force. Compared with the pre-boom period, total public
employment 'more than doubled,.absorbing--more than' 40 percent of -
net labor-force growth in this period. This expansion came, on top of
an earlier, sustained growth in public. employment; a key feature of
the Nasser period (Abdel Fadil 1980). This had important longer-run
consequences by.compressing .the market sector, of the economy,
while also shifting resources into the importables and home goods
subsectors. Allied to the -wage setting mechanism and the graduate
Egypt 333
Table 8.3 Growth in Value Added and Employment by Sector,. 1973-
1986
(1981/82 prices; aninual growtht rates in percent)
Employment
elasticity
Value Added Employment of output
Sector 197342 1982-86 1973-82 198286 1973-82 1982-86
Agriculture 2.5 2.3 -1.1 -0.4 -0.4 -0.2
Manufacturing 7.3 4.9 2.9 -3.1 0.4 . -0.6
Construction 10.0 2.8 .9.5 6.7 1.0 2.4
Services 6.3 4.1 4.6. .2.3 0.7 0.6
Transport 15.0 6.8k 5.5 0.5 0.4 0.1
Source: World Bank data.
employment guarantee scheme, this strongly affected the response of..
the labor market to the recession after 1982.
As regards the scale of migratory flows to the Gulf and other oil-
rich economies in the region, estimates vary widely. By the early
1980s, possibly as many as one million Egyptians were working in
other Arab countries (Amin and Awny 1985), implying a domestically
available labor force roughly 9 to 10 percent below the potential
upper bound. Other survey-based estimates (see table 8.2) point to no
more than 200,000 Egyptians. abroad in 1976, rising to 1.2 million in
1985 (Ferghany 1988). The bulk- of the labor that was bid away from
the domestic economy originated. from agriculture and construction.
The growth in. external migration extended beyond the boom years,
driven in the more recent period by demand from Iraq.
The period- from 1973 to 1982 was thus associated with a
combination of external migration,. growth in public employment,
consistent expansion of the informal sector, and broad.stability in the
share of employment in the import-substituting industry sector. The.
organized private sector (more than ten employees) also expanded
rapidly, albeit from a very small base,: but accounted for under 4
percent of net labor force growth in this period (table 8.2). These
developments allowed for high levels of aggregate.employment in the
economy in the 1970s. Despite labor force growth of around 3
334 RaguiAssaad and Shmon Commander
percent per annum, open unemployment (adjusted for seasonality)
ranged between 3 and 5 percent in the mid-1970s and early 1980s
(CAFMAS various years). The windfall was most clearly associated
with reduced 'employment . and relatively weak growth in 'the
agricultural sector (tables 8.2 and 8.3). The. agricultural labor force
declined by l.l percent per annum and;from 50 percent of the labor
force in 1973 to under 37-percent by 1982. Seasonal employment
variability fell, with peak period upswings in- labor demand being
filled by more:female and child labor (Richards -and- Martin 1983).
Furthermore, adverse shifts in the terms of trade against agriculture
did not result, as the important livestock subsector benefited from'
positive protection;
With greater internal mobility and a rapid growth in nonfarm labor
income, the boom period witnessed a narrowing of wage differentials
(a phenomenon extended in the post-boom period) across sectors.
Although the.. share of households below calorie and protein cutoff
levels was significantly higher in rural than in urban areas-7.9'.
percent. of rural households compared to 4.3 percent of urban
households in 1981/82 (Alderman and'von Braun 1984)-the data
indicate a very major fall in the share of the rural population in .
poverty (Adams 1985). Even so, survey data for 1984 suggest that
roughly. 25:to 33 percent of small-farm households (those. with less
than three acres) had incomes on or below a household adjusted
poverty line (Commander 1987).
Wages through Boom and Recession
With a windfall and an appreciation in 'the real exchange rate,
.upward pressure. on the real wage 'would. be expected. The degree to
which this occurs depends -on: relative. supply elasticities 'and
consumption weights. Both spending and resource movement effects'
would promote wage expansion. This appears to be valid in the
Egyptian context. -
Available wage data reveal that real wages tose substantially
between 1973 and 1982, except in,the government.sector, which
remained broadly constant (figures 8.4 and 8.5). The,, strongest
upward movement was for agriculture, where real wages rose by 11.
percent per annum, thereby trebling -during 1973-84. The rate of
Egypt 335
Figure 8.4 Real Wage by Sector of Economic Activity
(LE per week, 1973 prices)
9.00
7,00 .40
3.0C "e - - . : -
8.00 -r
5.00~~~~~~~~~~
4.00_ - - . - :
2.00
1.00- .
1973 1974 1975 1976 1977 1978 1979 1980 1981. 1982 1983 19B4 1985 1986 1987
-A Agri culture . ----Mansufacturing ---.-. Construction - -----Services
Sources: CAPMAS EWHW Bulletin; Assaad (1990) for construction wages; Handoussa
(1988) for government wages; Ministry of Agriculture for agricultural wages.
increase would be further enhanced if reductions in the standard
working day were incorporated. In the case of agriculture, the rate of
growth in the real wage can be largely attributed to labor supply
shocks with labor being bid away, to both the regional labor market
and the urban construction sector (Richards and Martin 1983).1o
* Hansen (1987) has rightly argued that in the Egyptian. context,
- mobility and search costs are low, which facilitates; labor transfers out
of the sector..Consequently, during periods of growth agriculture
releases labor to the rest of the economy, acting as a reservoir.during
recessions. Linked to external migration, this has resulted in a
10. The construction sector largely uses agricultural workers- for temporary
unskilled work With the boom, the temporary nature of the work was diluted,
and recenle survey data show that over 40 percent of permanent unskilled
c-onstruction labor originated in agriculture (Assaad 1990).-
336 RaguiAssaad and Simon Commander
Figure 8.5 Real Wages by Sector of Ownership
(LE per week, 1973 prices)
8.00
7.00.
..0
4.00 . * .
-3.00.
2.00
0.00 .r - - ---..
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 19B4 1985 1986 1987
Year
Govenmuent - Public enterprises Pi- - - Prvate sector
(10+ employees)
Sources: CAPMAS; .Handoussa (1988).
procyclical movement of wages where wages move above value added
trend in boom years and below it in the troughs. The fact that
agricultural wages continued to increase to 1985 can in part be
attributed to relatively strong extemal demand for. labor from Iraq and
Jordan in the 1980s.
-One consequence of the relative acceleration in agricultural wages:
was a major narrowing in wage and income differentials across the
sector. For other sectors, wage growth was significant, particularly in
private construction: and manufacturing. The government and public
enterprise sector shows some wage drift, but was mainly characterized
by implicit indexation to the cost of living. With the organized private
sector bound by publicly determined wage setting.rules7 this meant
that private sector wages effectively tracked those of the public
enterprise sector up to 1982.
The upward trend in real wages was drastically reversed post
1982/84. Indeed,, in construction the downturn came as early as 1979.
By 1987 construction wages were below 1974 levels. The declining
Egypt 337
-trend in other competitive markets-such as agriculture-is likely to
be sustained, albeit with some lag. As the growth in agricultural wages
was associated with regional migration, the consolidation of the
livestock subsector, and an expansion in rural nonfarm, particularly
public, employment, all of Which have contracted, this explains the 16.
percent per annum decline since the peak in 1985.11 For the
government and public sector, the trend has been unequivocally
downward, with real wages for government employees less -than half
their 1973174 levels. The contraction has been most severe for white
collar workers (see figure 8.6) and has been accompanied by higher
turnover levels and multiple. job holding (Handoussa 1988). Apart
from a contraction in differentials across- skill, levels,, a narrowing of
wages between the public and private.sectors has also occurred, as in.
the manufacturing sector. Most -striking has been! the sharp redu'ction-
in the differential between nonagricultural and agricultural wages (see,
figures 8.4 and. 8.5).
Given the size of' the public sector and the relationship of the.
organized' private sector's wage-policy to public sector wages, the
wage setting mechanism-is clearly critical. Here, a number. of factors
are -pertinent. First, the wage path from the early 1960s appears to
have been determined independently of productivity, being structured
.by noneconomic considerations. For manufacturing wages. over the
period 1973-81/82, there appears to be -a negative association with a
productivity term-measured as labor productivity-with the level of
money wages given by consumer prices and policy (annual growth in
nominal wage costs per' unit of output in manufacturing almost
'exactly tracked the CPI between 1974-and 1982).,With the limited
information that is available and taking the nonfinancial public
enterprise sector as a whole for the years 1973-83/84, until 1980/81
11. By 1981182. 18 to 20 percent of rural sector employment was public
employment (CAPMAS various years).
338 Ragui Assuad and Simon Commander
Figure 8.6 Real Wages by Sector of Ownership and White Collar/Blue
Collar Status
(LE per week, 1973 prices)
12.00
10.* - - :
8.00. -J;__,<^ '^ ,._. ,,
4.00 . _�
2.00
4.00
2-na~~~.
1973 1975 1977 1979 1981 1983 1985 1987
Year
Publicewhite ----Publicblue Prvate white -Private blue
edlar collar callar(104: collar(10+.
employees) emplayces)
Sources: CAP.MAS; Handoussa (1988). :
wages and productivity clearly moved broadly together, with wage
increases significantly exceeding cost of living adjustments. At that
point, productivity growth began to lag wages, which in turn lagged
the cost of living. HIowever, from the early 1980s onward, the
widespread use of incentive payments in some public enterprises and
in the private sector has meant that base wages can be a considerable
distance from actual wages, making meaningful estimation yet more
problematic.
Attempts to estimate either a basic or augmented Phillips curve
equation have been largely fruitless. Standard, underlying wage.
equations relate nominal wage growth negatively to the rate of
unemployment and as a positive function of the rate of consumer
price change with the latter term commonly lagged. Estimating a
nominal wage equation for manufacturing over the period 1972 to
1984 for Egypt, the unemployment term has not only the; wrong sign,
Egypt 339
but is also insignificant. The same was true when an economic activity
term was. inserted in place of the unemployment term. The.lagged
consumer price term was, however, significant, with an apparent 0.6
elasticity of wages to lagged prices.12
To summarize, all sections of -the labor market have been
characterized, to varying degrees, by a collapse in real wages after the
early 1980s. With the exception of agriculture and private sector
manufacturing, wages are now substantially below prewindfall levels
and falling. In the govemment sector, in particular, this has promoted
.absenteeism, moonlighting, and low morale. Downward pressure on
both public and government sector wages has been the main
adjustment mechanism, given the inability of achieving direct quantity
adjustments. The organized private sector appears largely to have
followed the same path. Layoffs, alongside lower capacity utilization,
have characterized the small-scale private sector, which has been
untrammeled by labor legislation.
Employment After the Windfall
The lack of reliable data restricts the robustness of any conclusions
concerning employment trends since 1984. Data from the 1986
population census point to an unemployment rate of about 12 percent,
as compared to 6 percent in 1982.13 Quarterly data (Labor Force
Sample Survey) for 1987 and 1988- indicate a fairly stable urban
unemployment rate of around 11 percent and a rural unemployment
rate fluctuating between 4 and 9 percent. Agriculture has offset this
12. The basic equation being:
w,- 0-aU-a Ut+apt+ (1-C2)&t : '
The best fit for the period 1972-84, using the rate of change of nominal
manufacturing wages as the dependent variable and the. rate of
unemployment and lagged consumer pnce change as independent variables,
was as follows (t - statistics in parentheses);
wmt = .03(1.05) + .04(1.17)U+ .6(1.7) Pt-i
R2 0.70; F =11.5; DW =1.89
13. The 1982 data are derived from the Labor Force Sample Survey. Some of
the discrepancy can be attributed to seasonal agricultural unemployment, with
the census being taken in a slack month, November, and the LFSS in May, a
peak employment period.
Egypt 339
but is also insignificant. The same was true when an economic activity
term was inserted in place of the unemployment term. The lagged
consumner price term was,. however, significant, with an apparent 0.6
elasticity of wages to lagged prices.12
To summarize, all sections of the labor market have been
characterized, to varying degrees, by a collapse in real wages after the
early 1980s. With the exception of agriculture and private sector
manufacturing, wages are now substantially below prewindfall levels
and falling.' In the government sector, in particular, this has promoted
absenteeism, moonlighting, and low morale. Downward pressure on
both pub!ic and. government sector wages has been the main
adjustment mechanism, given the inability of achieving direct quantity
adjustments. The organized private sector appears largely to have
followed the same path. Layoffs, alongside lower capacity utilization,
have characterized .the small-scale private sector, which has been
untrammeled by labor legislation.
Employment After the Windfall
The lack of reliable data restricts the robustness of any conclusions
concerning employment trends since 1984. Data from the* 1986
population census point to an unemployment rate of about 12 percent,
as compared to 6 percent in 1982.13 Quarterly data (Labor Force
Sample Survey) for 1987 and 1988 indicate a fairly stable urban
unemployment rate of around 11 percent and a rural unemployment
rate fluctuating between 4 and 9 percent. Agriculture has offset this
12. The basic equation being.
- a0 - a1Ut +aUpt + (l-Q2)Pt_1
The best fit for the period -972-84, using the rate of change of nominal
manufacturing wages . as the dependent variable and the rate of
unemployment and lagged consumer price change as independent variables,
was as follows (t - statistics in parentheses);
wmt = .03(1.05) + .04(1l7)U+ .6(1.7) Pt-I
R2 - 0.70; F - 11.5; DW = 1.89
13. The 1982 data are derived from the Labor Force Sample Survey. Some of
the discrepancy can be attributed to seasonal agricultural unemployment, with
the census being taken in a slack month, November, and thi LFSS in May, a
peak-employment period.
Bgypt 341
of. the unemployed. This constituted an "employment overhang" of
roughly 9.5 percent of the labor force in 1.986 (see figures 8.7 and
8.8). The vast majority - of the unemployed--47 percent of
unemployed males and 70 percent of unemployed females--are
concentrated among holders of intermediate . diplomas.-
Unemployment rates among graduates with intermediate diplomas or
above range--from over 20 percent for urban males to 50 percent for
rural females. In contrast, unemployment rates for illiterates range
Only from 1.5 to 5.8 :percent.
To some extent, the graduate. employment overhang-and hence
the unemployment rate-is exaggerated. .A significant proportion of
male graduates in the queue do have work, but see it as temporary.
The state is the. preferred employer, not for wage reasons, but for a
combination of status, security, and benefits, such as free medical care
the~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Unmly -.b
Figure 8.7 Distribution of the Unemployed by Educational
Attainment, Urban/Rural Status, and Sex, 1986. .
90
80
70
60D
~40
30
020tL W
Illterate Read and Primary Below Intermediate Above University
write intermediate intermediate
Educational attainment
- U Urban males 1 Urban females U Rural males 13 Rural females
Note: The below intermediate levcEl is equivalent to three years of schooling after
primary school,.the intermediale level is equivalent to secondary school, and the
above intermediate level is equivalent to two years of education after secondary school.
Source: 1986 Population Census, results of 20 percent sample.,-
342 Ragui Assaad and Simot: Commander
Figure 8.8 Unemployment Rate. by Education'al Attainmnent,
Urban/Rural Status, and Sex, 1986
:50
45 -
40
35 -
-30
~25
Ii
C20
15
C 10A1
illiterate Read and Primary Below Intermediate Above University
write intermediate intermediate
Educational attainment'
U Urban males - Urban females U Rural males o Rural females
Note: The below intermediate level is equivalent to three years of schooling after
primary school, the intermediate level i's equivalent to secondary school, and the
above intermediate level is equivalent to two years of education -fter' secondary
school.
Source: 1986 Population Census, results of 20 percent sample.- '
and priority access to subsidized goods and services. This results in a:
relatively low dropout rate from the queue: only 45 percent of
university graduates and 35 percent of intermediate diploma holders'
drop out during the waiting period. This low rate can be attributed to
the lack of alternatives for female and rural graduates (widespread bias
against hiring womnen in both the public and private sectors can partly
be attributed to legal requirements for provision of paid -maternity -
leave and unpaid leave for child care). As figure 8.8 demonstrates,
unemployment rates among femalc and rural graduates are much
higher than: among their male or urban counterparts. For this group,
the government is in effect the employer of last resort.
Egypt 343
The implications of the graduate employment guarantee for the
future distribution of the labor force and for the size of the nortmarket
sector are alarming. Table 8.4 summarizes a number of projections up
to the. year 2000. These projections are based on a number of
assumptions: an- annual dropout rate from the government
employment queue of 20 percent of university graduates two years
after graduation, and a 15- percent dropout rate for intermediate
graduates.14 Rates of growth in categories of graduates are based on
1981/82 to 1986/87 admission- rates, while nongraduate public
-mployment is assumed to remain constant.15 Three scenarios are
projected: the first assumes government growth equal to labor fofce
growth (2.6 percent per annum), the second assumes 1 percent per
annum growth in public employment, and the third assumes 4 percent
per annum growth in public employment (the trend rate for the -first
half of the 1980s).
The -first scenario yields graduate unemployment rising to -16.7.
percent of the labor force by the end of the period. Waiting periods
rise from the current five to six years to eight to nine years. Under the
second (low growth) scenario, graduate unemployment rises to 18.9
percent with public employment still accounting for 23 percent of the -
labor force in the year 2000. Under the third (high growth) scenario,
graduate unemployment and waiting periods rise slowly or not at all,
but public employment mushrooms to over one-third of the labor
force. Obviously- such projections ignore,. for example, weaker
crowding-out effects from a fall in government expenditure through
adopting scenario 2, and hence any nongovernment employment
multiplier, but in a restricted way they uinderscore the weight of the
employment obligation and its key importance. Apart from-the
_~~~~~~~~~~~~dt prvie by th Mii of
-14. This is derived from recent historical data proided by the Ministry of
* Manpower and Vocational Training that relates applications to employment
and the length of queueing. Dropout rates in the projections were not assumed
to rise faster due to the hard core of female graduates and the. difficulties,
particularly for intermediate diploma holders, of finding other employment.
15. Admission rates for-university and intermediates graduates in the period
1981/82-1986/87 were 3 percent and 4.5 percent, respectively; Nate also that 30
.percent af university graduates and 6 percent of other graduates are assumed
to be hired directly for ministries, while the rest are allocated after queueing
from the centralized pool;.
344 Ragui Assaad and Simon Commander
Table 8.4 Employment Implications of the Graduate Employment
Guarantee Scheme, Projections to the Year 2000
Year
Category Scenario. 1988 1990 1995 2000
* Public employment as 1 29 29 29 29
share of labor force 2 28 27 25 23
* (percent)a . 3 29 30 32 34
Unemployed gradutes as 1 10.6 12.1 14.9 16.7
.share of labor force. 2 11.0. 13.0 - 16.5 18.9
(percent) . .3 10.3 11.3 12.9 14.6
Waiting period (years)b 1 5.7 6.5 7.7 .8.5
. - - - .2 : 5.7 6.7 9.7 12.5
0 - 3.- S4 5.7 5.7 5.7
* Size of labor force (thousands) . 13,458 14,140 15,998 18,100
1987 = base year
Notes: Net growth rate of Scenario 1 Scenario.I Scenario 3
government employment: 2.6% 1.0% 4.0%
- : a. Includes both government administration and public enterprises.
b . Weighted average of university and intermediate-level graduates.
scheme's budgetary imiplications, the.costs enc'ompass chronic skill
MISmnatching and an inappropriate balance between'the size of the
market and nonmarket sectors. Abolition of the employment
guarantee is an effective precondition'.for a longer-term shrinkage in
the size of the nonrnarket sector.
The -weight of the government and1 public sector. in total
employment and the extension of e n to the organized
private sector has yielded an adjustment process through the wage
rather than via quantity adjustment. InAthis regaord, adjustmet in a
more classical sense has been left to the residual flex-price markets
character ized by labor mobility and an absence oflabor legislation.
This mainly boils down to the informal private sector, small-scale
pvmanufacturing, construction and, inj A somewhat different way,
rather tha vad -nottiyhowever,nnecesh s regar, dutetin
agnculture. Adjustment costs do not, however, necessarily register as
rising open unemployment. That is reserved largely for graduates
awaiting government jobs. Rather, falling real wages accompany lower
Egypt 345
levels of capacity utilization and rising underemployment, where the
latter is defined as the gap between desired and actual work time at
prevailing wages. This is clearly the case in construction, and we can
assume that the broad behavior of that sector is generally mimicked in
other flexible markets.
Flexible Market Adjustment: Construction
The dynamics of the Egyptian construction sector can be derived
from the level of aggregate investment in the economy and from labor
supply variations stemming from outmigration and subsequent return
migration. At the same time, the sector is important in terms of its
relationship with the agricultural labor market (see Assaad 1990 for a
more detailed account of the 1988 Construction Workers Survey, on
which this section is based),
Sectoral output has been strongly procyclicalt varying directly with
changes in aggregate investment. Investment demand has accounted
for 90 to 95 percent of construction output at any given time, while
the latter provides the largest (47 to 53 percent) component of gross
fixed investment. Consequently, the fall in investment from 28 percent
of GDP during 1978-82/83 to below 20 percent between 1986188 was
directly translated into a reduction of real value added in the sector.
Employment. in construction expanded substantially in the boom
years despite large-scale migration to other Arab countries. This
nigration resulted in skill mismatching and selective labor shortages,
which further forced wages up, inducing a stronger internal migratory
shift of workers. out of agriculture. After 1983 (figure 8.9) falling
value added did not-at least superficially-throttle off employment.
.However, the.apparent fall in value added per worker conceals the
principle response: lower levels of labor force utilization. For
construction, the employment cycle involves partial utilization of the
available-labor force in periods of slow growth, giving way to rapid
upturns in labor demand during periods of higher, investment-led
.growth, Underutilization of labor is replaced by shortages,;which are.
exacerbated by migration abroad in the boom years. The lagged
supply response in turn delivers, underemployment as the downside of
the cycle emerges. Labor is in effect variable capacity under
conditions of fixed coefficient production technology.
346 Ragul Assaad and Simon Commander
Figure 8.9. Trends in Construction Employment and Real Value
Added (index, 1980181 = 100)
16 - -
.120
100 . .
80
60
* 40 _
20
1970171 71-72 1973 1974 197' 1976 19T7 1978 1979 ' 8OIS S812 '93 83184 '84/85 ''5/5 8SS7
- Emplam. - ' V,.eadded
Sources: Assaad (1990); CAPMAS.
A simple distribution of construction labor by private and:public
sectors indicates, that by 1984, roughly 20 perce.nt of the labor force
were in the public sec'tor. That sector is marked by.the.dominance of
large general contracting firms. In contrast, over 75 percent of the
labor force was in the small-scale (less than iC) employees) private
sector. This subsector accounts for an even larger proportion of 5sie
labor (90 percent) and is marked by its flexible wage and employment
structure. Downturns in the investment cycle are transmitted directly to
this subsector. In other words adjustm'ent costs are.squeezed largely
out of the private, small-scale subsector segment. A similar effect
occurs more generally'with the entire informal sector.
A look at 'co-ntracts in construction reveials, as expected, a dual
market. At least 90 percent of public sector. workers hav'e formal
contractual arrangements, with over 70 percent guaranteed permanent
employment.-In c'ontrast, 97 percent of private sector workers atre
Egypt 347
hired on a casual basis without any formal contractual arrangement. It
is these workers who carry the brunt of adjustment. The period of
effective labor shortages of the 1970s has been conclusively reversed.
By 1988, casual construction workers on average could expect to work
no more than two-thirds of their available labor time (table 8.5).
Urban workers fared slightly worse than their rural counterparts, as did'
skilled workers compared with the unskilled. In short, the windfall-
generated labor scarcities and, low frictional unemployment have given
way to a labor supply glut and longer unemployment spells, ranging
from about 15 days on average for laborers to nearly 55 days for.
craftsmen and assistants.'
The. impact on wages has been profound. All subsectors show
rapidly falling real wages after 1984. While between 1972 and 1982
real wages climbed by 4.5 percent per annum, in the private sector real
wages fell by 3.8 percent per annum between 1982 and 1987, and
more than 9 percent per annum after 1986 (figure .8.10).
Table 8.5 Average Employment and Unemployment Spells for Casual
Wage Workers in Construction, 1988 -
Average Average
employment unemployment Employment Nunber of
Category spell (days) spell (days). ratioa (percent) observations
Skill level
Craftsman . 80.4 52.9. 60 262
Assistant 69.9 56.3 :55 96
Apprentice 103.7 36.1 74 . 58
Laborer 46.7 14.6 76 141
Urban/rural status
Urban 81.1 44.6 . 65 :335
Rural 59.4 - 23.8 71 222
Total 72.5 36.3 67 557
a. Employment ratio - El(E4-U), where E = the average employment spell and U = the
average unemployment spell.
Source:.Assaad (1990).
348 Ragui Assaad and Simon Commander
Figure 8.10 Real Wages in the Construction Sector
(1973 LE/week)
10.00 -
3.00 - .
2.00 =
6.. . . ..
4,00 - - -
3.00 -
1.00
o.oo - - ! 1 1
- 1973 1975 -1977 1979 1981 1983 1985 1987
Year
Public cntriscs - Private (iO+ 'asualIabor.
employcos)
Sources: Assaad (1990); CAPMAS.
Accommodating "capacity" change 'and assuming near, full
employment in 1982 and (optimistically) 70 percent employment in
1987, the rate of decline in monthly earnings nses to 11 percent per
annum.
The evidence presented above indicates unequivocally the nature of
the adjustment mechanism in a flexible market with fixed technology
and limited factor substitutability. Falling' labor utilizationmruns
alongside substantial downward pressure on the real wage. These
signals constrain the inflow of migrant labor from both upper Egypt
and agriculture. Survey data sbow that exit patterns from construction
differ widely by skill category, but that for unskilled labor, movement
out of the sector was toward services and to a lesser extent agriculture.
Egpt- 349
With a lag-and under current sectoral supply elasticities-this
translates into expanded unemployment and underemployment in the
* receiving sectors.
Conclusion
The oil and associated revenues windfall facilitated the maintenance
.of a near-full .employment economy. This was achieved. in part
through growth, in part through regional migration, in part through
maintaining rapid growth in the government and public sector work
force. In the latter sector, wages were driven primarily by adjustment
to the cost of living. These policies established major rigidities' in the
labor market and efectively, squeezed the cost of adjustment to lower
external transfers into the economy onto, in the- first instance, the
small-scale manufacturing and informal sectors, and subsequently
agriculture. In all cases, the highest- costs-as manifested in
unemployment and. income dependence-have been placed on new
entrants and women in the labor force.'
Egyptian statistics grossly underestimate the rate of femile- labor
force participation, and hence the aggregate activity rate (see
Commander 1987). Nevertheless, within this truncated measure,
women's share in unemployment in 1986 was twice as high as their
share in the labor force (23.5 percent to 11.6 percent respectively).
The female unemployment rate is strongly correlated with the
recruitment of graduates in .the public sector, given the limited
employment opportunities for educated women elsewhere. in the
. economy. . Women are also more able to stay in the queue for
government employment as they are generally secondary earners in
* the household.
The-windfall gains of the 1970s were largely. appropriated directly
by the state. The fiscal expansion this facilitated had predictable-
* Keynesian effect, but the accompanying structure of macroprices-the
exchange and interest rates-compressed traditional exportables'and
skewed -investment toward relative capital intensity. The. exportables
subsector of agriculture was the lagging. sector. Given the trade -
regime, deindustrialization did not result. Moreover, the strong public
spending effect was not necessarily associated with a.crowding out of
.private investment. The latter grew strongly in both absolute snd
350 RaguiAssaud and Simon Commlader
relative terms, admittedly from a low base. Shafik (1989) argues that
over 40 percent of public investment was allocated to infrastructure
through the.windfall period, and in an appropriately specified model,
this yields strong complementarities: a crowding-in effect with private
investmnent (Chhibber and van Wijnbergen 1988 teach similar
conclusions for Turkey).
- The employment effects of the windfall were limited, as .indicated
by tables 8.2 and 8.4.. The organized private manufacturing sector
expanded employment roughly two percentage points above
aggregate labor force growth during 1973-82, but most of the growth
was in services and construction. The organized private manufacturing
sector grew at only 1.8 percent per annum during the same period. A
* rising capital/labor ratio, alongside. low productivity of capital, has
meant that. the employment multiplier from private investment in
* manufacturing was low. There was a stronger employment multiplier
in the small-scale private sector,. the precise magnitude of which is
hard to. capture with available data. For public. sector manufacturing,
employment grew. at 4 percent per annum between 1973-82.
Productivity per worker declined significantly after 1979, despite. the
fact that hiring rules for public enterprises were more flexible than
elsewhere in the public sector.16 Wage growth was determined
independently of productivity. For the government sector, indexation
to the cost of living appears to have,held until the early 1980s.
Thereafter, wages have been sharply reduced, but the sector still
remains massively overmanned (Hansen and Radwan 1Q82 suggest
overmanning of at least 40 percent in the mid-1970s; the trend has
obviously been rising).
However, the demand for government and publicN sector,
employment cannot be attributed to the wedge between public wage
rates and the supply price of labor. Govemment wages are low, in both'
absolute and relative terms. Nonwage benefits probably continue to
- drive the demand for government employment. The effect on the
reservation wage is not obvious. Anecdotal evidence suggests that
governmentVemployment is often akin to ghost employment. Where
16. Productivity per worker is measured for the entire nonfinancial public
enterprise sector, and hence does not accurately reflect productivity in the
public. manufacturing sector.
Egypt 351
this is the case, the ultimate effect may be to lower the effective
reservation wage, but the lack of solid information on the dynamics of
the multiplc job-holding market compromises any conclusions.
The expansion of the public sector in the windfall reflected strong
procyclical government expenditure policy and was largely
dissociated from underlying permanent revenue.. Further, strong
employment growth was maintained countercyclicatly -up to -1985.
This* was partlX a function of the- graduate employment guarantee
scheme. The latter-under recent queueing conventions-alone yields
an annual growth rate of over 4 percent for government employment.
Almost full employment has, until recently, been the outcome of the
combination of this expansion and sizeable external: migration.
Neither option is sustainable. 'The recent acceleration in
unemployment is in effect the gap between the falling rate of growth
in private. sector employment and the now constrained ability of the
public sector to absorb the residual.
Adjustment measures for reducing fiscal imbalances involve
demand dampening. Austerity, falling real. wages, and currency
depreciation boost .external competitiveness. Adverse employment
effects would, in theory, be counteracted over the medium term by
expansion of tradables output.. In agriculture, at least, the ;exportables
subsector-has higher labor demand.
- Positive real interest rates and trade regime reforms could stimulate
switching toward higher labor intensity, but this scenario. is impaired
by a number of factors. First, the relative, price swing toward tradables
has been too small. In agriculture, for example, relative prices still
favor nontradables. Second, the tariff structure and interest rates
continue to deliver a bias toward capital. Third, even' assuming more
coherent macroeconomic adjustment. policies, the labor mobility
assumptions will not iold. The expansion of government employment
locks over 28 percent of the labor. force into the nonnarket sector, at
least on a first job basis. Aside from the productivity implications, any
hypothesized shift of labor into tradables production would be
severely constrained.- A combination of public employment and the
educational system yields a labor force inappropriately matched, in
terms of both skills and expectations, to the requirements of structural'
adjustment.
352 RaguiAssaad and Simon Comnmander
This analysis of the response. of the Egyptian labor market to boom
and recession indicates a range of employment and wage setting
mechanisms. The public sector and formal* private. manufacturing
sector are characterized by relatively inflexible empioyment rules.
Minimum wages are almost meaningless, but hiring and firing rules
unfoiced by the Labor Office restrict turnover, and in the case of the
private sector, create positive incentives for de facto small-scale units.
Combined with the interest rate and tariff structure, this results in the
organized private sector being characterized by high capital/labor
ratios. In the case of both the public enterprises and the organized
private sector, real wages have fallen sharply since 1985. Wage
flexibility may hold, but employment is regulated only via the rate of
recruitment, and in the case of the public sector, that rate is itself not
necessarily a decentralized variable. For the nonorganized. sectors,
markets have more standard competitive properties, albeit with
nominal wage inertia. -The dominant features of the adjustment
process to date, particularly the reductions in the public wage, result in
- further downward pressure on the wage in flexible markets. The
elasticity of earnings, in the latter with respect to the nonabsorption of
labor in other sectors of the economy is significant and negative. In:
due course this yields a fall in aggregate demand, but at a high cost.
UUnemployment and sharply. falling disposable income can only be
exaggerated (in the absence of new growth in net transfers into the
economy) by leaving the present labor market rigidities untouched.-
Egypt 353
APPENDIX
Figure 8.A1 Egypt Investment Deflator and its Components,
1973-87
(index, 1973 = 100)
SW
7C0-
600 -
Soo.
400-
1973 1975 1977 1979 1961 1983 19A5 1967
Year
-x-tnlCostructian 0 lmponi + DIuestk * nvestment C -DP
capitha gods capiti goads
354 Ragui Assaad and Simon Commander
References
Abdel Fadil, M. 1980. The Political E'conomy of Nassertism: A Study
in Emtiployment and Income Distribution in Urban Egypt,
1952-1972. Cambridge, U.K.
Adams, R. H. 1985. "Development and Structural Change in Rural
Egypt? 1952 to 1982." World Development 13(6): 705-723.
Alderman, H., and J. Von Braun. 1984. The Effects of thle Egyptian
Food Subsidy Systems on Income Distributiont and
Consumption. Report 45. Washington, D.C.: International
Food Policy Research Institute.
Amin, G., and E. Awny. 1985. International Migration of Egyptian
Labour. Report MR 108e. Ottawa: International Development
Research Center.
Assaad, R. 1990. The Structare of thie Construction Labor Market and
its Development since the Mid-1970s. Working Papers in
Planning No. 85. Ithaca, New York: Cornell University.
Braga de Macedo, J. 1982. "Currency Diversification and:Export
Competitiveness: A Model of the Dutch Disease in Egypt."
Journal of Development Economics 11(3j December): 287-
306.
CAPMAS. 1978. Population Census, 1976. Cairo. .
. 1988. Population Census, 1986. Provisional Results.
Cairo.
- - . t____ Various years. Labor Force Sample Survey. Cairo..
Chhibber, A., and S. van Wijnbergen. 1988. "Public Policy and
Private Investment in Turkey." Washington, D.C.: World
Bank. Processed.
Collier, P. 1988. "Macroeconomics aind the Labor Market." Oxford,
U.K. Processed.
Conimander, S. J, 1987. The State and Agricultural Development in
Egypt since 1973. Ithaca, New York: Ithaca Press.
Corden, W. M. 1989. "Macroeconomic Adjustment in Developing
Countries." World Bank Research Observer 4(1): 51-64.
Egypt 355
Corden, W. M., and J. P. Neary. 1982. "Booming Sector and
Deindustrialization in a Small Open Economy." Economic
Journal 92(December): 825-848.
Dervis, K., R. Martin, and S. van Wijnbergen. 1984. Policy Analysis
of. Shadow Pricing, Foreign Borrowing and Resource
- Extractioh in Egypt.. Staff Working Paper No. 622.
Washington, D.C: World Bank.
Dethier, J. J. 1989. Trade, Exchange Rate and Agricultural Pricing
Policies in Egypt, 2 vols. Washington, D.C.: World Bank.
Ferghany, N. 1988. In Pursuit of Livelihood: A Field Study of
Egyptian Migration. Beirut, Lebanon: Centre for Arab Unity
Studies4 In Arabic.
Handoussa, H. 1983. -Public Sector Employment and Productivity in
the Egyptian Economy. Geneva: International Labour
Organisation.
- - . _____1988. "The Burden of Public Service Employment
and Remuneration: A Case Study of Egypt." Geneva:
International Labour Organisation. Processed.
. 1989. "Egypt's Investment Strategy, Policies and
Performance since the Infitah," Processed.
Hansen, B. 1987. "A Full Employment Economy and its Responses
to Elxternal Shocks: The Labor Market in Egypt from World
War 2." Washington, D.C.: World Bank. Processed.
Hansen, B., and S. Radwan. 1982. Employment Oppottunities and
Equity in Egypt. Geneva: International. Labour Organisation.
Neary, J. P., and S. van Wijnbergen, eds. 1986. Natural Resources
and the Mncroeconomy. Oxford, U.K: Blackwell.
Richards, A., and P. Martin, eds. 1983. Migration, Mechanization
and Agricultural Labor Markets in Egypt. Boulder, Colorado:
Westview.
Shafik, N. 1989. "Private Investment and Public Policy in Egypt,
1960-1986." Oxford University, D.:Phil. thesis.
World Bank. 1983. Egypt: Issues of Trade Strategy arid Investment
Planning. Report 4136-EGT. Washington, D.C.
GHANA
P. Beaudry
N. K. Sowa
The. predominant .view underlying Ghana's Economic Recovery
Program (ERP), initiated in 1983, is tluat the country was most likely to
achieve - sustained growth by relying on market forces. Ghanaian
policym'akers believe that changes in relative prices, which are central
to the ERP, incite economic agents to reallocate resources as indicated
by market signals. However, the success of such a program depends
primarily on. the extent to which labor is allocated through the market
mechanism. If wages and labor mobility are insensitive to changing
market conditions, the ERP is unlikely to achieve its goal of structural
adjustment. This chapter examines whether the ERP.'s reliance on
market forces to reallocate and absorb labor was well-founded, and
whether the improvements in the Ghanaian economy since 1983 were
helped or hindered by the actual functioning of the labor market. This--
chapter also examines the relative impact the ERP has had on the
different segments of the labor market.
Most of our labor market analysis relies .on data from -the. first
round of the Ghana Living .Standards Survey (GLSS), -.which was
collected in 198788. This data set-contains information on. 1,526
households that encompass 7,637 individuals. The data set is-
especially well-suited for our analysis as its coverage of both the
allocation of. time and of the sources of . household income are
particularly extensive (for information on the design of the GLSS see
The :authors would like to thank Harold Coulombe, who provided excellent
research assistance for this chapter, and S. Horton, L. Dudley, A Martens, and R.
Kanbur fir their helpful comments.
357
358 P. Beaudry andNM K Sowa
Scott and Amenuvegbe 1989). However, because this data set consists
of only one cross-section,' many of the central questions associated
with the adjustment process can only be addressed indirectly.1
Therefore, we have also used aggregate time series data on earnings
and employment to complement the household survey data when
relevant. These data come mnostly from the Statistical Service of the
Republic of Ghana.
Historical -Background
Ghana's 1983 economic crisis was a culmination of several years
of decline. Some authors place the start of the decline soon after
independence, which was granted in 1957, while others place -it in the
1970s. Indeed,-some Writers' even think the decline started in the pre-
independence era, but only became acute, and therefore apparent, in
the postindependence' era.
:Nevertheless, at independence Ghana still had a vibrant-economy.
The country was a'world leader in the production of cocoa, and gold.
and timber exports were also good sources of foreign exchange.
earnings. It had a huge buildup (about US$269 million) of external
reserves, and .the balance of payments was. not problematic. At this
time, expansion/contraction in the money supply was tied 'to a
surplus/deficit of the balance of payments. Thus, the government
avoided indiscriminate expansions in the money .supply and held
inflation in check. The external value of the currency was relatively
strong, and was exchanged at par with the pound sterling.
Soon after independentce, the' Nkrumah government began a
number of projects aimed at -import substitution and semi-:
industrialization. The government embarked -on developing
infrastructure, such as roads, electricity, and piped water, in the belief
that such development would attract foreign investors into the country..
Basic 'education was' made free and compulsory-for all children of
school-going age.
Not. long after these projects began, .the price of Ghana's chief
foreign exchange earner, cocoa, started to fall on the world market. By
1. Since the GLSS data were collected between September 1987 and March 1988,
the sample does not provide a completely repiesentative sample of yearly activities,
and may therefore bias some of our inferences.
Ghana 359
the mid-1960s, the country's net reserves were in deficit. The
government borrowed heavily from the banking system to finance its .
growing fiscal deficit, which caused expansions in the money supply
that were not matched by growth in production. Thus, from an
inflation rate of virtually zero at independence, the rate had climbed to
20 percent by 1965. The falling price of. cocoa and the bad reserve'
position meant a shortage of foreign exchange. The first obvious
symptoms of the decline in lhe economy appeared in 1964, when
queues started forming in shops.and even at stadiums.(which were;
used as market places) for basic imported commodities like sugar,
milk, and soap. These harsh economnic conditions precipitated the
1966 coup that toppled the Nkrumah regime.
Unfortunately, the various governments that followed Nkrumah did'
not implement programs for long-term solution of the economy's
problems. Successive :governments persisted in providing ad hoc
solutions to satisfy the public's demand for consumables. This lack of
proper economic direction was probably one of the main internal
causes of the decline in the economy that led to the crisis of 1980s.
Ghana's economy depends heavily on agriculture, andnsis therefore
affected by uncontrollable factors like the weather. Agriculture -
contributes about 47. percent of the GDP -and employs about 53
percent of the labor force (much of the data for this section are drawn
from World Bank 1984b). Cocoa earns about:'70 percent of the
country's foreign exchange. Since the 1970s, bad weather and
inappropriate policies have led. to a decline in cocoa output. By the
beginning of the 1980s, cocoa output had dwindled to about'half of
its 1960 level. Table '9.1 provides some salient macroeconomic-'
indicators for 1978 onward. The price distortions.that affected the
whole economy were particularly harsh in the cocoa sector. The
government, through the Cocoa Marketing Board, -determines the
price of cocoa based on the world price and the exchange rate. of the
cedi. Thus, when for a long period the cedi remained at a fixed rate
and was grossly overvalued, cocoa fanners were grossly underpaid.
Furthermore, as the world price fell, Ghana's foreign exchange.
earnings fell; exacerbating the decline in farmers' earnings. These
factors led. some of the farmers to shift away from cocoa to other
agricultural products. Some of the farmers also smuggled cocoa . -
Table 9.1 Macro- Indicators, 1978-88
Indicator 1978 1979 1980 1981 1982 1983 1984 1985 1986 7987 1988
GNP per capita.
(percent growth) 0 -5.0 -2.9 -5,8 -10.0 -7.1 7.6 1.5 2.0 1.0 n.a.
Inflation
(percent growth) 73 55 S0 117 22 122 40 10 25 . 36 37
Money
(pctccnt growth) 69 16 34 51 23 40 54 46 48 45 13
Cedi end year
exchange rate
(per dollar) 2.75 2.75 2.75 2.75 2.75 30 50 60 90 115 227
Government balance
(cedis millions) -1,772 -1,696. -1,646 -4,440 -5,13k0 -4,514 -4,053 -5,453 -2,966 8,911 4,89
Trade balance
(USS millions) . 112.5 262.6 195.3 -243.6 18.3 .60.6 32.9 -36.3 .60.9 -124.7 n.a.
Cocoa produccion
(thousand tons) 265 281 254 220 179 158 175 ZZ0 225 234 289
Terms of trade
(1980 100) . 185.6 149.0 100.0 79.8 .69,2 89.5 104.5 99.7 83.2 n.a. n.a.
Net capital formation
(1975 cedis millions) 205.8 122.5 275.7 312.2 15.4. 14.9 125.9 191.8 . 180.3 222.5 n.a.
Debt service
(percent of exports) 6.5 . 6.0 8.3 7.2 9.6 .23.3 16.4 15.6 16.7 W9a n.a.
n.a. not available
Note: Cocoa production is for the season beginning in the year indicated.
a. = estimate.
Sources: IMF International Fitancial Statistics; World Bank 1984a; Ghana Statistical Service Quarterly Digest Statistics; World Bank
World Debt Tables.
Glhana 361
across the borders to the CMte d'Ivoire where they obtained higher
prices for their produce.
The relative price of maize and other food crops rose relative to
cocoa and some factors shifted into the production of these other
crops. However, the main effect was an outward migration. Between
1970.and 1980, agricultural production declined at the rate of about
0.2 percent per annum. Note, however, that even though production
declined, the sector's share of gross domestic output increased from
41 percent in 1960 to about 51 percent in 19822 The increase in
agriculture's share occurred because other sectors of the economy
declined at even faster rates than agriculture. Between 1960 and 1982,
the rates of growth of the other sectors were'-2.4 percent for industry,
-1.5 percent for manufacturing, and -7.5 percent for services.
With Ghana being at an early stage of development, manufacturiDg
and industry have never been large sectors of the. economy.
Manufacturing is usually undertaken on a smiall scale and employs
only a limited proportion of the labor force. Manufacturing employed
about 10 percent of the labor force in 1960, and about 12 percent by
the early 1980s. This sector comprises mainly small-scale, light,
nondurable consumer goods industries, including food processing,
textile, and basic metal processing industries. The big establishments
in this sector are mostly state-owned. The main problem that plagued
manufacturing was a lack of raw materials and spare parts due to the
shortage. of foreign exchange. This shortage was in turn largely
related to. the problems in agriculture. In addition, the state-owned
enterprises were grossly mismanaged to-the extent that instead of
earning income for the state, most of them depended on government
subvention for their salaries.
These problems in manufacturing manifested themselves in
shortages on the market, and. led to inflation. By the early 1980s, the
foreign exchange shortage and the import restrictions exacerbated the
shortage of consumables on the local mark6t and worsened inflation.
Between 1960 and 1982, government revenue as a percentage of
GDP dropped from 17.5 to 5.6. This fall resulted largely from the tax
base being heavily dependent on cocoa exports, which were faIling..
Despite falling revenues, expenditures increased and the budget was
consistently in -deficit. The government kept on borrowing from the
362 P. Beaudry andN. K. Sowa
banking system to finance the deficit. This by itself was not surprising
because the financial system was not developed enough to allow the
government to borrow from any other source. Unfortunately, high
government borrowing crowded out private sector borrowing and thus
further. depressed the private ' sector's chances for growth.
Furthermore, this borrowing was accompanied by a high level of
monetary expansion that fueled inflation.-
In 1972, when the military, govermment of the National Redemption
Council led. by General Archeampong repudiated some of Ghana's-
external debts, the, country was blacklisted as not creditworthy.' Thus,
for most 'of the 1970s and early 1980s, Ghana could not attract
external loans. Whereas 'the& lack of external funds was part of the
reason why imported spare parts and raw materials were not available,
it also meant that Ghana hada 'relatively light debt burden. In 1970, its'
debt service was 1.1 percent of the GNP and 5.0 percent of exports;
by 1982, these ratios were 0.2 and 6.8 percent, respectively. Thus, at
the time the. government introduced the ERP, Ghana did not'have.
much of a debt burden. Other'low-income countries in Sub-Saharan
Africa had, on average,'a debt to GNP ratio of 13.4.
In 1982, further shocks compounded Ghana's long-run problems.
'(inflation, declining terms of trade, overvalued currency, and political
instability). The country was hit by the worst drought in its history;
bush, fires. destroyed farms and crops;' and almost every conceivable
item-food, water, electricity, and so on-was in short supply. These
problems were exacerbated, when about a 'million Ghanaians were
repatriated from neighboring Nigeria. The plight of these refugees
and the near famine situation in Ghana drew the attention of the
international community.
In an effort to solve some of these problems the government, with
the help of the IMF.and the World Bank, instituted the' Economic
Recovery Program.
Consequences 'of the Crisis
The immediate consequence of Ghana's economic decline was.the
general impoverishment. of the nation as a whole. Most indicators
point to a drop in the standard of living. Per capita GDP, in constant
1975 prices, dropped from a level of c634 in 1971 to c395 in 1983..
Ghana 363
Most people could not afford the basic necessities of life. By 1983, the
near famine situation caused most people to develop. "Rawling's
chain" (protruding clavicles). Kalabule-a situation in which
suppliers take advantage of consumers during periods of scarcity-
became the order of the day.
The index df food production per capita with 1971 as-100 dropped
to about 72 in 1982. Rough estimates (Green 1987, table 20) show a
great increase in poverty: the number of. urban people below . the
poverty line increased from 30 to 35 percent in the late 1970s to 45-to
50 percent by the mid-1980s. For rural people, the situation was even
worse. The percentage of people living under the poverty line in rural
areas increased from sonme 60 to 65 percent in the late 1970s to 67 to
72 percent by the early .1980s.
The crisis also manifesied itself in the deterioration of human
capital. Health standards, which had generally improved during' 19b6-
80, deteriorated severely in the early 1980s. Life expectancy at birth
increased from 46 years in 1970 to about 55 years by the end of the
1970s, and then droppied to 53 years in the beginning of the 1980s.
The daily calorie supply as a percentage of requirements dropped
from 88'percent in the 1970s to 68 percent in 1983.'-The drop in
health indicators in the 1980s was due mainly to a shortage of drugs,
food, and other supplies, and also to the "brain drain" in the medical
profession.
Education likewise had generally improved since independence, in
that the' number of school-age children who went to school had
increased. However, the quality of education -dropped due to-the
shortage of textbooks and other educational equipment and to the
emigration of teachers.
- Another significant consequence of the economic crisis of-the
1970s and 1980s was its effect on manpower development and labor.
The high rates of inflation were not matched by increases in the
nominal'wage. Thus, over the years workers saw their-real incomes
being eroded. This affected mainly those on wage, incomes, and
caused-most of them to take on second and -third jobs. The most
common second job was trading. At the height of the kalabule period,
those who benefited most were traders. The trading was not necessarily
in actual wares. Some people made huge profits just by knowing
364 P. Beausdry and N. K. Sowa
someone in a position to give them chits for obtaining essential
commoditie's.2 Th'ese chits were then sold to the actual traders for
cash. Such dishonest-acts did not enc'ourage manpower development
in the country. School dropouts who t'urned themselves into
"businessmen" became, better-off .than their counterparts who
continued their schooling.- Skilled personnel like doctors, engineers
and teachers, who'could not engage in these acts of kalabule, fled,
.mainly to other African countries and Asia. Nigeria wa's the, main
beneficiary of the Ghanaian brain drain,.though most- of the medical
doctors ended up i'n Saudi Arabia. Estimates suggest that Saudi Arabia
had more' Ghanaian doctors than Acc'ra. Most, teachers went to Nigeria,
as the oil bo'om in that country in the 1970s led to the eistablishment
of new schools. By the beginninig of the.1980s, Ghanaians-of every
class and skill were leaving in droves for Nigeria. When an economic
crisis hit Nigeria in the 1980s, the government expelled almost a.
million Ghanaian refugees.
Summnary of Policies Under the ERP
The government of the ProvisionalI National* Defense Council
launched the Economic Recovery Program in April -1983 as a
* edium-term plan with the following objectives (Government .of
*Ghana 1983, pp. 15-16):
12. to restore incentives for production of food, industrial raw materials and
export commodities and thereby increase their output to modest but realistic
levels;
2. to increase the availability of essential conisumer goods and improve the
distribution system;
3. to increase the overall availability of foreign exchange in the country,
improve its allocation me-chanism and channel it into selected high priority
activities;
4. to lower the rate of inflation by pursuing prudent fiscal, monetary and
trade policies;
5. to rehabilitate the physical infrastructure of the country in support o;f
directly poductive activities; anl
6. to undertake systematic analyses and studies leading towardis a major
restructuring of economic. institutions in the country.
2. One of the methods of rationing scarce commodities was by issuing notes of
rights to purchase.
Glhana 365
In pursuit of the above objectives, the government introduced the
following policies:
The exchange rate policy was aimed mainly at regularizing the
external value of the cedi. In the beginning this involved large
devaluations, and later flotation of the cur:ncy for the rate to
be determined at an auction through market forces.
* The fiscal policy was aimed principally at eliminating the high
budget deficits. This policy included widening the tax net and
cutting government expenditure.
* The investment policy was aimed at encouraging foreign
investors through a system of incentives provided by a new
investment code.
* The prices and incomes policy was aimed at removing
distortions in the economy. The pricing policy involved.
eliminating price controls, whereas the incomes policy was
intended to adjust workers' incomes to prevent the erosion of
real income by inflation.
* The divestiture policy involved the selling of some state-owned
enterprises. --
* The agricultural policy aimed at increased production of food,
selected raw materials, and export crops. -
The Macroeconomy Since the ERP
A useful first step is to distinguish between the ERP's stabilization
impacts and its adjustment impacts. As table 9.1 shows, between 1983
and 1985 inflation fell from 122 percent to 10 percent per annum and
GNP growth per capita rose from -7.1 percent to 1.5 percent annually.
Such drastic changes certainly reflect a period of effective
stabilization. However, these macroeconomic improvements were:
probably caused by good weather following the drought of .1983. For
exanple, cereal production tripled between 1983 and 1984 and the
production of starchy staples rose 40 pereant. Nevertheless, since 1985
per capita GNP growth has been maintained at i to 2 percent
annually. This period of sustained growth in GNP is in sharp contrast
366 P. Beaudry and N. K Sowa
with the five years of continual decline preceding the ERP, thus 1985
may be considered the beginning of the program's adjustment phase.
Besides growth, the other 'most notable' changes following the
implementation of the ERP were. the achievement of a trade surplus in.
1986 and a government budget surplus in 1987.
The performance of investment under the ERP has also been
relatively successful (table 9.1): in 1983, the ratio of net capital
formation to GNP was less than 0.5 percent, but by 1987 it had
reached close to 4.0 percent. Improvements in the legal regime
governing private investment have probably been an important factor.
For example, the 1984 Petroleum Explorati6n and Production Law,
the new investment code of 1986, and the 1986 Minerals and Mining
Law all created generous incentives to prospective investors.
The. ERP's lasting effect on inflation has been much less impressive
than its\effect on growth. Inflation has been:accelerating since 1985
- and bhad t9ached almost -40 percent per year in 1988. This reflects the
large increases in the money supply, which averaged more than 30
percent per year between 1983 and .1987. Moreover, the monetary
dynamics. underlying this inflationary spiral mnay be- difficult to
reverse.
The continual.depreciation of the cedi is likely to be at the center
of this spiral. The cedi depreciated by more than 800 percent between
1983 and 1988, which caused a rapid increase in the, price. of many
goods, especially in urban areas. The monetary authorities' need to
finance the fiscal deficit before 1987 had forced them to respond to
the price increases by printing money. The resulting expansion of the
money supply led to further depreciation -of the cedi,: and
consequently further pressures for price Lncreases.
Overall, the ERP's macroeconomic aspects have been relatively:
successful. However, the extent to which the ERP has provoked-real:
structural adjustment in the allocation, of labor is= unclear. The
following sections will try to address this question, but first we will
examine the characteristics, of the labor market.
Characteristics of the Labor Market'
: This section provides basic. information' on 'the structure of Ghana's
labor force (for an earlier study of the Ghanaian labor market see
Ghana 367
Ewusi 1978). Table 9.2 presents estimates of laaor, participation rates.
To avoid omitting certain segments of the labor force, we have chosen
a restrictive definition for an economically inactive person.;The
population of working age encompasses everyone aged seven or older,
and a person is classified as inactive if either (a) the person is a
student, or (b) the person has neither worked nor looked for work in
the preceding week for reasons of sickness, age, household work, or
unwillingness to work,3 All other individuals are considered active.
The table indicates that for people older that 25, participation rates.
are slightly higher for men than for women. For the prime-age males
category (25-60 years of age), the participation rates exceed 80
percent. This is quite close to the level Ewusi (1978) recorded using
the 1970 population survey. Overall, the data indicate that 42 percent
of the population over seven years of age is economically active. -
As table 9.2 suggests, the 'problem of child labor is still severe in
Ghana, especially. for females. In rural areas, 30 percent of. girls aged
7 to 16 were- reported to be economically active, and are thereore not
enrolled as students. The level for boys of the same age group -is 21
percent. In urban. areas the problem, is somewhat less severe: 14
percent of girls and 8 percent of boys are economically active. These
high figures reflect many households'- inability to live on the earnings
of the head of the household only.. As a result, the children are put to
work to supplement household income. Furthermore, the costs of
books and school uniforms are often very high relative to household
income, which. is a further disincentive to school enrollment..
Table 9.3 breaks down the active population by . employment -
sector, level of education, and sex. Household farming is by far the.
largest sector of employment, followed by household businesses.
Public sector employment and employment in the private sector
outside the household total only 17 percent of employment. The
remaining 83 percent. of the employed population work in household
businesses or farms.
3. A residual category, "other reasons" for neither working or scarching, also
qualified the person as inactive.
368 P. Beaudiy andN K. Sowa
Table.9.2 Labor Participation Rates by Age and Sex
Age Location Mate Female Total
7-16 Urban 8.3 13.9 11.2
Rural 21.4 29.5 25.2
17-25 Urban 60.0 55.7 57.9
Rural 69.5 74.6 72.3
26-45 Urban 89.3 76.8 82.2
Rural 89.4 85.0 87.0
46-60 Urban 83.7 73.1 78.7
Rural 87.7 76.5 81.4
>60 Urban - 55.0 32.5 43.8
Rural 70.0 55.5 63.3
Total Urban 39.2 38.0 38.6
Rural 41.7 46.4 44.1
Total 41.0 43.9 42.5
Source: Authors' calculations from the Ghana Living Standards Survey (1987-88).
The government sector, the state -sector, and, to a less extent, the
private nonhousehold sector all employ a much higher percentage of
educated workers than the housThold sectors (business and farming).
Of those working on household farms, 75 percent have either not
received any formal education or have received only a primary
education, while in the government sector this group represents only
14 percent of the work force. The .government is particularly
important in the employment of highly educated Ghanaians: over 50
percent of those with a postsecondary education work for the
government.
The last two columns of table 9.3 compare the sectoral composition
of employment for men and women. Not included in this table is the
difference. in education between men and women: women have on
average received less education than men; fewer than 30 percent of
women have. been. educated beyond the primary school level, while
almost 50 percent of men were educated beyond this level. This result
-accords with the finding noted previously that girls are more likely
Table 9.3 Employment Sector by Level of Education and Sex
(percee.t)
Education Sex . . .
Mliddle Post-
Sector . Aone Primary school Secondary secondary Koranic Male Female Total
Government 12.6 *1.4 52.6 14.0 18.6 '0.9 719 28.1
1.7 1.0 11.7 33.7 53.3 95 . t05 3.9 7.0
State 19.6 *6.5 58,7. *4.4A. *8.7 '2.2 0.4 19.-6
0.6 1.0 2.8 2.3 5.3 4.8 2.5 0.6 15
Private 30.2 14.9 . 46.5 5.1 . *1.4 *1.9 76.7 23.3
4.1 10.4 10.4 12.4 4.0 19.1 11.3 3.1 7.0
� BaHousehold farming 65.8 9.9 .22.5 1.2 0.6 '0.1 . 47.8 52.2
65.7 51.3 37.3 21.4 13.3 95 52.5 . 52.2 523
Household business 48.7 9.5 37.2 2.2 1.1 1.9 26.2 73.8
14.0 14.3 17.9 11.2 6.7 42.9 - 84 21.4 15.2
Household business
and farming 48.7 14.8 33.9 1.4 *0.6 . 0.6 37 . 62.1
10.7 16.9 12.3 5.6 2.7 95 9.1 13.6 115
Unenp!ayed 32.7 9.5 435- 7.1 6.6 '0.6 50.0 50.0
3.4 5.2 7.6C 135 14.7 4.8 5.8 5.2 5.
Total . 52.4 10.1 31.5 2.9 2.5 2.9 . 47.7 523. 100.0
*Fewer than five observations. .N =3063
Nole: The first row in each sector gives the percentagc in that sector that corresponds to a particular educational level, while the
second shows the percentage with that level of education that work in the sector. For example, in cell 1, 12.6 percent of government.
employees has no formal education, while 1.7 percent of those with no formal education work for the government.
Source: Authors' calculations from the GLSS.
370 P. Beaudry and N. K. Sowa
than boys to become economically active before the age of 16. The
lower levels of educational achievement for women may help to
explain the much lower percentage of women in' the government, state
enterprise, and private nonhousehold sectors, By contrast, women are
much more likely than men to be self-employed in business activities.
Over 33 percent of women-compared to 17. percent of nien-work
in household business.
One of the surprising figures in table 9.3 is the high level of
unemployment. At 5.5 percent of the active population, this figure is
higher than official figures, and therefore requires some explanation.
Table 9.4 provides information on this. group. Within the grouip we
classified as unemployed, only 26 percent were actually searching for
a job. This is the group usually referred to as the unemployed.
Among the others classified as unemployed, 37 percent said they were
not searching for a job as they were either waiting to start a new job or
waiting to receive an answer from an employer, while the remaining
37 percent said they were not searching for a job because they did not
believe jobs were available. This last group, often referred to as
discouraged workers, represents about 2 percent of the total active
population.
Table 9.4 Unemployment Status
(percentage of unemployed)
Searching. Waitin g
Sex Location for job for job Discouraged
Male Urban 46.2 43.6 10.3
Rural 24.4 35.6 40.0
Female Urban 23.8 35.7 40.5.
Rural . 11.9 33.3 54.8
Total Urban 34.6 39.5 25.9
Rural 18.4 -34.5 -47.1
Total 26.2 36.9 36.9
Source: Authors' calculations from the GLSS. N=168
Gleana 371-
Notwithstanding (lie difficulties associated with measuring the
unemployment rate properly, many people consider the level -of
unemployment to be an inappropriate measure of-the underutilization
of labor. One better measure of underemployment may be the
fraction of the employed wanting more work. Table 9.5 indicates -that
only 4 percent of employed women. and 5 percent of employed men
were actually searching for more work during the previous week.
Correspondingly, almost 80 percent of employed people considered
that they had, enough work. Another 11 percent of the working
population indicated that they were not searching for additional work
since they did not believe that any was available. Together, the figures
on unemployment and underemployment do not indicate that finding
work is an extremely critical problem for most Ghanaians. :
'rable 9.6 'presents data on household labor income (the definition
of "Nwage" under table' 9.1.0 corresponds-to labor income). The male-
female distinction in this table refers to the sex of the- head of
household. The average household labor. income per month includes..
inc6me-received from both main and secondary jobs for all members
Table 9.5 Underemployment
(percentage of 'employed)
Sea rch for
additional' Enough
Sex Location wo-rk work Discouraged Others
'Male Urban 6.4 72.3 11.7 9.6
Rural 4.6 79.6. .9.7. ' 6.2
Female Urban 4.1. 72.1. 11.9 11.9
Rural 3.4 -78.0 1 1.8 6.9
Total Urban 5.3 72. 11.8 10.7
Rural 3.9 78710.8 6.5
Total. 4.3 77.1 7.6
Source: Authors' calculations from the GUSS.' N=2895
372 P. Iieaudry atd N. K. Sawa
of the household.4 Urban hiouseholds earn, on average, 50 percent
more labor income than rural households. Similarly, male-headed
households earn up to 50 percent more than female-headed
households in the same geographical region.
The fourth column of table 9.6 helps assess the relative importance
of labor income by. showing labor income as a percentage of total
income. Total income was calculated by summing labor income,
business income (profits), and net farming income (these figures
should be considered as only indicative, since double counting and
undeclared income is likely). Nevertheless, the data indicate that labor
income is an important source of income for all the categories shown
in table 9.6, with labor income being most important in urban areas,
especially Accra.
Table 9.6 Household Labor Income by Sex of Head of Household
and Region
Hoausehold Labor
labor income as Number Average
Sexrof Income . perceniage of earners income Number
-head of per month of total . . per per earner Household in
Region household (cedis) income household - cedis) size sample
Accra Male 14,287 88 1.4 10,205 4.5 99
Female 8,311 89 1.0 8,311 3.4 28
Urban Male 14,116 50 1.0 14,116 4.9. 249
Female 10,652 71 -. 0.8 13,315 4.3 116
Rural. Male 9,053 39 1.5 6,035 4.9 137
Coastal Female . 6,707 28 1.2 5,589 4.1 71
Rural Male 9,025 . 33 1.3 6,942 5.3 390
Forest Female 5,146 38 0.8 6,432 4.6 183
Rural Male 6,748 43 1.3 5,191 6.2 216
* Savanna Female 6,965 54 0.8 8,707 4.8 38
* Urban refers to households in towns with a population. over 5,000 other than Accra.
Source: Authors' calculations from the GLSS.
4. Household sector workers were also asked how much money they received for
their work. The answer to this question is considered to be labor income.
Glhana 373
One possible explanation for the male-female household income
differential observed in table 9.6 is that female-headed households
haveo fewer contributors to total labor income as indicated in the fifth
coltimn of table 9:6. The sixth column shows household income
divided by the average number of contributors. The sex-related
differential is almost eliminated, in this weighted data. Although
female-headed householdsr may be poorer, the results suggest that this
differential may be related mainly to a labor supply effect rather:than
to. limited olpportunities.
Table 9.7 examines the determinants of monthly labor earnings
more closely by decomposing individuals' earnings into hourly
wages, hours worked per week, and weeks worked per year. At. the
individual level,'women again have monthly labor earnings below
those of men (except women in urban areas other .,than Accra).
However, women also work less in remunerated employment than
men, both in terms of hours per week and weeks per year. Once labor
earnings are scaled to hourly earnings, -the sex-related wage
Table. 9.7 Personal Earnings from Main Job by Sex and Region
Monthly Weeks Houerly
earning Hours p&year earning
Rigion Sex (cedis) worked worked (cedcis)
Acra 12,459 50. . 41.9
Female 6,557. 40.3 37.3 88.
Urban Male 10,102 :40.9 42.2 90.1
Female 11,514 33.8 40.5 . 101.7.
Rural Male -4,119 32.2 39.0 .44.3
Coastal. Female 3,732 30.8 37.5 45.5
Rural 5,720 35.6 42.2 . .1
Forest Female 4,712 28.9 40.3 45.2
Rural Male L ,482 38.2 44.3 65.9
S Female 114,861 33.2 39.7: 44.7
Sourace A MaUb f the GS.2.4.7
Sourale: AuhrsMaleculation from th -LSS.36.9-
374 P. Beaudry and N. K Sowa
differentials are considerably reduced. In particular, women in Accra
earn almost the same per hour as men, although their monthly
earnings are on average nearly half those of men.
One interesting aspect of the average hourly earnings figures in
table 9.7 is the extent to which they are similar in Accra and other
urban areas and again for different rural areas. This suggests that
migration nmay play an effective role in equilibrating wages between
these different local labor markets (the urban rural wage differential
observed in this table is much-smaller than that usually observed; see
Squire 1981 for some comparisons). Consequently, this provides
some indirect preliminary support for the notion that relying on
migration to respond to the changes in relative wages caused by the
adjustment program may be reasonable.
Tables 9.8 and 9.9 provide further information about the
determination of wages and the labor supply. Table 9.8 indicates that
hourly earnings are ol average highest for people working in the
household business sector. This may be because many household
businesses are in urban areas, where the average hourly wage is much
higher than in rural areas. Employees of the government and of state
enterprises also tend to receive higher hourly pay than the-average
worker. Again, this may reflect a composition rather than a sector
effect. Table 9.9 shows that Workers with more formal education
Table 9.8 Personal Labor Earnings and Hours Worked on Main Job
by Sectors
Sector of Hourly Weekly
employment earnings (cedis) h1ours worked
Govemment 96.0- 40.2'
State .91.2 - 46.6
Private 52.2 45.4
Household farming 39.6 31.5
Household business 105.8 36.9
Household businiess and farmirng 68.6 29.8
Source: Autbors' calculations from the OLSS.
GClana 375
Table 9.9 Personal Labor Earnings and Hours Worked on Main Job
by Education
Level of Hoourly Weekly
edtication earnings (cedis) hours worked
None 56.9 - 34.5
Primary 53.1 33.0
Middle 68.1 36.7
Secondary .164.8 40.0
Postsecondary 130.6 . 33.3
Koranic 603 40.3
6. 40J
Source: Authors' calculations from the GLSS.
receive .higher hourly earnings. Since both the government and the.
state enterprise sector employ a large percentage of educated woikers,
this could be the cause of this sector effect. However, any reliable
. appraisal of these different explanations requires multivariate analysis,
which is undertaken in .the following section. Note that in none of the
tables 9.7, 9.8, or 9.9 does a strong link emerge between wages and.
hours worked. This suggests that labor supply elasticities are quite
small.
The preliminary information extracted from the GLSS provides
some insight into the functioning of. the labor market.. Although
Ghanaian workers are mainly concentrated in the household sectors,.
especially in farming, market forces are.apparent in the allocation and
remuneration of labor. For example, hourly earnings are quite similar
within rural areas and within urban areas,, most people seem to be able
to find a sufficient amount of work, and, on the basis of hourly
earnings,. women do not seem to be paid less than men.
The Determinants of Earnings and Labor Supply
This section examines the determinants of labor earnings and labor
supply in more detail. Table 9.10 examines the determinants of
hourly earnings for all individuals that reported labor income.
Following 'the human capital literature (see Mincer..1974. for an
Table 9.10 Hourly Wage for the Main Job
0 - - - ~~~~~~~~~ (t) - {~~~~~~2)- (3) :
Dependent. :HWM Standard HWVM Standard HM. Standard
variable es&;te error estimate error estimate err
Experience 0-029 0.007* 0.020 0.008* 0.023 0.007
Experience squared -0.0003 .0.0001 -0.0001 0.0001 -0.0003 0.0001
Primary 0.038 0.021 * 0.032 0,021 0.026 0.020
Middle .0.070 0.032* 0.056. 0.032 0.026 . 0.031
Secondary 0.176 0.044* 0.140 0,044* 0.102 0.042*
- Teaching techinical ed. 0.105 0.076 0.065 0.075 0.061 0.072
* Postsecondary 0.118 0.085 0.125 0.84: 0.058 0.081
Professional vocational ed. 0.252 0.150* 0.194 0.148 0.125 0.143
Accra 0.353 0.134 0.133 0.137 -0.154 0.136
Urban 0.794. 0.101* 0.577 0.105- 0.343 0.104*
Rural-coastal -0.010 0.111 -0.095 0.111 -0.156 0. 108
Rural-forest 0.197 0.094* 0.126 0.093 0.192 0.090*
Ghana national: 0.241 0.106* 0.256 0.104* 0.270 0.100.
Migrant -0.130 0.066* -0.120 0.065* -0.118 0.063*
Head of household 0.298 0.079* 0.283 0.078* 0.259 0.075
Women . ; 0.060 0.074 0.084 0.073 : -0.058 0.074
Seasonal - 0,S28 0.114 0.59 .114 0.387 0.111
Tenure -0.014 0.003 0.010 0.004
Tenure government 0.049 * 0.007 -0.005 0.010
Tenure private business 0.026 0.009* -0.002 0.0 11
Tenure household business 0.0 14 - 0.004o -0.020 - 0.005'
Union. 0.589 0.215
Government 1.101 0.202*
Private business 0.484 0. 162*
Household business 0.904 0.126
Pension, 0.066 0.152
Mining 0.598 0.223
Manufacturing 0.317 - 0.127
Service 0.284; 0.109*
Construction'. 0.249 0.171
No. obs. 1,801 : 1,801 1,801
R 0.157 0.185 0.255
HWM = log of hourly wage on main job
Significantly different from zero at 90 percent confidence level
Ghana national = I if the individual has Ghanaian citizenship, otherwise 0
Migrant = I if the individual has moved within the last five years, otherwise 0
Head of household = I if the individual is the head of the household, otherwise 0
Women = I if the individual is a female, 0 if a male
Seasonal = I if the individual has worked less than 20 weeks in the last 12 months, otherwise 0
Union= I if the individual has worked in a unionized firm, otherwise 0
Pension = I if the individual has a pension plan, otherwise 0
Tenure govemment = tenure in the government
Tenure private business = tenure in the formal private sector
Tenure household business tenure in the informal business sector
378 P. Beaudry anidN. K Sowa
introduction), column (1) of table 9.10 presents the estimated
coefficients for a regression of hourly earnings on a set of individual
characteristics. Most of the results are standard. The first two
coefficients represent the effect on wag-.s of total labor market
experience: this effect is positive and decreasing. The next five
coefficients represent the effect on wages of a year of education at
different levels of schooling. All these coefficients indicate a positive
return to education. The highest returns to education are associated
with formal secondary education. Postsecon-dary education, including
teaching and technical education, show large positive returns, but the
coefficients are not significantly different from zero at the 90 percent
confidence level. The variable denoted prof-voc is also an education
variable. This dummy variable is equal, to 1 if the individual has
completed a professional or vocational, training program. The returns
to this type of training are also positive.
The next set of variables is included to capture differences in wages
that arise due to geographical considerations. Such differences can
result from either different costs of living or from different local labor
demand conditions. The latter source of differelntials can be
maintained in the long run only if migration flows are unresponsive.
The four geographical dummy variables included in the regression are
mutually exclusive. The residual category is rural savanna. The largest
location differential is associated with living in urban areas other than
Accra (towns over 5,006 in population). The results cf this regression
also indicate that living in Accra is associated with a premium when
compared to living in rural areas (however, this result is not robust).
Among rural areas, the differences are not very large'. Working in the
forest region,-which is the cocoa produdtion region, is nevertheless
associated with a small premium compared'to the other two rural
regions. Although the rural savanna region is very poor, these results
indicate that the rural coastal region may be even poorer.
A final set of individual-specific dummy variables capture earning
differentials due to information networks-or discrimiination. Ghanaian
nationals and heads of household are shown to receive higaer than
average hourly earnings, while recent regional immigrants (those who
have lived in their current place of residence for less than two years)
receive lower than average earnings. Further confirming our previous
Gihana 379
observations, womeen are not observed to receive lower hourly earnings
than men. Finally, seasonal workers (individuals working less than 20
weeks a year) receive higher than average hourly wages. This last
result suggests that seasonal workers enter the market only in good.
times and leave in bad times, indicating a relatively frictionless labor
market.
To examine the:link between employment-specific capital and
earnings, column (2) of:table 9.10 includes a set of tenure variables.
The tenure variable is the number of years that individuals have
worked in their current job. This variable is also allowed to interact
with sector dummies: the public sector, the private nonhousehold
sector, and the household business sector, with the household farming
sector . omitted. In most sectors, the -hourly wage has a positive
correlation with tenure, but contrary to standard expectations
concerning tenure variables, the estimates do not indicate a positive
tenure profile for household farm sector workers. Most learning in the,
household farming sector jobs. is therefore possibly of the general
human capital type, and consequently may make labor mobility in this
sector relatively inexpensive.
Column 3 of table 9.10 adds a set of employer-specific variables to
our individual-specific set of explanatory variables. Assuming thie
labor- market was without friction and that employees had no
aggregate preferences with respect to types of employment, then
employer-specific-variables should not be expected to have any
explanatory power. However, many of our employer-specific variables
are observed to have large and significant effdcts on. hourly earnings.
These effects may represent either compensating wage differentials or
'labor market distortions. The large and significant effect of unions on
wages is a common example of an interference with market forces,
however, this interpretation is sometimes debated (see: Lewis 1986).:
The increase in wages associated with unionization is estimated to be
close to 60 percent.. This estimate is much larger thmin similar estimates
for industrialized countries, which are usually .of the order of 10 to 20
percent (Lewis 1986). Comparing wage difftrentials associated with
sectors of employment, we find that all sectors of employment receive
a premium relative to the household;farming sector. The premium is
highest among public sector employ" ;'s, followed by household
380 P. Beaudry and N, K. Sow.
businiess employees, and finally by private sector employees. However,
a fraction of these premiums are likely to be a compensation for the
loss in direct consumption associated with household farming.
Similar to the sector effects, the industry dummies also indicate -a
negative wage differential for agricultural employees (the agricultural
industry is the residual categor). The employees nf the mining
industry are observed to receive: the highest industry-related
differential. This last observation accords quite well with- a
compensating differential interpretation of indust-ry effects. However,
the other industries-manufacturing, services, and constTuction-do
not exhibit major interindustry differentials. The final employer-
specific variable iniclud&d in this regression is a pension dummy. This
vanable indicates whether the employee is covered by a pension plan,
and it is included to captutre the: notion of a formal employment
relationship. This variable is not observed to have a significant effect
onl wages, which may be due to its correlation with other variables.
The inclusion of employer-specific variables in our earnings
equation has changed some of the estimated codfficients of the
individual-specific variables. For example, the Accra dummy -is no
longer significant once employer variables are taken, into account.
This suggests that the previous premium we associated with living in
Accra was probably caused by the greater likelihood of being
employed in high'paying sectors rather than an across-the-board
compensation. Moreover, this result indicates that migration may -be
eliminating geographical earnings differentials. Another consequence
of the introduction of the employer variables has been the reduction
in the returns to schooling. This may be caused by either differential
returns to education in different jobs, or by education being used as a
selection 'device in certain jobs. SThe examination of sector-specific
earnings functions will help assess the plausibility' of each of these
explanations.'
In brief, these first regression results indicate that bourly labor
earnings conform relatively well to a combination of human capital
theory, the theory of equilibrating migration flows, and to tihe theory
of compensating differentials, that is, returns to human capital are
positive, . regional differences are not extremely important,. and
Ghana 381
compensating differentials are related mostly to being off the farm,
especially in the mining industry.
Labor Supply
In order to further depict the overall functioning of the .hanaian
labor market, an examination of. labor. supply decisions is useful.
Table 9.11 presents estimates of the. determination of weekly hours
worked on the. main job. The examination of hours worked in both
main and secondary jobs will be discussed later.
Column (1) of table 9.11 presents estimates of a labor' supply
function .using ordinary least squares (OLS).1 The -estimated elasticity
*-drf labor-supply with respect to hourly earnings is negative. However,
this estimate may be severely biased, since unobserved individual
heterogeneity is likely to be correlated with hourly earnings. Column
(2) therefore presents results from a two-stage least squares (2SLS)
estimation of the same labor supply function.5, These results indicate a
positive and 'significant effect of hourly earnings on the, supply of
labor,, although the effect is quite small. The other results of column
(2) worth noting are that women work significantly fewer hours.per
week than men, and that workers in .Accra work on average'more
hours a week. Column (3) presents estimates for a reduced-form labor
supply function, that is, the hourly earnings variable is replaced by the
earnings function. These results suggest that our positive estimate of
the elasticity. of labor supply is most likely the result of the lower wage
farmers receive and the lower hours worked by farmers (this
aggregation bias will be discussed further).
Main and Secondary Jobs
Both tables 9.10 and 9.11 have examined the determination of
hourly earnings and hours worked for main jobs'only, however,
almost 30 percent of the people in our sample reported earnings from
a secondary job. Therefore, column (1) of table 9.12 presents
estimates of the determiniation of total monthly earnings from-both
main and secondary:Jobs (using the extended set of variables), and
5. The set of instruments used for this bstimation are the employer-specific
variables and the tenure variables.
Table 9.11 Hours Worked per Week on Main Job
! ' (1] ) , (2) (3)
Dependent -FRS Standard H HRS (2SLS) Standard - HRS Standard
variable estimate error - estimate eror estimate error
HWM -0.150 0.01 1.1 94 0.039*
Experience 0.007 0.003* -0.003 0.004 0.001 0.004
Experience squared -0. 0001: 0.00004* -0.00001 0.0001 -0.0001 0.00005
Pnlmary -0.008 0.009 -0.021 0.012$ -0.016 0.010
Middle 0.025 0.014* 0.001 0.018 0,002 0.015
Secondary 0.032 0.020 -0.029 0.026- -0.019 0.020
co Teaching technical ed.: 0.005 0.034 -0.031 0.043 -0.027 0.035
Postsecondary. 0.030: 0.038 -0.011O 0.048 -0.001 0.039
Professional vocational ed. 0.068. 0.067 .-0.018 0.085 -0.018 0.070
Accra. 0.296 0.060* 0.175 0.077' 0.046 0.066
Urban . 0.142 0.046* . 0.131 0.065* -0.09- 0.050*
Rural-coastal -0.1 69 0.050* -0.165 0.063 -0.165 0.053*
Rural-forest .. -0.071 0.04* .-0.39 0.053 -0.090 0.044
Ghana nat. 0.203 0.o47*- 0.120 0.060 0.188 00.049
Migrant .-0.028 0.030 0.016 0.038 0.004 0.030
Head of household 0.136 0.035* 0.033: 0.046 0.070 0.037*
Women* -0.169 0.033* --:0.1900' 0.042 : -0.174 0.036*
0 - Seasonal -0.142 0.051* -0.32 0.068* -0.227 0.054
S6'asonal:~~~~~~~~~~~~~~ 0.68 0.054 . ..
Tenure 0.002 0.002
Tenure govemrnment -0.00003 0-005
Tenure private business 0.005 0.005
Tenure household business 0.003 0.003
Union -0.073 0.105
Government -0.001. 0.098
Private business -0.123 0.079
Household business: -0-098 0.061
Pension 0 171 0.074
-Mining 0.080 0.108
�� Manufacturng 0.165 0.062
Service - 0.290 0.053
Construction 0.330 0.083
No. obs. 1,801 -1,801 1,801.
R2 0.179 0.074 0.138
HRS = log of hours per week worked on main job
HWM = log of hourly wage on main job
Significantly different from zero at 90 percent confidence level
See table 9.10 for definitions of other dependent variables.
384 P. Beaudry and N. K. Sowa'
Table 9.12 Monthly Wages and Hours Worked on Main and
Secondary Jobs
(I) -{(2)
Dependent WMS Standard HIRSSM Standard
varlable estfiate error estinatea error
Experience 0.028 0 .0007 0o002 0.003
Experience squared -0.0004 0.0001 .0.0001 0.0001
Primary 0.027 0.018 -0.013 0.009
Middle - 0.OI1 0.027 -0.0001 0.01A.
Secondary: 0,056 0.038. 0.019 0.020
Teaching technical ed. 0.036 0.063 -0.030 0.033
Postsecondary . 0.075 0.072 -0.001 0.038
Professional vocalional ed. 0.091 0.127 -0.018 . 0.067
Accra 0.017 0.122 0.004 0.064
Urban 0.246. 0,093* -0.129 0.049
Rural-coastal -0.160 0.098 -0.164 -0.052
PRural-forest . . 0.179 0.081 -0.045 0.043
Ghana national 0.530 0.089 0.171 0.047
Migrant -0.117 0.056' -0.016 0.029
Head of household' 0.303 0.067 0.076 . 0.036k
Women -0.210 0.066 -0.178 0.035
Seasonal -0.203 0.099* -0.228 0.052
Tenure 0.013 0.003* 0.002 0 5O"i
Tenure government -0.007 0.009 --0.0003 0.005
-Tenure private business .-0.007 0.010 - 0.002 0.005:
Tenure household business -0.012 0.005 0.001 .0.002
Union .0.362 0.192. -0.116 0.102
Government . .1.313 0.188* . 0.195 0.099-
Private business 1.121 . 0.152 . 0.330 0.080
Household business 1.370 0.123 0.121 0.065
Pension 0.094 0.136 0.123 0.072
Mining 0.230 0.202 -0.006 . 0.107
Manufacturing -0.040 0.119 0.061 . -0.063
Service 0.128 0.104 0.158 0.055'
Construction' 0.145 0.158 0.217 0.083
Second job 0.505 0.087 0.322 0.046*
No. obs. 1,807 1,806
R2 - :- 0.379 0.180
WMS _ log of monthly wage on main and secondary job
HRSSM = log'of total hours on main and secondary job
. Significantly different from zero at 90 percent confidence. level
See tabl 9.10 for definitions of other:dependent svariables.
Glhana 385
column (2) presents results for total hours worked per week on both
jobs (using the reduced form estimates). The sample remains that of
all workers, irrespective of whether or not they had a second job.
The regression results for total monthly earnings closely resemble
those for hourly earnings; however, a few differences are apparent.
First, women receive considerably lower monthly earnings than men,
even though their hourly wages were not observed to be below
average. Second, the earnings differential associated with unionization
on the main jobs is considerably reduced when total monthly earnings
are considered instead of hourly wages. Third, workers in all sectors
other than the household farming sector receive similar monthly
earnings, in particular, public sector employees are no longer observed
to receive a premium. Finally, individuals with a second job receive on
average 50 percent more monthly earnings than workers with only
one job.
The size of the income effect associated with a secondary job
warrants-closer examination. Economic theory predicts thatlin a well-
functioning labor market, the income effect of a second job should be
the result exclusively of the greater number of hours worked. From
column (2). in table 9.12 we observe that most of the income effect
associated with a second job is possibly the result of an hours effect,
that is, the size of the hours effect and the size of the income effect are
not significantly different. Consequently, hourly earnings are possibly
similar on both jobs. This observation, combined with a previous
observation about individuals' general lack of desire for more work
(table 9.5), supports a competitive interpretation of labor allocation.
Sector-Specific Earnings Functions
To examine whether different segments of the labor market behave
differently, tables 9.13-9.15 present estimates of earnings functions
and labor supply functions for specific subsamples. Table 9.13
presents results for the- household farming sector, while tables 9.14
and 9.15 present results for the formal and informal segments of the
labor market, excluding the household farining sector. In table 9.14, a
job has been defined as formal if either (a) it is a public sector job, (b)
it is a unionized job, (c) it is covered by a pension plan, or (d) it is
covered by social security benefits. For all three tables, column (1)
386 P. Beaudry and N, K Sowa
Table 9.13 Household Farming Sector
(1) (2) (3) (4)
Dependent HWM HRS WMS HRSSM
variable estimate estimate cstimate estimate
.-IWM -0.083
Experience 0.034 - 0.006 0.043 0.003
Experience squared -0.0003 -0.0001. -0.0005 -0.0001
Primary 0.081 -0.037 0.036 - -0.043
Middle 0.010 0.022 0.040 0.018
Secondary . 0.017 0.043 0.012 0.042
Teaching technical ed. 0.086 -0.056 0.044 -0.070
Postsecondary -1.635 -0.592 . -1.667 -0.470
Professional vocational ed. 0.317 0.122 0.333 0.148
Accra -0.933 -0.119 -0.713 -0.034
Urban 0.203 -0.300* -0.055 -0.311*
Rural-coastal -0.173 -0.142 -0.219 -0.142
Rural-forest 0.275 -0.057 0.298 -0,073
Ghana national 0.343* 0.109 . 0.488 0.082
Migrant - -0.188 -0.086 -0.221 * -0.067
Head of household 0.275 . 0.072 0.308* 0.048
Women -0.046 -0.246& -0.297- -0.249
Seasonal 1.594 -0.235 0.276 -0.365
Tenure 0.005 0.008 . 0.001
Tenure government
Tenure priivate business
Tenure household business
Union
Government
Private business
Household business
Pension
Mining
Manufacturing
Service
Construction
Second job 1.464 0.509
No. obs. 734 734 735 735
R2 - 0.156 0.117 0.220 0.139
HWM = log of hourly wage on main job
HRS = log of hours worked on main job.
WMS = log of monthly wage on main and secondary job
HRSSM = log of total hours on main and secondary job
Significantly different from zero at 90 percent confidence level
See table 9.10 for definitions of other dependent variables.
G/tana 387
Table 9.14 Formal Sector
(1) (2) (3) (4)
Dependent HWM HRS WMS HRSSM
variable estimate estimnate estimate estimate
HWM -.-0031
Experience 0.020 0.001 0.013 0.0002
Experience squared -0;0004' 0.00003 -0.0003 -0.000004
Primary -0.106l 0.029 -0.033 0.032
Middle 0.155 -0.025 0.055 -0.054
Secondary -0.205 - -0.047 0.094 -0.076
Teaching technical ed. 0.075 -0.037 0.034 -0.025
Postsccondary 0.047 0.007 0.062 0.011
Professioal vocational ed. 0.028 -0.013 0.028 0.003
Accra 0.098 0.139 0.223 0.112
Urban 0.215 0.020 0.137 0.004
Rural-coastal 0.263 -0.241 0.035 -0.203
Rural-forest 0.215 -0,032 0.197* 0.055
Ghana national 0.015 0.044 0.081 0.082
Migrant 0.015 -0.004 0.023 -0.017
Head of household 0.181 0.052 0.098 0.071
Women 0.036 -0.107 -0.077 -0.108w
Seasonal -0.638 0.08.* -0.480* 0.018
Tenure ..012 0.014 0.000
Tenure government
Tenure private business
Tenure household business
Union 0.199 0.205 -0.082
Government 0.352 0.190 -0.032
Private bus.
Househuld business
Pension
Mining 0.029 -0.104 -0.071
Manufacturing 0.075 0.078 0.147
Service -0.145 -0.074 -0.005
Construction -0.181 -0.053 0.136
Second job - 0.288 0.184
No. obs. 311 311 313 313
R2. 0.276 0.205 0.313 0.251
HWM =.0og of hourly wage on main job
HRS = log of hours worked on main job
WMS = log of monthly wage on main and secondary job
HRSSM - log of total hours on main and secondary job
Significantly different from zero at 90 percent confidence level
See table 9.10 for definitions of other dependent variables.
388 P. Beaudry and N. K. Sowcu
Table 9.15 Nonagricultural Informal Sector
(I) (2) (3) (4)
Depenidernt HWM IIRS WMS HRSSM
variable estimalfe estimate estiitte estimaie
HWM 0.015
Experience 0.025 -0.006 0.026 0.002
Experience squared -0.0004 0.00008 -0.0004' -0,00004
Primary -0.015 -0.002 0.018 0.009
Middle 0.013 0.005. -0.007 0.004
Secondary 0.132 -0.033- 0.063 -0.002
Teaching technical ed. -0.266 0.143 - 0.107 -0.067
Postsecondary 0.009 0.047 0.065 0.031
Professional vocational ed. 0.265 -0.139 0.114 -0.125
Accra -0.610 0.230 -0.404 0.081
Urban 0.046 0.019 -0.004 -0.052
Rural-coastal -0.589 -0.161 -0.534 -0.091
Rural-forest -0.314 -0.093 -0.354 0.059
Ghana national 0.316 0.294 0.758 0.286
Migrant -0.086 0.045 -0.159 0.029
Head of household' 0.224 0.143 0.372* 0.137
Women -0.163 -0.111 -0.120 -0.075
Seasonal 0.246 - -0.231' -0.233 -0.236
Tenure -0.006 0.004 0.004
Tenure govemment
Tenure private business 0.018 0.010 0.001
Tenure household business
Union
Government
Private business -0.491 -0.275 0.253
Household business
Pension
Mining 1.018 0.396 -0.014
Manufacturing 0.467 -0.180 0.009
Service 0.399 -0.045 0.161
Construction o.516 0.163 0.141
Second job 0.356 0.308
No. obs. 754 754 756 .756
R2 0.120 0.070 0.159: 0.142
HWM = log of hourly wage on main job
HRS = log of hours worked on main job'
WMS = log of monthly wage on main and secondary job
HRSSM = log of total hours on main and secondary job
Significantly different from zero at 90 percent confidence level
See table 9.10 for definitions of other dependent variables.
Glhana 389
presents regression results for our extended hourly earnings function,
column (2) presents results for the two-stage estimation of labor
supply, column (3) presents results for the total monthly earnings
equation, and column (4) is the reduced form regression results, for
total weekly hours.6
The results of tables 9.13-9.15 indicate quite different returns to
education in the different segments of the labor market. Earnings in
informal jobs (excluding household farming) do not indicate any
significant returns to education, and earnings on household farms are
positively related only to primary education. By contrast, earnings in
formal employment relationships are much more closely related to
education, especially to middle aind secondary school education.
Another interesting result is the effect of living in Accra. Table 9.14
indicates that formal sector employment does not provide a premium
for individuals living in Accra, however, an informal sector worker
earns less in Accrn than elsewhere. This result may be due to a Harris-
Todaro type of equilibrating mechanism (Harris and Todaro 1970),
whereby migration decisions are based on the comparison of a
probability weighted sum of earnings in the formal and informal
sectors. Since the probability of formal employment is greater in
Accra, equilibrating migration flows cause informal sector jobs in
Accra to receive less than average earnings. This suggests that
informal sector workers in Accra may be among the most
disadvantaged workers in urban areas.
The comparative results between sectors also indicate the possibility
of different forms of organization or production technologies in these
different segments of the labor market. In formal jobs, seasonal
workers are paid less and the tenure profile is positive. In informal and
in farm jobs there is no tenure profile, and seasonal workers receive
higher average hourlv earnings. Thus, the presence of job-specific
investments appears to be relevant only in the formal sector,
consequently, mobility is probably more costly in -this sector.
6. The regression results of these tables could be biased because of sample
selectivity problems; however, preliminary work to correct this type of bias has not
produced very different results. Results by Stelcner and others (1987) for Peru,
nevertheless, suggest that this bias may be important for assessing public-private
wage differentials.
390 P. Beaudry and N, K Sowa
The comparison of the role secondary jobs play in the different
segments of the labor market is also quite striking. In both tables 9.14
and 9.15, the effect on monthly earnings of a secondary job is almost
totally attributable to an increase in hours worked. This again suggests
a labor market without strong frictions, which is not surprising for the
informal sector. However, the effect on earnings of having a
secondary job in the household farming sector cannot be attributable
only to.an increase in hours worked. Average hourly earnings for a
farmer's second job are therefore probably much higher than on the
first job. This resulit may indicate either the presence of a high fixed
cost associated with finding a second job, or simply that the earnings
on primary farming jobs are grossly underestimated because of direct
consumption. If the latter hypothesis is maintained, secondary jobs
provide a good estimate of farmers' returns to work and, in this case,
indicate that household farm workers may not be at a great
disadvantage relative to workers in other sectors of the econoimy.
A final observation from the sector-specific estimations is that the
labor supply elasticities are all insignificantly different from zero. This
suggests, as noted earlier, that the estimate of this elasticity derived
from table 9.11 may be biased upward due to the pooling of the
different sectors of employment.
In view of the results presented in this and the previous section, the
Ghanaian labor market offers some signs of a relatively fluid labor
market, for example, the seasonal earning differential and the absence
of important earnings differentials for.secondary jobs are signs of a
well-functioning market. However, the differential patterns of earnings
between sectors indicate the possibility of segmentation within the
labor market, that is, the rationing of formal sector jobs.
The Role of the Labor Market in the Adjustment Process
We now will try to assess the link between the labor market and the
process of adjustment under the ERP. In particular, we will examine
how the labor market responded to the adjustment program. Overall,
the macroeconomic indicators presented earlier indicate that the
stabilization aspects of the ERP have been mostly successful. However,
to examine whether any fundamental structural adjustment has really
occurred, we must examine the program's impact on the sectoral
G/lana 391
composition of output. A primary goal of the adjustment program has
been to liberalize the price system in the hope of stimulating
production in sectors where Ghana possesses a comparative advantage.
Table 9.16 presents data on sectoral shares of production between
1978 and 1987 (these shares are evaluated at purchasers' value in
current dollars). Since 1983, the sectors of economic activity that have
grown fastest are the industrial sector, the government sector, and the
cocoa production sector. These changes in the sectoral composition:of
output accord quite well with the objectives of the ERP. For example,
the change in the terms of trade in favor of cocoa. production,
resulting in part from the depreciation of the cedi, pushed cocoa.
production from 4 percent of GDP to almost 9 percent of GDP
between 1984 and 1987. The improvements in the world price of
cocoa between 1982 and 1984 were initially an important factor in
this increase, although since 1984 the world price of cocoa has been
decreasing (see table 9.1).7 The large increase in Ghana's main export
has contributed greatly to the trade balance surpluses registered in
1986 and 1988. Similarly, the rapid growth of the industrial sector,
especially manufacturing, reflects the adjustment program's goal of
increasing the capacity utilization rates. These rates are nevertheless
still quite low.8
Part of the recovery in the industrial sector may be due to the
rationalization of the foreign exchange system. Since 1983, the
foreign exchange system has gradually changed from a fixed
exchange system with controls to an auction system. Within the
auction system, firms that value foreign exchange the most are served
first (under the constraint of having the required capital). Therefore,
the allocation of imported intermediate goods has probably improved
in favor of the most productive firms within the industrial sector,
thereby encouraging gruwth. Small firms are,. nevertheless
disadvantaged by the new system. The 100 percent up-front payment
requirement for foreign exchange is often prohibitive for small firms
7. These figures probably overestimate the real increases in the production of
cocoa. Part of the increase is only a redirection of production from smuggling toward
official channels.
8. The size of the manufacturing sector in relation to total production was still
lower in 1987 than it was in 1970.
392 P. Beaudry andN K. Sowa
Table 9.16 Percentage Distribution of Gross Domestic Product by
Kind of Economic Activity, in Producers' Values, 1978-81, 1984-87
(current cedis millions)
Sector 1978 1979 1980 1981 1984 1985 1986 i9874
Agriculture 47.7 45.6 48.4 43.8 40.8 34.6 34.3 35.7
Cocoa production 7.0 8.1 5.5 1.4 4.1 5.5 8.0 8.9
Forestry 4.8 5.0 4.8 3.7 3.4 3.3 3.7 4.6
Mining 0.6 0.8 1.4 0.5 1.2 1.1 1.7 1.8
Manufacturing 8.6 8.7 7.2 5.9 6.4 11.5 11.2 9.9
Electricity 0.4 0.4 0.6 0.5 0.8 1.2 1.8 1.8
Construction 2.5 2.3 2.3 1.9 2.2 2.9 2.5 2.4
Transportation 2.6 2.7 2.2 2.1 6.4 5.3 -4.3 3.7
Trade 13.3 13.8 14.6 30.0 28.3 24.8 19.1 18.5
Finance 3.1 3.6 2.8 2.0 1.7 2.3 3.0 2.6
Government 8.2 7.5 8.5 6.2 3.9 6.2 8.2 8.1
* Provisional
Sources: 1978-81: World Bank (1984a); 1984-87: Ghana Statistical Service
Quarterly Digest Statistics (1988).
(firms are not allowed to borrow from the banking system to buy
foreign exchange), which may have greatly limited their expansion.
The growth in. the government sector observed in table 9.16 may
seem paradoxical given that one of the objectives of the ERP was to
cut this sector. The main cause of this observed growth is the increase
in public sector pay. Increases were given mainly in an attempt to
improve efficiency. The level of wages in the public sector as of 1983
was generally so low that most workers needed other jobs to
supplement their income, which often reduced the hours they worked
for the government. The results from the estimated earnings functions
presented earlier indicate that these increases have probably now
brought. government sector pay in line with private sector pay.
Nevertheless, comparison of the pay structure in Ghana with that in
neighboring countries may suggest the need for further increases for
educated workers.
Ghana 393
Table 9.16 indicates that the major declining sector since5 1983 is
commerce. This observation must be carefully qualified. Evaluated in
constant 1975 dollars, the share of commerce in total production has
actually been r'ising under the ERP. This seemingly contradictory
observation is due to a sharp decrease in the relative price of services
during the period. The actual quantities of services, including
commerce, have been rising quickly rather than declining. Following a
period of contraction, such an expansion in the- volume of commerce
is normally expected. However, a fall in the relative price of trading
services is somewhat surprising. This change in relative price is most
likely a direct consequence of the ERP's objective of stimulating
exports and, as will be shown, it has had a large impact on the
remuneration of service sector workers.
The labor market's role in redirectin,g production observed under
the ERP will first be analyzed by examining the wage trends associated
with the different sectors of the economy. Table 9.17 presents time
series data on real monthly earnings by sector of production. These
data come from establishment surveys, and must therefore be
considered as indicative as they cover only a small fraction of the
economy.
The observed trends in wages have followed the pattern of output
quite closely. Real wages were falling- rapidly before the ERP and
started improving immediately after. However, both the downward and
upward trends have been much more accentuated for wages than for
outputs. Wages rose some 50 percent between 1983 and 1987, while
GNP growth per capita rose by just over 10 percent during the sarme
period.
The sectors where real wages have gained mrst during the
adjustment program are agriculture, mining, and transportation: all
sectors favored by the ERP as a means of improving external balance.
Real wages in these sectors have reached levels close to or above those
prevalent in 1978. In contrast, real wages in the commerce sector were
still 25 percent lower in 1987 than in 1978. One possible explanation
of the poor wage performance in the commerce sector is that this
sector probably served as the main absorbing sector during the crises
of the early 1980s. The relative importance of the commerce sector in
the total value of production (current prices) went from 13 percent of
Table 9.17 Average Monthly Earnings by Sector, 1978-86
(constant cedis, 1978 - 100)
Sector 1978 1979 1980 1981 1982 1983 1984 1985 1986
Ag-.culture, forestry,
f.sheries 100.0 70.3 82.9 58.6 53.2 38.0 70.3 76.3 126.9
Mining, quarrying 100.0 79.8 87.3 56.5 46.8 53.4 58.3 1641 153.0
Manufacturing 100.0 75.7 78.2 45.1 42.1 31.6 59.2 7?.9 100.1
ElectriciLy, water, gas 100.0 85.0 102.2 63.1 54.6 28.4 46.7 70.4 112.2
Construction 100.0 76.2 83.3 54.8 46.0 35.2 44.7 67.0 102.6
Commerce 100.0 85.1 81.3 36.5 44.2 28.6 34.1 56.1 785
Transportation, storage,
communications I00.0 84.0 93.8 50.6 47.8 33.4 49.5 70.1 163.2
Finance, insurance,
real cstate 100.0 65.3 67.1 39.9 39.6 25.5 38.4 61.6 103.8
Services 100.0 78.9 86.1 49.1 42.2 30.8 47.2 60.1 113.0
All industries:
Public sector 100.0 81.3 88.8 53.9 46.2 36.7 51.9 76.7 130.4
Private sector 100.0 82.2 86.2 46.0 48.6 32.8 59.1 76.7 98.9
All sectors 100.0 82.2 88.3 52.5 46.6 36.2 53.2 76.5 124.5
GNP/Capita 100.0 95.0 92.0 87.0 78.0 72.0 78.0 79.0 81.0
Sources: 1978-81: World Bank (1984); 1982-86: Ghana Statistical Service Quarterly Digest Sradstics (1988).
Ghania 395
production in 1978 to 29 percent in 1983. This influx of labor into
the commerce sector placed enormous pressure on earnings within the
sector, which has not as yet been completely reversed. Another factor
lhat contributed to the decline of carnings in the commerce sector was
lhe reduction in botlh corruption and the excessive usc of market
power by lraders (kalabuile). These chalnges were a direct consequence
of Flighl. Licutenlant J. J. Rawlings' ascent to power.
Thle real wage flexibility observed in table 9.17 for all sectors of
production offers further evidence that the Ghanaian labor market
responds to competitive forces, and therefore, favors the hypothesis
that the functioning of the labor market has probably-facilitated the
sectoral reallocations sought by the ERP.
For .a. more detailed assessment of how the returns to labor may
have changed since 1982 and thereby created the incentives for
workers to relocate, table 9.18 presents information on the changes in
earning patterns over the course of adjustment. Column (1) of table
9.18 presents results for the estimated hourly earnings function for
workers who began their jobs between 1983 and 1988. Column (2) is'
the estimated hourly earnings function for workers who began their
jobs before 1983. Notice that both equations are estimated using data
for the same year (the data come from the GLSS for both
regressions). Consequently, the proper interpretation of the difference
between these two columns is not obvious. Under the assumption that
market forces are much more effective at the margin of new
employment, we can interpret the differences between these two
columns as indicating changes in patterns of earnings (this is
obviously a questionable interpretation).
The results shown in table 9.18 indicate that regional differentials
have changed during the adjustment process. While the data indicate
that premiums are associated with jobs acquired before 1983 in urban
areas other than Accra and in the forest region; these premiums-are
not apparent for jobs acquired after 1983. The changes in the patterns
of regional earnings may be.the result of competitive pressures due to
migration flows. A second sign of changes in market forces since
1983 is the higher premium associated with being in a union for
recently hired workers. The union differential is about 30 percent for
workers hired before 1983, but over 60 percent for workers hired after
396 P, Blcaudry and N, K. Sowa
Table 9.18 Changes in Pattern of Earnings
- U) -(2)
11MW 11 MW
D apoLd(enI fitetre <5 Staindard l aiiurc > 5 Standard
varilable exst iane error esiV1timate crtr
Experience 0,027* 0.012 0.011 0.011
Experience squarcd 0*0003 0.0002 -0.0001 - 0.0001
Prlmnlry 0.060 0.032 0.014 0.026
Middle 0.017 0.047 0.010 0.042
Secondary 0.11000* 0.061 0.085 0.063
Teaching technical cd. -0.071 0.211 0.127 0.097
Postsecondary 0.075 0.096 0.097 0.156
Professional vocational ed. 0.252 0.222 0.166 0.191
Accra -0.493 0.207 0.132 0.193
Urban 0.065 0.169 0.495 0.136
Rural-coastal -0.268 0.189 -0.147 0.133
Rural-forest -0.083 0.166 0.326 0.107
Ghana national 0,187 0.150 0.408 0.135
Migrant -0.176 0.100 -0.042 0.082
Head of household 0.411 0.119 0.101 0.099
Women 0.272- 0.121 -0.268* 0.095
Seasonal 0.250 0.134 1.188 0.231
Tenure 0.013 0.063 0.006 0.005
Tenure government 0.070 0.103 -0.006 0.014
Tenure private business -0.057 0.097 -0.019 0.015
Tenure household business 0.046 0.077 -0.016 0.007
Union 0.659 0.313 0.297 0.317
Government 0.994 0.404 1.080 0.292
Private business 0.735 0.306 0.903 0.330
Household business. 0.963 0.295 0.748 0.178
Pension 0.160 0.237 0.027 0.199
Mining 0.665 0.301 0.434 0.367
Manufacturing 0.178 0.207 0.322 0.169
Service 0.148 0.182 0.350 0.146
Construction 0.427 0.275 0.043 0.227
Second job
No. obs. 710 1,090
R2 0.281 0.278
HMW= log of hourly wage on main job
Significantly different from zero at 90 percent confidence level
See table 9.10 for definitions of other dependent variables.
Ghatna 397
1983. This suggests thal wages for recently hired workers In the
unionized sector have adjusted much less rapidly Iltan In the
nonunionized sector.
The changes in industry-related earniing diffierentials observed in
table 9.18 lend some support to our interpretatioll of thc difference
between the columns as indicating changes in the patterns of earniings.
As in table 9.17, in table 9.18 the mining differential has increased
and -the service differential has decreased (botlh differentials are
compared to agriculture). The table also suggests that the returns to
education linve inicreased over the course of adjustment (this result is
not. statistically very significant). In particular, the return associated
with a year of primary education has tripled- for recently acquired
jobs. This, increase in the returns to education may reflect in part the
government's-policy of stretching the wage scale within the public
sector as a means of keeping ils most educated workers. In 1983, the.
ratio of relative wages within the public sector was at most 1:1.8. This
ratio was much smaller than the equivalent ratio for government
employees in neighboring countries, and therefore probably
contributed to Ghana's loss of many educated workers. Another
explanation for the increased returns to education is the reduction in
corruption and kalabule, which -mainly benefited unskilled labor.
The data on real wage trends indicate that sectoral changes in
production were mainly accompanied by changes in wages, however,
these data do not provide information on whether adjustments in
employment also occurred. Table 9.19 presents information on the
intersectoral flow of workers. Among workers who changed jobs since
1983, column (1) of.Table 9.19 gives the distribution of the industrial
sector of origin, and column (2) gives the sectoral distribution of
arrivals. The major flow is observed to be toward the agricultural
sector. This coincides with the growth in the agricultural sector,
especially the production of cocoa as targeted by the- ERP. All our
evidence therefore suggests that the ERP's reliance on market signals
as a way to achieve structural adjustment has probably been helped by
a labor market sensitive to profitable opportunities.. Thus, the
possibility of exporting the Ghanaian experience probably depends, in
part, on both the extent to which the targeted country's iabor market
has the same degree of flexibility as that observed in Ghana (the
. . . . . .~~~~~~~~~~~~~~~~~~~~~
398 P !Jaaidrya adNd K. , Sowla
Table 9.19 Employment Flows Between Sectors of Economic Activity,
1983
(percent.)
(1) 012) (2) - (1)
S9ecwrl Orlg hi Desdntia on N e,t
F7orminig 23,95 49.81- 25.86
Forestry, mIning 3.04 2.3 -0.76
MaQnufacturing 20.15 10.1 -10.01
Construction, transporlution 13.18 6.97 -6.21
Services 39.67 30.8 -8.87
Tolfil . 100.00 100.00 0
Source: Authors' calculations from GLSS.
economic crisis in Ghana in the early 1980s may have contributed to
the observed flexibility) and on the political regime in power.9 In the
case of Ghana, the reduvtion in overt corruption is potentially a major
factor in the improved allocation of resources..
The Distributional Aspects of the Adjustment Program
The central element of the ERP is. the change in relative prices in
favor of the tradable goods sector. To assess the distributional impact
of such a change, it is helpful to consider a highly stylized model of
the Ghanaian economy and to examine whether the theoretical
predictions correspond to our observations. In particular, this exercise
will, permit us to link together many of our previous results and to
identify a group likely to be the most adversely affected by the
adjustment program.
The simplest model of the Ghanaian economy is a two-sector
model with one tradable good sector and one nontradable good
sector. The tradable good cdrresponds to agricultural production and
9. Loxley (1988) appropriately emphasizes that a crisis based on supply
distortions, as opposed to excess demand, is essential to the understanding, and
possible exportation, of the Ghanaian experience,
Ghlana 399
the nontradable good corresponds to services.10 The impact on labor
earnings of a price change in favor of agricultural products depends
on the degree of factor mobility. In the short run, labor is most likely
to be immobiUe, while in the long run our observations suggest that it
is probably quite mobile. The other major factor of production is
land, which is obviously, immobile. Given such a simple model of the
economy, the theoretical predictions of the impact of a price change
in favor of agriculture, that is,. the exportable good, are
straightforward. In the short run, the returns to labor in the service
sector will fall, in terms of both tradables and nontradables, while labor.
earnings in the agricultural sector will rise. Workers will react to these
changes by moving away from urban areas, where services are
concentrated, and toward rural areas, especially export-oriented rural
areas. Those hardest hit by the change will therefore be the
unprotected workers in urban areas.
The data analyzed in the previous sections are generally consistent
with the predictions of this simple model. First, the trends in real wages
observed in table 9.17 reflect major gains for agricultural workers
relative to service industry workers. Second, the changes in regional
earnings differentials derived from table 9.18 indicate that the returns
to working in Accra and in other urban areas fell. significantly over the
course of adjustment. Finally, the returns to unionization, that is,
partial protection from market forces, also increased since the
introduction of the ERP.
Another implication of the mod.1 is that the patterns of migration
should have changed during the course of adjustment. Table 9.20
helps assess this possibility. Column (1) indicates that prior to 1970,
46.5 percent of net migration was directed toward Accra. As predicted
by theory, columns (2) and (3) indicate that the patterns of migration
were completely reversed in the 1980s. Instead of being a region of
net inflow, Accra has become the major source of migrants. Between
1982 and 1987, net migration out of Accra accounted for almost 60
percent of net outward migration in Ghana. The main destination of
10. The manufacturing sector can be omitted from this discussion given its small
size (less than 10 percent of production).
400 P. Beaudry and N, K Sowa
Table 9.20 Net Migration Flows
(percent)
(1) (2). (3)
Up to 1970. Up to 1987 1982-87
Region Destination Origin Destiniation Origin Destination Origin
Western : 21.1 74.0 39.0
Central 22.4 11.0 -6.3
Accra .46.5 15.0 57.6
Eastern 17.0 3.0 27.1
Volta 28.9 31.0 15.3
Ashanti 12.3 . 40.0 20.3
Brong-Ahafo . 20.0 13.0 13.6
Northern & Upper 31.8 13.0 20.3
Totals . 100.0 100.0 100.0 100.0 100.0 100.0
Sources: (1): Ewusi (1984); (2) and (3):'Authors' calculations from the GLSS.
migrants has become the western region, which is the region of
expanding cocoa production.
The combination of theoretical predictions and empirical
observations clearly suggests that those most adversely affected by the
ERP are th.- young, informnal sector workers in urban areas, especially
those working in Accra. This group also includes older workers
retrenched from the public and private sector over the course of
adjustment who have been forced to integrate into the informal sector.
Furthermore, since informal sector businesses employ mostly women,
female-headed households have probably been hit severely by the
ERP. Therefore, any policy aimed at alleviating the costs of
adjustment should probably target the informal sector in Accra (this
does not imply that this group is the poorest, only that it has probably
been hit the hardest)."1 However, the effect of any such policy should
11. The Program of Action to Mitigate the Social Costs of Adjustment
(PAMSCAD) has already begun to target this gtoup for belp..
Glana 401
be considered in a general equilibrium setting given the
responsiveness of migration flows.
The ERP's impact on overall poverty is much harder to evaluate.
The major changes have been in favor of agricultural workers and
against informal sector service workers. Since the relative poverty of
each group is difficult to assess, the overall impact is unclear.12 A
valid appraisal of the relative poverty of each group would require a
detailed analysis of consumption data, but this is beyond the scope of
this chapter. However, casual observation suggests that agricultural.
workers are poorer than informal sector workers, and therefore that
the ERP has most likely reduced the inequality of incomes within the
country..
Conclusion
The main aim of this chapter was to assess whether the functioning
of the Ghanaian labor market accords well with the ERP's market-
oriented export promotion policies. Our assessment is essentially
favorable, even though widespread poverty remains an important
problem in Ghana. Our main finding is that although most workers in
Ghana are either self-employed or employed by household members,
labor allocation and remuneration seem to be affected by market
forces, and that.a market-oriented policy for structural adjustment
may therefore be appropriate. In particular, the determination of both
hourly and monthly earnings appears to conform somewhat to the
theory of compensating wage differentials. Moreover, the evidence
indicates that migration may have responded quite promptly to labor
earnings differentials between regions, and may therefore be
considered as an effective equilibrating force in the long run.
With regard to the specific period covered by the adjustment
program, we have found that changes in labor eamings and allocations
mainly reflect the objectivMs sought by the.ERP. There have been
substantial relative wage gains for agricultural and industrial workers
in relation to service sector workers (especially the retail trade sector).
Accordingly, the main reallocation of workers has been toward the
12. Benjamin and Deaton (1988) conclude for the Cote d'Ivoire that a price change
in favor of cocoa production does not significantly affect the distribution of income.
402 P. Beatdrty an1d N. K. Sowva
agriculture sector, in particular, the Greater Acora regioni has been
losing workers while the western region (the main cocoa production
region) has been gaining them. However, these migration flows have
not as yet been large enough to equilibrate earnings across sectors and
regions. The young, informal sector workers in urban areas, especially
in Accrr:, are still feeling the adverse effects of the ERP. Helping part
of this group to relocate to the faster growing regions of the economy
might be warranted.
Finally, even though we believe that a flexible labor market
probably helped achieve the macroeconomic improvements observed
in Ghana during the 1980s, this factor may be neither the most
important factor nor a sufficient condition for further improvements.
On the one hand, factors such as the reduction in corruption and the
inflows of new capital may have played more important roles than that
of the labor market. On the other hand, the probability of maintaining
growth based on an export promotion policy depends foremost on the
world prices for Ghana's exports, which have been mostly unfavorable
lately. Therefore, an overall assessment of the ERP requires taking into
account the program's different facets, and not only its link with the
labor market, even though the latter provides support for the program.
References
Deaton A., and D. Benjamin, 1988. The Living Standards Survey and
Price Policy Reforrns: A Study of Cocoa and Coffee
Production in Cote d'Ivoire. Living Standards Measurement
Study Working Paper 44. Washington, D.C.: World Bank.
Ewusi, K. 1978. "The Size of the Labor Force and Structure of
Employment in Ghana." Technical Publications Series No.
37. Legon, Ghana: Institute of Statistical, Social, and
Economic Research.
. 1984. "The Dimensions and Characteristics of Rural
Poverty in Ghana." Technical Publication No. 43. Legon,
Ghana: Institute of Statistical, Social, and Economic Research.
Glewwe, P., and D. de Tray. 1988. The Poor During Adjustment: A
Case Study of Cote d'lvoire. Living Standards Measurement
Study Working Paper 47. Washington, D.C.: World Bank.
Glhana 403
Government of Ghana. 1983. Economnic Recovery Programn, 1984-
86. Vol. 1. Accra.
Green, R. H. 1987. Stabilization and Adjustment Programmes and
Policies: Gh2ana. Helsinki: World Institute for Development
Economics Research.
Harris, J. R., and M. P. Todaro. 1970. "Migration, Unemployment
and Development." American Econiomic Review 60.
Lewis, H. G. 1986. Uinioni Relative Wage Effects: A Survey. Chicago:
University of Chicago Press.
Loxlcy, J. 1988. Ghtana: Econtotmtic Crisis and tse Lonig Road to
Recovery. Ottawa, Canada: The North-South Institute.
Mincer, J. 1974. Schooling, Experience, and Earniings. New York:
National Bureau of Economic Research.
Scott, C., and B. Amenuvegbe. 1989. Sample Designs for thne Living
Standards Surveys in Ghaana and Mauritania.; Living
Standards Measurement Study Working Paper 49.
Washington, D.C.: World Bank.
Squire, L. 1981. Employment Policy in Developinig Countries: A
Survey of Issues and Evidence. New York: Oxford University
Press.
Stelcner, M., J. van der Gaag, and W. Vijverberg. 1987. Public-
Private Sector Wage Differential in Peru: 1985-1986. Living
Standards Measurement Study Working Paper 41.
Washington, D.C.: World Bank.
World Bank. 1984a. Ghana: Policies and Program for Adjustment. A
World Bank Country Study. Washington, D.C.
. 1984b. Toward Sustained Development in Sub-
Saharan Africa: A Joint Program of Action. Washington, D.C.
10.
KENYA
William J. Milne
iMonica Neitzert
Kenya, like many other developing countries, suffered through the
economic crises of the mid-1970s and early 1980s. International
economic events have a strong impact on Kenya's economy. The
sources of the economic turmoil included fluctuations in coffee and
tea prices; the rapid increase in oil prices in the mid-1970s and 1979-
80; the world recession of 1981-82 due, in part, to the rise in world
interest rates; the collapse of the East African Community in 1978;
and the drought of 1984. While Kenya-s real GDP did not fall as it
did in many other African countries, changes occurred that signaled
the need for structural adjustment. These included a fall in real
earnings in all sectors of the monetary economy, a deterioration of the
trade balance in 1980-81, and a rapid expansion of external debt.
The labor market has adjusted to these economic events, and
apparently this adjustment has, in part, lessened the impact on
economic growth. Indeed, while real wage rates have declined,
sometimes quite dramatically, unemployment rates in urban areas do
not appear to have increased substantially from the late 1970s to the
mid-1980s. Of course, with continuing declines in real wages, poverty
and basic needs provision are of critical concern. In addition, balance
of payments problems and the size of the external debt remain serious
constraints for sustained medium- to long-run economic growth.
The authors would like to thank Kenya's Ministry of Planning and National
Development for access to some of the data used in this manuscript. Research
assistance was provided by D. E. Hyatt.
405
Table 10.1 Some Macroeconomic Indicators, 1974-89
Real Real
Real Per effective Growth of Growth of per capita External
GDP capita Trade exchange Inflaryn wage real consumption debt Government
growth growtha balanceb ratec rate eniployment ,age8e growth (millions of dficit as
Year - (% (%) K; millions) (1975=00) ( %) (%) (%] USS end year) of DP
1974 3.2 -0.6 -199 106 15.3 . 8.5 -5.1 11.1 999.7 n.n.
1975 2.4 -1,4 . -175 100 15.6 -0.9 -0.6 -41 1,1083 n.a.
1976 4.5 0.7 -115 107 10.4 4.7 5.2 -5.8 1,303.6 . 2.7
1977 8.2 . 4,4 -89 98 12.8 5,3 -4.2 6.2 1,745.6 1.8
1978 7.9 4;1 -304 84 12.5 l;4 -1.4 12.8 2,321.6 4.6
1979 5.0 1.2 -248 85 8.4 6.7 1.0 -1.2 2,8703 4.7
1980 4.0 0.2 -513 84 12.9 3.4 0.9 -3.5 3,51 17 6.3
1981 6.0 2.2 -453 101 12.6 1.9 3.6 -7.8 3,389.0 8.4
1982 7,6 3.8 -332 103 22.0 2.1 -13.2. -1.8 3,518.4 8.7
1983 1.5 . -2.2 -253 114 - 14.5 4.5 -6.1 -4.3 3,763.7 4.3
1984 0.7 -3.0 -320 114 9.1 2.4 0.4 2.2 3,725.4 5.8
1985 4C9 1.3 -385 115 10.8 4.9 -2.2 -8.4 4,402.9 4.3
1986 5.5 1.9 -351 106 5.6 - 3.9 3.0 12.0 4,945.2 7,5
1987 4.8 1.3 -641 120 7.1 3.5 0.1 3,7 5,966.9 3.9
1988 5.2 1.7 -813 126 10.7 3,7 1.9 3.0 5,888.1 4.6
1989 5.0 1,5 -1,219 n.a. 10.5 2.4 1.3 -0.7 n.a. n.e.
n.a. = not available
Nole: The data used throughout this chapter are from revisions to the national accounts provided by th: Central Bureau of Statisdcs, Government of
Kenya. They may not, therefore, match the figures in the historical publications.
a. Per capita growth is the percentage change in real GDP per capita. Computed by the authors as the difference between the growth rate of real GDP
and the estimated population growth rate.
b. The trade balance is the visible balance (that is, merchandise only) in millions of Kenyan pounds.
c. An increase in the index implies depreciation.
d. The inflation rate is based on an annual average of the consumer price index for the average of three income groups in Nairobi.
e. The real wage rate is defined as the total wage employment annual wage rate for all modern sector employees (both private and public) divided by
the average consumer price index for Nairobi.
Sources: Kenya, Republic of, Economic Survey and Statistical Abstract (various issues); World Bank (1990).
Kenrya 407
Origin and Nature of the Adjustment Problem
The period since independence in 1963 through 1973 was
characterized by rapid growth, although with substantial fluctuations,
in the Kenyan economy. Real GDP growth averaged over 6.5 percent
per year-and per capita real GDP growth averaged approximately 3.0
percent per year. Much of this growth can be attributed to increases in
the land available for cultivation, and a movement from large to small
farms and from low- to high-value crops. In addition, land was
transferred from foreign to Kenyan hands through a settlement
scheme involving one million acres during this period. Within the
manufacturing sector, growth was good due to the "easy" phase of
import substitution.
By 1973 Kenya had adjusted some of the easiest options for
further. growth, both in agriculture and in manufacturing. The post-
1973 period can be broken into three subperiods, namely, the
relatively successful reaction to the first oil crisis (1973-78), the
relatively more difficult stabilization and response to the second oil
crisis (1979-84), and the beginning of structural adjustment (1985
onward).
Since 1973 economic events in Kenya have been influenced by
several international economic events. As table 10.1 shows, per capita
real GDP declined in both 1974 and 1975 in response to the rapid rise
in world oil prices in 1973-74. In Kenya, the f.o.b. price of crude oil
doubled in 1973 and almost tripled between December 1973 and the
end of the first quarter of 1974. In addition, international growth was
slower and a widespread drought adversely affected agriculture.
Besides the decline in real income per capita, .these events increased
the inflation rate and caused a large decline in the growth rate of wage
employment in 1975 and falling real wages in both years.
Beginning in 1976, world coffee and tea prices rose dramatically
due to frost in Brazil, which gave a temporary boost to growth. In
1976 and 1977, coffee prices were 130 and 317 percent above their
1975 levels, respectively, while tea prices were 23 and 134 percent
above their 1975 levels in the same two years. The result was that
between 1975 and 1977 Kenyan coffee exports increased sixfold and
tea exports tripled (both in value terms). The result of these increased
408 William J. Mi lte and Monica Nelizert
exports was a significant turnaround in the balance of trade in 1976
and 1977 (table 10.1). During 1976-78, Kenya was able to maintain
an average annual growth rate of real GDP of close to 6 percent, whichi
.implied relatively strong real GDP per capita growth. However, along
with this strong growth came inflation, which persisted even after the
adjustment to the oil price shock of 1973-74.
The period 1979 through 1984 presents a different pattern of
economic growth. The collapse of the East African Community in
1978 significantly reduced Kenya's exports to Tanzania and Uganda
and also led to lower gross fixed capital formation in 1979. There was
no net improvement in real per capita income during this period. The
second oil price shock more than doubled the world price in 1979. In
addition, many developed countries followed restrictive monetary
policies in an effort to control inflation in the early 1980s.; This
resulted in high world interest rates and the worldwide recession of
1981-82, which led to a significant reduction in Kenyan exports, and
therefore slowed growth. Also in August 1982, a coup attempt caused
temporary political instability, which reduced investment spending and
subsequently 'hurt tourism. In 1983, the world oil price fell, but
interest rates remained high, causing Kenya's external debt service
requirements to be large. In 1984, due to the worst drought 'in 40
years, real GDP per capita declined by 3 percent, a reflection of the
continuing importance of agriculture in the Kenyan economy.
The year 1985 marked the beginning of a turnaround in Kenya's
economic fortunes. Good weather, continued declines in world oil
prices, and rising coffee prices led to increases in real GDP per caprta.
Through 1989, the Kenyan economy grew strongly, although the
trade balance continued to deteriorate. Interestingly, in 1988 tourist
earnings overtook coffee exports as the main source of foreign
exchange.
As table 10.1 indicates, exchange rate policy has also been
important. The government announced a small nominal devaluation
of about 5 percent against the U.S. dollar in February 1981, and a
larger devaluation of over 15 percent in September 1981. The next
devaluation took place in December 1982 and there have been no
publicly announced devaluations since, rather, the policy is. one of
continued depreciation of the Kenyan shilling versus a trade-weighted
Kenya 409
basket of currencies. With the mini coffee boom in 1986, the currency
appreciated (in both real and nominal terms) before the downward
trend in the rate continued. With the continuing deterioration of the
trade balance and inflation rates in Kenya running higher than
worldwide inflation, the shilling depreciated by almost 13 percent in
nominal terms against the U.S. dollar in 1988 and another 16 percent
in 1989. A comparison of exchange rate movements and:the trade
balance highlights one of the economy's ongoing problems: despite
the continuing depreciation of the shilling, the trade balance deficit
has increased. While in the first half of the 1980s the government
controlled imports as a stabilization measure, there has not been any
significant export penetration into new markets and the restructuring
of imports does not appear to have occurred. Furthermore, with the
sharp drop in coffee prices in 1988 and 1989, the. balance of
payments problem continues to be serious.
Table 10.1 also indicates another significant problem:. the
significant erosion in both real consumption per capita and in real
wages paid to employees in the modem sector. (The fact that these
series move roughly together suggests that although urban
employment.in the modem sector is only a small part of total
employment, the trends in real wages rates are an indication of the
aggregate effects.) Consequently, the plight of the working poor
remains a serious issue. However, this decline in real wages, which is
due, in part, to government policy as discussed later, has allowed wage
employment growth in the modem sector to remain strong (at a rate
close to the rate of growth of the labor force) throughout the period.
The government deficit, its financing, and the growth of external
debt is also of concern. Since 1980, the deficit as a percentage of GDP
reached a peak of 7.5 percent in fiscal year 1986-87. It has declined
somewhat since, although for fiscal year 1988-89 it still stood at 4.6
percent, which has caused concerns about inflationary pressures.
Internal borrowing froni nonbank financial sources to finance the
deficit has proven difficult. At the margin, the government finances
the remaining part of the deficit through bank sources, both the
central bank and the commercial banks. This tend.s to be inflationary
and points to the need to develop a functioning capital market so that
sonie financing can be done through the nonbank sector. In this
410 Williarn J. MI/ne arid Monica Nelizert
regard, the government has recently set up the Capital Market
Authority to explore ways to make trade in treasury bills and bonds
more efficient, including the establishment of a secondar market.
In terms of reducing the deficit, the largest source of government
revenue remains indirect taxes, since income taxes are hard to assess
and collect. These indirect taixation receipts provide more than half of
total revenues, with sales taxes being the most important source. Other
major sources of indirect taxes are import, excise and export duties.
As of January 1990, the sales tax has been replaced with a value added
tax, which has the advantages of increasing the base to include some
professional and other services as well as removing the bias against the
manufacturing sector caused by the sales tax. On the expenditure side,
the government faces constraints since the major portion of the budget
is in the areas of education and health, both important to the
development process.
External debt is also of concern. Between 1972 and 1978, the end
of the coffee boom, the external debt (measured in U.S. dollars) grew
on average at 36.0 percent per year. During the next six years, the
average growth of the debt slowed to 7.9 percent per year, but then
accelerated to 17.0 percent per year in 1984-87. Since then, however,
external debt has fallen slightly, partly due to debt forgiveness. Recent
loans have also been on fairly concessional termns, for example, in
1988, the interest rate on new commitments from public creditors was
1.9 percent with a maturity of 21.8 years. These concessionary loans
are an important part of the fiiiancing of the deficit. External debt
service (foreign loan payments plus foreign interest payments as a
percentage of total export earnings) increased frora 28 percent in
1981 to almost 38 percent in 1988.. Since Kenya's primary exports of
coffee and tea and its tourist industry are very sensitive to world
economic conditions, this debt service percentage can easily reach
unmanageable levels, A slowdown in world economic growth could
put Kenya in a very precarious economic situation.
Table 10.2 sets out some information on structural change in the
economy from 1972-89. The periods chosen in this table coincide
with particular events that affected Kenya's economy. The first period,
1972-76, includes the first oil price shock and its aftermath; the
period 1977-78 includes the coffee boom; 1979-S1 includes the
Kenya 411
second oil price shock and the drought of 1979-80; the period 1982-
84 encompasses t1he political turmoil of 1982 and the massive drought
of 1984; the final period reflects the beginning of several sectoral
structural adjustment programs and the mini coffee boom of 1986.
As development occurs, one expects the secondary sector
(including manufacturing and building and construction) and tertiary
sector (including the service industries) to increase at the expense of
the primary sector (traditional, agriculture, and so on). While table
10.2 clearly demonstrates the declining share of the primary sector, it
Table 10.2 The Industrial Structure of the Kenyan Economy,
1972-89
(percentage of GDP in constant prices)
Category 21972-76 1977-78 1979-81 1982-84 1985-89
Traditional economy 6.20 .5.82 S.56 5.51 5.51
Monetary economy
Agriculture 33.45 33.38 30.88 30.72 29.08
Forestry and fishing 0.84 0.89 0.99 1.15 1.08
Mining and quarrying 0.34 0.29 0.27 0.22 0.26
Manufacturing 9.84 11.98 12.55 12.54 12.99
Electricity and water 0.63 0,74 0.81 0.81 0.87
Building and construction 4.93 4.24 4.56 3.83 3.27
Trade 12.28 11.28 11.30 10.25 11.17
Transport and communication 5.30 5.25 5.28 6.24 6.13
Finance, insurance, and Tcal estate 4.79 5.26 6.68 7.07 7.55
Other 6.92 5.93 5.65 5.76 5.56
Private household services 0.61 0.78 0.99 1.13 133
Government 13.87 14.16 14.48 14.77 15.19
Total 100.00 100.00 100.00 100.00 100.00
Note: Other includes the ownership of dwellings, other services, and imputed bank
serviceV charges.
Sources: Kenya, Republic of, Statistical Abstract (various issues); Central Bureau of
Statistics revised data.
422 WIllian J. Milne atd Monica Neltzert
is primarily the tertiary sector that has expanded during the period.
Although Kenya is the most industrialized country in East Africa, the
share of manufacturing, at roughly 13 percent of GDP, is still
relatively small, and except for tlle. increased activity during and
immediately after the coffee boom of 1977-78, the growth of
manufacturing output during 1979-89 was slow. A disturbing feature
of the growth of the tertiary sector is the continuing growth in the size
of the public sector. Even with the pronouncement in Sessional Paper
No. 1 of 1986 on Economic Management for Renewed Growth that
employment growth in the public sector would be curtailed, there is
not much evidence of this to date. However, the* government is
committed to recommendation 7.6 of the Sessional Paper, which states
that "employment in government will no longer be guaranteed for'
graduates of university and training programs."
In sum, the period since 1974 has been one of fluctuating
economic growth. However, real GDP has not suffered a dramatic
decline as was the case in many other African countries. The onset of
successful structural adjustment is still relatively recent, and continuing
short- and medium-term problems include the, decline in real wages,
persistent balance of payments difficulties, inflation, and external debt.
The main long-term issue is attaining a growth rate adequate to keep
up with rapid population growth.
The Period of Adjustment: Targets, Instruments, and
Results
As already discussed, the period 1979-84 was one of stabilization
without proper adjustment (van der Hoeven and Vandemoortele
1987), while the period since 1985 has seen more successful efforts at
adjustment. The types of policies in the two periods were not
markedly different, but the difference in effect is probably due to the
intensity of application and the cumulation of different policy
changes.
Fiscal policy was an early priority due to the problems of imported
inflation dating back to the first oil crisis. However, as mentioned
earlier, Kenya has a fairly narrow tax base and only limited options
for cutting government spending. The budget deficit rose in 1980/81
and 1981/82, which led, via the increase in government borrowing, to
Kwiya 4J3
cancellation of an IMF standby agreementi. Throughl the period 1983-
85, the government followed a restrictive stance combined with
reasoniably tighlt monetary policy, wlile continuinig price incentives to
the agricultural sector. In 1985, the government launchedl ic Budget
Rationailization Programme. Through this programrn the governmenlt
aimed to control budgelary expencditures more closely. Budget
resources, whether recurrent or development, were to be allocated
according to well-defined priorities. This resulted in sonic projects
with low potential benefils being cancelled or postponed, and newC
development projects were only funded if they were "productive
investments of high priority." The government's budget deficit
increased during 1986-88, However, much of this was a result of
education expenditures (reform of the school system, which began in
1985, and increased university intakes), the hosting of the All-Africa
Games in 1987, and the general election of 1988. The government has
instituted cost sharing in education and health as part of the structural
adjustment process. Nevertheless, the deficit remains large, and
internal deficit financing from nonbank sources remains a very
serious problenm that impedes adjustment. Finally, there is still no
evidence that the government has curtailed hiring university graduates
who are otherwise unable to find a job (see, for example, Mills 1988).
As regards financial policy, by 1980 the government had become
aware of the problems created in the financial system due to the
continuation of very low and inflexible interest rates, usually negative
in real terms. Interest rates rose substantially from 1980 to 1981,
increasing by some 400 basis points. Current government policy is to
keep real interest rates positiVe to encourage saving. Over the longer
run the government is committed to having interest rates determined
by market forces. However, to date there are still interest rate ceilings,
which may lead to some investment projects being undertaken that are
not as productive as desired, since the capital market may operate
inefficiently. These ceilings may also make it more difficult for the
small-scale enterprise sector to obtain funds. The government
implemented a financial sector reform package in 1989, which aimed
at full interest rate liberalization by mid-1991. In addition, the' reforms
strengthened the central bank's supervision of financial institutions
414 William J. Milmte and Monica Nedizert
and implemented restructuring for ten troubled financial institutions
and two development finance institutions.
As regards exchange rate policy, the currency had a tendency to
slide into overvaluation in the late 1970s,- particularly during the
coffee bo( n. Beginning in 1981, the currency underwent major
devaluatio4is; however, the two 1981 devaluations were undone by
increased inflation. In 1982, wage policy changed to reduce inflation
compensation, and subsequent nominal devaluations did result in real
depreciation of the currency. This has still-not led to an increase in
manufacturing profits: the increased costs of imported inputs,
combined with low substitution possibilities between domestic and
imported inputs, caused problems (Vieira da Cunha 1987). However,
the measure, did address some of the problems of exports.
On trade policy, the government began to shift away from its focus
on import substitution and began a gradual move to import
liberalization. Beginning with the 1983 budget, the government
reduced import duties and import quotas, and by 1985 duties
remained on only some 12 perce"t of imported items. Nevertheless, in
1988 import licenses in two of the four schedules wYere still very
tightly controlled in a manner tantamount to quotas. In 1988 the four
schedules were increased to five, automlatic licensing was introduced in
all but one category, and the licensing system was streamlined. Tariffs
were also cut in 1988, and most specific tariffs were replaced with ad-
valorem ones.
In an attempt to promote exports, the government established
export processing zornes; set up manufacturing under bond, through
which imported inputs-are duty free provided the output is for export;
and introduced the export compensation scheme, which repays imnport
duties on intermediate imports for goods produced for export.
However, implementation had its problems: the export compensation
.scheme entails long payment delays, and the high initial costs of
manufacturing under bond has led to very few applications (according
to unpublished World Bank documents, only 38 had been approved
by 1990).
Nevertheless, the trade balance continues to deteriorate and limited
foreign exchange reserves can constrain economic growth. The
Kenya 415
disequilibrium in the external sector is of considerable concern, and so
far the policies have not achieved better balance.
As for wage policy, the government is heavily involved in wage
setting in the modern sector, both public and private. The 1970s and
particularly the 1980s marked a watershed in wage policy, from the
deliberate high wage policy adopted following indepenedence, to a
wage policy more concerned with aiding employment growth. Real
wages in production fell 25 percent between 1973 and 1976 after the
first oil shock (Vieira da Cunha 1987), but there was more real wage
resistance to the second oil shock, due partly to indexed wages
catching up with inflation. After 1982 the government reduced
sharply the wage indexation allowed by the Industrial Court, and
managed to enforce this in wage contracts. The ability to cut real
wages in a period of stagnant growth likely helped to prevent further
inflation, and also coincided with the government's desire to maintain
employment growth. It also probably gave a disincentive to rural-
urban migration.
The government intensified its structural adjustment policies in the
late 1980s, with an agricultural reform package (1987), industry and
trade reforms (1988), and financial reform (1989). The agricultural
policies aimed at improving the supply of Key inputs, especially
fertilizer, and producer incentives, deregulating markets, improving
public investment and expenditure in agriculture, and reforming
parastatals. The agricultural reforms have had mixed success.
Another aspect of structural adjustment was the removal of price
controls. Price controls, where they exist, can lead to shortages and low
profits. On the specific list of price-controlled items, only 12 items
remain: charcoal, salt, maize and maizemeal, sifted maizemeal, milk,
fats and edible oils, bread, wheat flour, tea, rice, sugar, and beer and
stouts. Further, at the end of 1989 the price of beer at some
establishments was decontrolled and early in 1990, the price of
fertilizer was decontrolled. In terms of the consumer price index, price
controlled items constitute 22 percent of total expenditures for the
middle-income group.
In sum, in the first half of the 1980s, Kenya implemented
stabilization programs and the second half of the decade showed some
evidence of structural adjustment. Further policy changes are required,
416 William J. Milne and Monica Neitzert
however, to permit adjustment on the basis of the "right" prices. The
devaluations and recent policy decisions have made for an exchange
rate that is more responsive to market conditions, and interest rates are
also more responsive to market conditions (despite the interest rate
ceilings, which were scheduled to be eliminated by July 1991). It is,
however, too early to evaluate the success of the structural adjustment
policies.
The Structure of Kenya's Labor Market
This- section describes broad patterns in the labor market, namely,
employment patterns and participation and unemployment rates, and
describes some of the labor market institutions. Kenya remains a
predominantly rural country, with some 80 percent of the population
living in rural areas. Further, much of the urban population is
concentrated in Nairobi, Mombasa, and Kisumu. Data from Fallon
(1985) and the sixth Development Plan (Kenya, Republic of, 1989), as
set out in table 10.3, indicate that the poj9iation in rural areas tends to
be self-employed, while wage employment predominates in urban
areas. Fallon provides some broad groupings of employment in urban
and rural areas based on the Presidential Committee on
Unemployment (1982/83) and Livingstone (1981), and these
definitions are used in the development plan..
As the data in table 10.3 indicate, although small-scale agriculture
is the predominant source of jobs- in rural areas, modern wage
employment is still important. This reflects some large-scale
agricultural activities, such as ranching, tea estates, and coffee
plantations. In urban areas, the growth of the infornal sector is
striking, with an annual average growth rate of 12.8 percent during
1980-88 and a projected average annual growth rate of 10.6 percent
through 1993
Given Kenya's rapid population growth, the big question in the
long run is whether there will be enough jobs for: the rapidly
expanding labor force. Labor force growth can be expected to be
high through the turn of the century. Indeed, the Development Plan,
1989-93 (Kenya, Republic of, 1989) predicts an annual average
growth rate of 4.3 percent in the labor force. This makes sense, as
through 1984 the total fertility rate stood at 7.7 percent or over, and
Kenya 417
Table 10.3 Employment in lJrban and Rural Sectors of Kenya,
Selected Years
1980 1988 1993
Number Number Number
Type of employment (O0's) Percent foo0 os) Percent (00 0's) Percent
Rural
Modem wage employment 544 9.5 443 53 499 4.9
Small farm and rural infonral 4,458 78.6 6,490 782 7,793 76.4
Urban
Modern wage employment 501 8.8 924 11.1 1,183 11.6
Urban informal 168 3.0 441 53 730. 72
Total 5,671 100.0 8,298 100.0 10,205 100.0
Notes: The estimate for modem wage employment in rural areas seems high for 1980,
suggesting a definition change. There is other evidence from the World Bank (1988)
that indicates that total wage employment in rural areas comprised 7.9 percent of total
employment in 1985. This seems more in line with the development plan's estimate
for 1988.
Sources: 1980: Fallon (1985, table 3); 1988 and 1993 projections: Kenya, Republic
of (1989).
consequently, potential labor force entrants are already born.
Although evidence from the 1989 Kenya Demographic and Health
Survey (Nairobi: National Council for Population and Development)
indicated that rates have declined, this will not affect the rate of labor
force growth until well into the first decade of the next century. The
other important consequence of this high fertility rate is that young
people dominate potential new entrants to the labor force (in 1989,
projections indicated that approximately 50 percent of the population
was under the age of 15). These young people typically have high
unemployment rates due to their lack of experience. With a fast
growing supply of labor and with real wages constant or declining, the
issue of poverty is a serious one.
Another factor is the increase in the female urban participation rate,
which rose from 38.8 percent of the working age population in 1977-
78, to 55.8 percent in 1986 (table 10.4). Note, however, that the
change in the reference period from one day in 1977-78 to one week
in the 1986 survey causes some noncomparability. By contrast, the
Table 10.4 Labor Force Participation Rates, Selected Years
(percent)
Urban Rural
Male Female Male Female
Age 1977-78 1986 1977-78 1986 Age 1977-78 1988-89 1977-78 1988-89
15-19 23.9 19.6 23.0 31.8 8-14 55.0 78.1 55.3 82.6
20-24 80.3 73.7 37.9 53.7 15-24 69.0 84.7 79.4 92.8
25-29 93.4 94.5 47.4 69.4 25-64 91.3 97.0 92.0 96.8
65+ 84.2 86.9 76.5 82.5
co 30-34 97.1 98.8 44.2 64.2 Total 83.4 87.2 86.9 91.0
35-39 98.8 96.4 40.1 .61.2
40-44 98.6 99.5 39.1 59.9
45-49 97.9 97.4 47.8 60.2
50-54 89.6 95.3 44.1 53A
55-59 90.t 84.8 34.5 48.1
60-64 87.1 74.0 30.8 47.5
Total 83.9 82.2 38.8 55.8
Sources: Kenya, Republic of (1986a, basic report, table 6,1, table 5.1); Urban Labour Force Survey, 1986 (table 7-1; includes active
and passive job search); Rural Labour Force Survey, 1988-89 (preliminary results)..
Kenya 419
male participation -rate in urban areas did not change appreciably
between the two surveys, and for prime age males, the participation
rate is well over 90 percent. There is a smaller difference between the
female participation rates in the two surveys in rural areas. Not
surprisingly, the participation rate for both males and females in rural
areas is very high; however, the definition for participation is working
a minimum of one hour in the past week in the 1988-89 survey.
Table 10.5 sets out unemployment rates from the 1977-78 and
[986 Urban Labour Force Surveys. Two alternative definitions are
available for 1986, either including or excluding those who engaged
in passive job search. As the African Employment Report 1988 (ILO
1989) indicates, the structure of unemployment,has some important
features. First, young participants have the highest unemployment
rates: those aged 15 to 24 represent two-thirds to three-quarters of the
unemployed. Second, women have,an unemployment rate nearly twice
that of men and their unemployment rates remain remarkably high
even beyond 25 years of age, while male unemployment rates fall
dramatically. Evidence by educational attainment indicates that
unemployment is higber among those with secondary school than
those with no formal education. In this survey, nearly half of the
unemployed males had completed secondary school. By contrast,
university graduates had very low unemployment rates due, in part,. to
the government's policy of hiring university graduates. This
phenomenon of high unemployment among school leavers is also
.consistent with evidence from Collier and Lal (1986); who find long
delays between leaving school and starting work.
Table 10.5 also permits a comparison of unemployment rates in
1977-78 and 1986. Note, however, that the definition of
unemployment in this case is quite narrow (although other choices are
available in both surveys); it only includes active job search and does
not include underemployment. The unemployment patterns by. age
and sex are quite similar for both surveys: both exhibit a U-shaped
curve with age. The overall rate for men declined slightly between the
two surveys, while that for females increased significantly (this may be
due to the increased female participation rates in urban areas). Thus'
overall, the unemployment rate has not changed markedly..
420 William J. Milne and Monica Neitzert
Table 10.5 Unemployment Rates by Age and Sex, 1977/78 and 1986
(percent)
Age 1977-78 1986
Men
15-19 32.2 30.5
20-24 22.2 21.3
25-29 5.6 5.1
30-34 1.9 2.9
35-39 1.8 1.1
40-44 0.7 0.4
45-49 1.1 2.0
50-54 1.3 0.5
55-59 0.2 4.9
60-64 3.'. 0.0
65+ 2.2 n.a.
Average 6.2 5.6
Women
15-19 21.0 22.4
20-24 11.4 22.6
25-29 2.9 7.6
30-34 1.7 4.1
35-39 1.4 4.0
40-44 0.2 2.1
45-49 0.4 0.0
50-54 0.0 1.8
55-59 1.6 0.0
60-64 0.0 0.0
65+ 0.8 n.a.
Average 5.9 9.5
Both sexes
15-19 24.7 24.6
20-24 17.1 21.8
25-29 4.4 6.1
30-34 1.8 3.3
35-39 1.6 2.0
40-44 0.7 0.7
-45-49 1.0 1.5
50-54 1.3 0.8
55-59 1.4 4.1
60-64 3.0 0.0
65+ 2.0 n.a.
Average 6.1 6.9
n.a. = not available
Note: Data are based on a one-day reference period.
Source: CBS, Urban Labour Force Survey, 1986 (table8.6).
Kenya 421
Labor market institutions have important effects on the operation
of labor markets. One key feature is the way wages are determined.
For small-scale agriculture, rural nonfarm enterprises, and the urban
informal sector, wages are market-determined and respond to supply
and demand conditions. However, for modern large-scale agriculture,
modern industrial enterprises, and the commercial and public secto-rs,
wages are set through Industrial Court's wage guidelines. Hence,
collective bargaining forms the basis of the industrial relations system
in these sectors. A number of laws cover working hours, vacations, and
the minimum wage. While union membership is not compulsory, it
accounts for roughly 50 percent of production workers in large-scale
manufacturing. All workers working in unionized establishments,
whether or not they belong to the union, are covered by the collective
agreement. If collective bargaining and voluntary arbitration fail, the
Industrial Court can rnmandate settlements. Thus, the government plays
a large role in wage setting behavior, both through the Industrial
Court's wage guidelines and because the Central Organization of
Trade Unions has government positions on it.
Government policy that sets guidelines for wage negotiations were
introduced in 1973. The original guidelines (a) limited wage
adjustments to full indexation based on the cost of living index plus a
real growth rate that did not exceed the rate of income growth in the
economy as a whole; and (b) required wage agreements to cover a
minimum period of two years,. while allowing only predetermined
yearly adjustments.
After the economic turndown in 1974, the guidelines were revised
in 1975 with the result that productivity increases could no longer be
used to justify wage increases, and full cost nf living increases were
allowed for only the lowest paid groups. Hiowever, given union
opposition and the threat of a general strike, the government reversed
its actions and also abandoned an attempt to reduce real wages in
1976. With the guidelines of 1979, average productivity increases
again could not be passed through, and overall wage increases could
be no more than one-half -the rise in the cost of living. With the
downturn in the early 1980s and the devaluation of the currency in
September 1981, another revision of the guidelines limited the overall
wage increase for all income groups covered by a contract to three-
422 William J. MiMne and Monica NeiJzert
quarters of the rise in the cost of living. The 1982 revision allowed for
separate compensation for housing, which may have reduced the
guidelines' effectiveness. The latest development plan (1989-93)
reiterates government support for the wage guidelines and the notion
of a two-tier wage policy, in which wages paid in rural small-scale
agriculture are market-determined, While in the modern sector wages
are determined through the Industrial Court's wage guidelines.
Measuring the impact of the wage guidelines is difficult as the
average rate of wage increase in contracts settled by the Industrial
Court has typically been below the targets contained in the guidelines.
This is likely due to the continuing large growth in the supply of
labor, which holds the growth of nominal wages in check. However,
the fall in real wages since 1982, following the 1982 revision of the
guidelines, has been quite marked.
Another component of wage setting behavior is for civil servants.
The government reviews and adjusts these salaries roughly every fivc
years. The latest of these reviews was in 1985. The reviews take into
account the cost of living and the affordability of the implied wage
bill for the government. The net result has been that government
employees' wages in real terms have suffered a marked decline since
the mid-1970s.
This section has outlined the structure of Kenya's labor market.
Fast population growth means that employment creation is an
important policy priority. Noteworthy is the increase in the female
participation rate in urban areas during 1977-78 to 1986. Rural
participation rates have not changed significantly. The structure of
unemployment in urban areas indicates that there has been a
significant increase in female unemployment; but a compensating
decrease in male unemployment, so that overall unemployment rates
have not changed. Individuals with some secondary school education
tend to have the highest unemployment rates. Institutional
arrangements in the labor market affect primarily the urban formal
sector and operate through government-inplemented wage gridelines.
The Adjustment of Labor Markets
This section examines labor market adjustment during the 1970s
and the 1980s, including real wages and employment in aggregate and
Kenya 423
by sector of GDP, wages and employment in the public and private
sectors, wages and employment in the formal and informal sectors,
and finally migration.
Sectoral Wages and Employment
One way to examine sectoral adjustment is through the pattern of
sectoral real wage and employment growth. Table 10.6 sets out real
wage increases in various industries in the private and public sectors
during 1974-89, and table 10.7 shows the wage employment share.
These data are based on a survey of employment in the modem sector
(although some of the informal sector is also included), and include
only urban areas. As the table shows, the public sector has suffered a
greater real wage loss than the private sector, although some industries
in the latter sector (for example, mining and quarrying and
construction) have experienced even greater real wage erosion.
Table 10.7 shows the expansion of the service sector, the decline in
agriculture, and the stagnation of the manufacturing industry in the
private sector. While one expects the share of employment in
agriculture to fall during the course of development, structural
adjustment requires a resource shift toward tradables that includes a
large part of modem agriculture, some of modern manufacturing, and
possibly transportation and communication services.
In line with the government's policy on holding wage increases
below the rate of inflation, the minimum wage in real terms fell by
over 40 percent between 1981 and 1985 as table 10.8 indicates. The
minimum wage has fallen quite markedly relative to the average wage,
although the latter also fell. The fall in real w.ages was probably
important in helping to prevent a rise in unemployment.
Another way to examine adjustment in the labor market is through
data on labor productivity and real wages. Figure 10.1 sets out these
data for the modern economy (including both the. prn'-ace and public
sectors). As the figure indicates, the real. wage index dropped
dramatically between 1981 and 1983, while the productivity index did
not show such a severe decline. Part of the explanation behind this
difference is the deflator used in computing the real wage index. The
CPI is used for the real wage, while the GDP deflator for the monetary
economy is used to measure productivity. Vieira da Cunha (1987)
Table 10.6 Real Wage Rates by Industry
(index, 1974 = 100)
Community,
Trade. Transport & Finance, soctal, and
Agricualre Mining & Manufac- Electricity resrauranis, communt- insurance, personal
Secwroyear & forestry quarrying turing & water Construction & hotels cation & real eslare services Total
Private sector
1974 [00.0 100.0 100.0 - 100.0 100.0 100.0 100.0 100.0 100.0
1975 103.3 105.8 98.3 - 100.2 98.4 83.1 95.6 89.3 98.2
1976 111.9 101.1 97.7 - 101.9 95.5 79.0 96.5 96.9 101.6
1977 99.4 68.9 98.7 - 98.5 93.1 78.0 93.0 90.9 98.9
1978 109.2 60.7 94.8 - 104.5 97.1 81.4 85.8 88.3 102.7
1979 108.3 60.1 89.2 - 97.0 98.9 83.2 90.0 94.3 104.1
1980 118,7 56.8 92.3 - 107.2 107.9 81.5 92.0 101.3 114.2
1981 105.7 54.0 83.3 - 98.3 94.9 75.6 87.9 91.5 103.0
1982 93.3 47.2 78.5 - 74.2 82.3 80.1 72.1 79.9 92.9
1983 91.5 44.4 78.3 - 74.4 79.9 68.6 69.8 79.9 90.4
1984 92.1 43.5 76.7 - 72.5 80.0 67.1 69.2 85.3 90.9
1985 89.9 40.6 74.5 - 70.4 76.8 66.5 68.8 84.2 89.1
1986 97.4 41.8 74.4 - 68.1 78.6 67.0 68.8 87.4 90.4
1987 101.5 44.6 77.7 - 76.3 79.7 74.1 72.7 92.4 93.8
1988 105.5 47.1 80.2 - 70.6 85.5 73.7 72.6 89.0 96.3
1989 105.1 60.1 77.7 - 71.3 83.5 64.3 74.0 91.5 97.4
Public sector
1974 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
1975 103.1 93.5 96.8 103.0 113.0 97.9 103.2 95.9 93.9 97.7
1976 156.5 90.9 90.3 95.8 112.5 102.2 103.4 103.8 104.9 108.4
1977 145.2 134.2 81.6 81.1 108.5 81.7 95.2 97.8 103.0 103.4
1978 142.6 168.6 83.4 79.6 101.6 95.9 97.7 104.1 103.6 104.1
1979 134.2 167.3 85.0 85.2 102.8 76.6 105.0 92.6 97.6 100.4
1980 100.4 169.6 88.4 83.3 73.6 87.4 103.2 105.6 93,6 94.7
1981 88.5 149.2 75.3 88.7 77.6 87.0 92.4 100.5 91.8 91.3
1982 78.9 142.7 69.4 70.9 75.3 80.9 83.7 90.0 79.3 80.9
1983 79.2 127.4 65.0 68.4 74.0 81.2 84.3 82.7 77.4 79.3
bi 11984 73.7 114.2 66.5 70.0 77.4 75.1 81.5 83.2 73.6 76.9
1985 72.1 106.7 62.9 68.7 72.8 68.9 76.2 84.0 70.8 74.0
1986 76.1 103.5 61.3 74.7 60.7 67.9 79.9 91.9 75.5 78.2
1987 76.6 105.1 62.1 88.9 59.3 71.0 78.8 89.1 71.2 75.8
1988 81.3 95.1 62.5 89.5 57.0 68.6 80.8 90.6 72.7 76.5
1989 75.7 101.7 66.3 91.7 64.2 69.7 79.3 85.3 73.0 77.7
-indicates no employment in the electricity and water industry in the private sector, hence no wage rates
Note: The public sector includes the central government, the Teachers' Service Commission, parastatal bodies, local govemments,
and firms owned by the public sector through majority control.
Sources: Emnployment and Eamings (various years); Kenya, Republic of (1990).
Table 10.7 Employment Shares by Industry
(percent)
Communiry.
Trade, Transport & Finance, socdal, and
Agrncutture Mining & Manufic-. Electnrcizy restaurants, communti- insurance, personal
Secorf year & forestry quarry(ng turing & water Construction & hotels cation & rea Iestate services Total
Private sector
1974 43.1 0.6 16.5 - 5.9 11.2 3.5 3.8 15.5 100.0
1975 41.1 0.6 17.2 - 5-2 10-9 3.5 4.2 17.4 100.0
1976 39.5 0.6 17.6 - 6.0 11.6 3.6 4.2 17.0 100.0
1977 39.2 0.5 18.0 5.6 11.5 3.7 4.6 17.0 100.0
1978 36.2 0.4 20.2 - 5.5 11.4 4.0 5.0 17.4 100.0
1979 35.4 0.4 20.5 - 5.9 11.8 4.3 5S1 16-7 100.0
1980 32.3 0.3 20.9 - 5.9 12.4 4.3 6.0 18.0 100.0
1981 32.1 .0.3 21.6 - 6.0 12.5 3.5 5.8 18.1 100.0
1982. 31.0 0.3 21.5 - 5.9 12.8 3.6 6.4 18.3 100.0
1983 31.4 0.4 20.7 5.5 13.2 3.7 6-4 18.7 100.0
1984 31.4 0.4 20.7 4.7 13.7 3.5 6.6 19.0 100.0
1985 31.0 0.5 20.6 - 4.3 14.0 3.4 6.7 19.4 100.0
1986 31.1 0.6 20.7 - 4.0 14.2 3.3 6.5 19.6 100.0
1987 30.7 0.6 21.0 - 4.0 14.3 3.0 6.4 19.8. 100.0
1988 29.4 0.5 21.1 - 4.5 14.6 3.2 6.6 20.1 100.0
1989 28.6 0.5 21.1 - 4.7 15.0 3.3 6.7 20.0 100.0
Public sector
1974 14.4 0.2 5.9 1.7 4.6 0.5 8.7 10 63X 1000
1975 13A1 0.2 5.4 2.2 4.6 0.6 8.5 1.1 64-2 100.0
1976 12.7 0.2 5.8 2.4 4.8 0.6 8.3 1.3 63.8 100.0
1977 14.3 0.3 6.2 2.6 5.1 0.6 7.6 1.5 62.0 100.0
1978 13.9 0.2 6.4 2.4 6.8 0.8 7.8 1.6 603 100.0
1979 14.3 0.2 6.2 2.3 6.8 1.0 7.4 1.8 60.1 100.0
1980 12.5 0.1 6.3 2.1 6.7 0.9 6.8 137 62.8 100.0
1981 12.8 0.1 6.1 2.1 5.9 1.0 7.5 1.7 62.7 100.0
1982 11.1 0.2 6.1 2.7 5.6 1.1 6.5 1.8 64.8 100.0
1983 10.2 0.3 6.0 3.2 5.5 1.1 6.4 1.8 65-5 100.0
1984 10.0 0.3 6.2 3.2 4.1 1.0 6.3 . 2.2 66.8 100.0
1985 9.6 0.3 6-1 3.1 4 2 hO 6.1 2.3 67.3 100.0
1986 9.2 0.3 6.u 3.0 5.1 1.1 6.2 2.6 66.5 100.0
1987 9.2 0.1 5.9 3.1 5.1 1.3 6.3 2.6 66.4 100.0
1988 10.1 0.1 5.7 3.1 5.3 1.3 6.1 2.6 65.7 100.0
1989 9.3 0.1 5.8 3.3 5.2 1.3 6.1 2.7 66.2 100.0
-indicates no employment in the electricity and water industry in the private sector, bence no wage rates
Note: The public sector includes the central government, the Teachers' Service Commission, parastaral bodie, local govermments,
and firms owned by the public sector through majority controL
Sources: Employment and Earnings (various years); Kenya, Republic of (1990).
428 William J. Milne and Monica Neltzer*
Table 10.8 Growth Rate of the Real Minimum Wage, 1973-88
(percent)
Growtlh Growth
Year rate Year rate
1973 -8.7 1981 -11.2
1974 19.0 1982 -13.7
1975 8.1 1983 -16.3
1976 .9.4 1984 -4.3
1977 -11.3 1985 8.4
1978 3.7 1986 -5.3
1979 -7.7- 1987 3.8
1980 15.4 1988 2.1
Source: Kenya, Republic of (various years).
Figure 10.1 Productivity and Real Wages, 1973-89
(index 1973 = 100))
115-
105 -~~~~~~~~~~~~~~~
ProductMty - -- - Real 'Wag
75. --
1973 1975 1977 .1979 1981 1983 1985 1987 1989
Year
- Producthrlty -- - - Real WVage
Kenya 429
discusses this different behavior in some detail. HIc argues that the
failure of real wages to decline before 1982 led to substantial pressure
on manufacturing sector profits. This was relieved somewhat after
1982 when real wages began to fall: however, although real consumer
wages (nominal wages deflated by the CPI) fell, real product wages
(nominal wages deflated by the manufacturing GDP deflator)
continued to rise until 1985. Thus, the wage-productivity gap still did
not provide strong relief for manufacturers.
Public Sector Wages and Employment
Public sector wages and employment are another important facet of
structural adjustment. As indicated earlier, in Sessional Paper No. 1 of
15986 (Kenya, Republic of, 1986b) the government announced a
policy to reduce the growth rate of public sector employment. Two
important issues are relevant here: first, in 1988 employment by the
public sector was over 50 percent of modem sector employment, and
second, the wage bill as a proportion of total govemment expenditures
is extremely high (approaching 70 percent). This makes it difficult for
the government to purchase supplies and equipment (nonwage
operating expenses) that can lead to a more productive civil service
(Mills 1988).
Through the first half of the 1980s, employment growth in the
public sector was very high, averaging an annual growth rate- of over 4
percent, but increased to over 4.7 percent per year during 1985-88.
Most of this additional growth was through the Teachers' Service
Commission, where employment grew at over 7 percent per year
because of the implementation of a new education system. Thus, on
the employment, front, there is little, evidence of a government
commitment. to slowing the growth of public sector jobs. This sector
remains the largest source of jobs in the modem sector. The inability
to slow employment growth is probably one factor behind the relative
decline in public sector wages (table 10.6).
The. real wage losses in the public sector have likely led to morale
problems in the civil service, including the possibility of some civil
servants holding more than one job (O'Connell 1987). Indeed, in the
early 1970s, when strict wage guidelines were introduced, a civil
service review commission allowed members of the civil service to
430 Wiliam J. Milne and Monica Neitzert
operate private businesses. The real wage losses are particularly
obvious after 1980. Only in 1986 did some turnaround in these real
wage rates occur and, in general, rising real wages continued through
*1989.
Public sector employees tend to hold on to their jobs despite real
wage losses for several reasons. First, the job provides security: being
fired from a public sector job is very difficult. Second, a public sector
job provides a base for a job search. Finally, employment in the public
sector has oenefits such as access to loans at subsidized rates.
Another problem is with the wage structure of the public service.
On the low enid, wage levels tend to be higher than in the private sector
and, therefore, an oversupply of workers exists. At the other extreme,
for example, engineers and researchers, the pay is substantially lower
than what they could obtain in the private sector or at a parastatal.
Consequently, the government civil service has difficulty retaining
highly skilled workers. Of course, the wage is only part of the
compensation package for government workers. They also receive two
important fringe benefits: a noncontributory pension scheme and a
housing allowance.
The combination of rapid employment growth and fringe benefits
has resulted* in an increasing proportion of government expenditure
going to salaries. This leads to significant inefficiencies in the delivery
of services to the public. Furthermore, it cieates problems for the
government in terms of meeting targets for overall expenditure
growth. In this sense, there is little evidence of structural adjustment.
Informal Sector Earnings and Employment
Of course, modern wage employment- accounts for a relatively
small proportion of aggregate national employment. To consider the
overall trends in employment one must also consider the informal
sector and small-scale farming activities. The definition of the urban
informal sector used in Kenya is that it "consists of semi-organized
and unregulated activities undertaken by self employed persons in the
open markets, in market stalls, in undeveloped plots or on street
pavements within urban centres. They may or may not have licenses
from local authorities." (Kenya, Republic of, 1981, p. iii).
Kenya 431
The most comprehensive information on informal sector
employment comes from the annual Central Bureau of Statistics (CBS)
survey, and is summarized in table 10.9. These figures must be treated
cautiously, however, for several reasons. First, the coverage of the
survey has changed since 1972 in ways that are not completely clear.
Initially, the survey covered only four cities: Nairobi, Mombasa,
Nakuru, and Kisumu. By 1975, the survey covered 11 municipalities
and sampled other towns with populations exceeding 2,000 and rural
trading centers with populations less than 2,000 (Ogundo 1977, p.
79). Further, the nature of the sampling technique is unclear. By 1980,
some 17 municipalities and all towns-exceeding 2,000 population were
fully covered while rural trading centers were sampled. Because of the
steady broadening of the. base on which the informal sector
employment has been estimated, the figures for the early 1970s are
not comparable with those of the 1980s. .(Also there is a discrepancy
between the figures published in the Economic Survey, as reported
here, and the Development Plan, as set out in table 10.3.) However,
there can be little doubt that employment expanded rapidly in the
urban informal sector* after 1981. In 1988 alone, the growth was
almost 11 percent. This coincides with the period of somewhat slower
growth in wage employment. On the basis of the CBS survey and an
estimate of the urban labor force in 1986 (based on participation rates
from the Urban Labour Force Survey) and a demographic projection
model (Milne 1986), the urban informal sector accounts for 21.4 of
total urban employment.
An independent estimate of the size of the informal sector can be
made from the Urban Labour Force Survey, which showed that in
1986, the informal sector (self-employed nonprofessionals and casual
workers) accounted for 21.1 percent of urban employment (excluding
the small group of people who reported work in both sectors).1 This
1. The distinction between the informal and formal sectors is defined by
occupation and employment status. An individual is assumed to be in the formal
sector if he/she is working in one of the following occupations: engineer, technician,
medical, nursing, physical and life science, human relations and other professionals,
public administrators, private and personnel management, accountants, post office,
radio and television, and was either an employer, was self-employed, or was a public
or private sector employee. The other respondents are assumed to be in the infonnal
sector. This definition of the informal sector is quite narrow and this is due to the
432 William J. Milne andt Manica Neftzer:
estimate of the size of the informal sector seems rather small
compared to other countries, but it excludes workers in small-scale
establishments.2 The other 78.9 percent of urban workers had jobs in
the formal sector. Examining the data by gender shows that the
categories do not represent the population evenly. A larger share of
women than men find employment in the informal sector (28.0 versus
17.8 percent), while men are overrepresented among formal sector
workers (82.2 versus 72.0 percent).
The government wants to encourage growth in the informal sector
as it believes this will be the, source of jobs for the expanding labor
force, and it has emphasized this sector (along with the rural small
farm sector) in policy. Table 10.10 examines the success of this
policy, comparing the formal and informal share within'occupational
categories over time, and shows that the informal sector's share lhas
increased overall. The shift into the informal sector was greatest in
agriculture, repair, and transportation, but decreased for sales and
professional personnel. This suggests that the incentive structure has
changed since 1977 and that rents that had been earned in some
informal occupations have diminished.
The pattern of earnings.in the informal sector is more difficult to
measure. However, some data are available from the report on Srnall
Scale Enterprises in Rural and Urban Areas of Kernya (Kenya,
Republic of, 1985a) and indicate that informal sector wvages, in real
terms, increased through the late 1970s before beginning to decline
along with the deterioration in the economy. Table 10.11 compares
male average wages by occupation. The table shows that real wages
fell most dramatically for self-employed sales and agricultural workers
and for wage workers in clerical and production occupations. Real
wages increased s-ubstantially for self-employed service and
narrowness of the questions in the survey. That is, there is no way to identify workers
in small firms that mnight be classified in the informal sector. Consequently, the
inf. ja! sector, as defined here, is substantially underestimated, since it includes
only self-tmployed nonprofessionals, and casual workers.
2. According to a summary of research work on the informal sector by J. Charmes
as reported in Turnham and others (1990, p. 21). Employment in the informal sector
in Asia and Africa is often in excess of 40 to 50 percent of nonagricultural
employment, while the comparable figure for Latin America is above 30 percent.
Table 10.9 Informal Sector Employment, 1972-89
(thousands of people)
Total Manufacturing Services Urban Rural
Year Old New Old New Old - New Old New Old New
1972 33.9 n.a. 2.6 n.a. 31.3 n.a. n.a. n.a. n.a. n.a.
1973 41.4 n.a. 3.2 n.a. 38.2 n.a. n.a. n.a. -n.a. n.a.
1974 76.2 n.a. 0I.9 n.a. 65.3 n.a. n.a. n.a. n.a. n.a.
1975 74.1 n.a. 9.6 n.a. 64.5 n.a. n.a. n.a. n.a.. n.a.
1976 94.9 n.a. 13.8 n.a. 81.0 n.a. n.a. n.a. n.a. n.a.
1977 .103.9 n.a. 15,3 n.a. 88.6 n.a. n.a. n.a. n.a. n.a.
1978 113.9' n.a. 17.0 n.a, 96.9 n.a. 80.7 n.a. 33.2 n.a.
1979 121.6 n,a. 17.2 n.a. 104.4 n.a. 88.7 n.a. 32.9 n.a.
1980 123.2 n.a. 18.2 n.a. 1 Osl0 n.a. 91.7 n.a. 31.5 n.a.
1981 157.3 n.a. 25.9 n.a. 131.4 n.a. 105.5' n.a. 51.8 n.a.
1982 175.4 n.a. 28.8 n.a. 146.6 n.a. n.a. n.a. n.a. n.a.
1983 182.9 n.a. 29.3 n.a. 153.6 n.a. 134.3 n.a. 48.6 n.a.
1984 197.8 n.a. 31.7 n.a. 166.1 n.a. 145.0 n.a. 52.8 n.a.
1985 215.9 254.5 n.a. 43.5 n.a. 211.0 n.a. 166.6 n.a. 87.9
1986 n.a. 281.0 n.a. 49.9 n.a. 231.1 n.a. 182.7 n.a. 98.4
1987 n.a. 312.2 n.a. 58.4 n.a. 253.8 n.a. 202.1 n.a. 110.0
1988 n.a. 346.2 n.a. 66.1 n.a. 280.1 n.a. 223.1 n.a. 123.2
1989 n.a. 390.0 n.a. 74.4 n.a. 315.5 n.a. 251.2 n.a. 13n.7
n.a. = not available
Sources: Kenya, Republic of, Economic Survey (1989, p. 48; 1986); Kenya Informal Sector Survey 1985 (tables IA & 4).
434 William J. Milne and Monica Neitzert
Table 10.10 Occupational Distribution of Urban Employment, 1977
and 1986
(shlare of total employment)
1977 1986
Employmernt
category Formal Iiiformal Fornial Informial
Professional 11.24 0.73 17.85 0.24
Administrativelclerical 22.82 0.45 19.38 0.12
Sales 3.04 9.39 4.52 7.76
Services 17.82 1.01 . 18.90 2.18
Agriculture 1.18 1.07 2.13 3.96
Production 3.88 2.64 3.07 2.51
Repair 9.61 2.25 . 8.39 3.22
Transportation 12.55 0.34 5.48 1.11
Total 82.14 17.88 79.92 21.07
Note: Figures do not sum exactly to 100 due to rounding.
Source: Calculations based on Kenya, Republic of, Urban Labour Force Survey (1977-
78, 1986).
production workers and wage workers in professional, administrative,
repair, and transport occupations. Whereas in 1977 average male
wages from self-employment were substantially higher than for wage
workers in sales, agriculture, production, and repair, by 1986 the
situation was reversed. This is consistent with the view that the
economy had undergone stabilization, but little structural change and
resumption of economic growth. The self-employed sector absorbed
excess labor, with a consequent fall in the sector's relative earnings.
For occupations likely to be in tradables (agriculture and production),
1986 real average wages were below 1977 levels (with the sole
exception of self-employed workers in production). This again is
consistent with a lack of successful structural adjustment.
The rural sector, even though it comprises close to 80 percent of
employment, is even more difficult to analyze empirically as so few
surveys are available. The Integrated Rural Surveys were undertaken
in the 1970s ano a Rural Household Budget Survey was undertaken in
Kenya 435
Table 10.11 Average Wage Rates by Occupation, Urban Males, 1977-
78 and 1986
(1977 Kenya shillings per hour)
1977-78 1986
Nvn- Non;-
Wages & professionial, Wage & professiorial,
casual self- casual self-
Occupatioln workers7 emitployed workcr2 emnployed.
Professional 9.36 * 12.91 **
Administrative/clerical 8.14 6.87 4.37
Sales 3.17 20.88 2.21 7.37
Service 3.18 1.95 3.29 22.50
Agriculture 3.91 10.52 2.72 6.69
Production 3.77 6.14 2.17 10,75
Repair 3.71 12.42 5.67 9.72
Transportation 4.33 5.38 10.02
1. Includes self-employed in professional occupations.
* Very few observations.
'$ No observations.
Notes: The data are unweighted. The weighting corrects for overenumeration of
Nairobi, which was likely a problem in both surveys. The result is that the data are
somewhat more representative of Nairobi residents than the average urban dweller.
Includes self-employed in professional occupations.
Source: Calculated for Kenya, Republic of, Urban Labour Force Survey (1977-78,
1986).
1981-82. One measure of how the rural sector has fared through this.
period is through an examination of the prices paid to farmers
compared to the prices they pay for inputs and consumer goods.
Table 10.12 presents data on agriculture output prices, input prices,
and the agriculture terms of trade during 1977-89. As these data
indicate, the terms of trade have deteriorated since 1980. Noteworthy
in these data is the coffee boom of 1977-78, where the index of the
output price was above the input price, and the mini coffee boom of
1986, jwhen the trend in the terms of trade was temporarily reversed.
436 William J. Milne and Monica Neitzert
Table 10.12 Prices and Terms of Trade in Agriculture, 1977-89
(1982 = 1OO)
Outplut Input Terms of
Year price price trade
1977 79.7 72,6 109.8
1978 80.3 78.0 102.9
1979 79.6 85.4 93.2
1980 83.7 77.2 108.4
1981 90.8 87.1 104.2
1982 100.0 100.0 100.0
1983 -113.6 111.4 102.0
1984 130.0 131.8 98.6
1985 136.8 146.3 93.5
1986 149.0 150.7 98.9
1987 150.3 158.8 94.7
1988 168.7 170.5 98.9
1989 176.4 181.2 97.4
Note; The input price is a weighted average of purchased inputs and an index of
purchased consumer goods in rural areas. The terms of trade is the ratio of the output
price to the input price.
Source: Kenya, Republic of, Economic Survey (various issues).
Rural-Urban Migratdon
Migration between rural and urban areas is also a means of
adjustment in the labor market. The 1979 census suggested that the
rural to urban migration rate (that is, the percentage of the rural
population that migrates to urban areas) is around 1 percent a year.
Given the large differential between urban and rural wages, it is
somewhat surprising that the net migration rate is not higher.
Nevertheless, Barber and Milne (1988) indicate the determinants of
district to district flows are, at least in part, consistent with the human
capital model. They find that distance is an important deterrent to
migration and that economic opportunity variables (especially in the
destination district) are important.
Kentya 437
The 1986 Urban Labour Force Survey provides some further
information about migration. For those who were employed in 1986
and who had moved between 1977 and 1986, table 10.13 shows the
distribution of individuals by the period of move within each
occupation. The period 1977-79 was the coffee boom era, 1980-83
were the years of deteriorating income growth, and the years 1984-86
marked the resumption of steady bu' slow income growth. One would
expect that mobility would be highest during the economic shocks of
1980-83 and for mobility to remain fairly high as the incentive
structure adjusted during 1984-86.
Of those employed in either the formal or informal sector in 1986,
53 percent had made their last move since 1977 (12 percent had never
moved). Apparently, mobility has increased in the 1980s, and has
affected formal and informal workers relatively evenly (informal
sector workers constitute about 21 percent of the employed in the
survey and of the workers that have moved). Professionals and service
workers account for almost 50 percent of those who have moved in
each period. Since the public sector accounts for the bulk of
Table 10.13 Year of Move by Occupation and Sector, 1977-85
(percent)
1977-79 1980-83 1984-85
Formal Informal Formal Informnal Formal Informal
Occupation sector sector sector sector sector sector
Professional 16.01 0.00 37.15 0.30 45.71 0.84
Administrative/
clerical 20.78 0.00 45.54 0.00 33.69 0.06
Sales 9.76 19.56 16.62 28.46 -6.56 18.98
Service 15.08 0.96 31.65 2.87 44.94 4.49
Agriculture 11.38 5.55 8.24 23.59 23.49 27.75
Production 13.63 10.62 22.33 18.22 20.72 14.49
Repair 17.32 4.19 28.84 11.22 25.97 12.45
Transport 16.58 3.28 33.57 5.90 32.40 8.28
Note: Percentages sum across rows to 100 (except for rounding errors).
Source: 1986 Urban Labour Force Data, weighted.
438 Williarti J. Milne and Monica Neitzert
professional and service employment, it is not obvious from this table
that individuals are shifting into tradables. Nevertheless, the table does
indicate a geographically mobile labor force that responds to changes
relatively quickly: for all occupations; except those in formal
agriculture, mobility rose dramatically in 1980-83 compared to
1977-79. The movements correspond to wage incentives: table 10.10
showed that agriculture and sales were the only two sectors where
informal sector earnings exceeded formal ones, and these account for
a large share of informal sector migration.
The government is concerned that the large floWs of people to the
major urban areas of Nairobi, Mombasa, and Kisumu will create
substantial unemployment problems. In Sessional Paper No. 1 of
1986 (Kenya, Republic of, 1986b), a chapter is devoted to the rural-
urban balance that predicts that by the year 2000, almost 30 percent
of the population will live in urban. areas. The policy that the
government has set in place is aimed at encouraging migrants to settle
in secondary towns and smaller urban centers. They hope to achieve
this by developing urban infrastructure in these towns and by,
providing more support for rural trading centers.
Table 10.14 shows the distribution of the labor force by current
and previous residence. The table shows that only a small portion
(13.5 percent) of the surveyed labor force were natives of one of the
five largest. cities in Kenya in 1986. A larger proportion of town
residents, about one-third, were town natives. Further, an equal portion
of the labor force residing in towns or cities in 1986 (about 42 percent
each) had migrated from rural areas at some time. The table also
shows migration.between towns and cities is considerable. About one-
fifth of those who lived in towns in 1986 had previously resided in
large cities, while 39.8 percent of city residents had migrated from
smaller towns. The 1977 data indicate that the proportion of town
residents who had come from larger. cities was almost the same as in
the 1986 survey, 19 percent, while the share of large city residents that
had emigrated from either towns or rural areas was somewhat higher
than in 1986, 91.7 percent compared to 82.6 percent in 1986. This
suggests that.flows to larger centers still account for a larger, but
diminishing, share of migration, and flows to the towns from both
rural areas and cities are significant.
Kfenya 439
Table 10.14 Labor Force by Current and Previous Residence,1986
(percentt)
Current residenice
one of tie Oriier
Previous residetce 5 largest cities town
Other country 3.8 2.3
City (1 of 5 largest) 13.5 20.3
Town 39.8 36.0
Rural area 42.9 41.5
Total 100.0 100.0
Source: 1986 Urban Labour Force Survey, weighted.
Interestingly, as Fallon (1985) indicates, migration among rural
areas is not large even though substantial differences in income exist.
For. example, in the 1981-82 rural household budget survey, Western
and Nyanza provinces had the lowest monthly incomes but the
greatest population pressures; This curtailment of migration is likely
due to cultural and tribal differences between the regions that make
migration difficult.
This section has considered the adjustment of the labor market in
Kenya by examining changes in real wage rates and employment by
sector, and examined agricultural terms of trade and migration as an
adjustment mechanism. On the surface, despite the government's
policy direction, the Kenyan labor market has apparently not
restructured in a significant way. While the evidence suggests that real
wages have dropped significantly and that urban unemployment has
not changed significantly, the informal sector, rather than tradables, is
absorbing the labor slack. The terms of trade in rural areas worsened,
but not dramatically. The data presented on mobility suggest that the
Kenyan labor force is indeed highly mobile, and that relative wages
are important determinants of mobility. However, the data on wages
show that there has not been a clear change in relative wages between
tradables and nontradables, thus as of 1986, there had not been any
obvious shift into the tradable goods sector. This supports the earlier
discussion, which argued that structural adjustment only intensified
440 Williamn J. Milne and Monica Neitzert
after 1985, too late to have marked effects on the 1986 labor force
survey.
Implications of Labor Market Adjustment
This section considers other economic issues related to labor
market adjustment. While Kenya's labor foice had not restructured in
a significant way, at least by 1986, adjustment policies and. other
economic factors are significantly altering the returns to the various
factors of productioni. These changes in returns have imposed costs on
certain groups of individuals who have also tried to minimize their
losses with behavioral changes. This section examines the impact of
adjustment on gross fixed capital formation, on women, and on the
distribution of income and poverty and, analyzes estimated earnings
functions.
The Effect on Capital Formation
Investment or gross fixed capital formation measures the addition
of new physical assets and the replacement of worn out structures and
machines. The need for new capital goes hand-in-hand with necessary
increases in labor productivity. Table 10.15 sets out gross investment
over the period 1974-89.
These figures clearly show the sharp decline in investment
spending following the collapse of the East Africa Community and
again following the political instability in 1982. Indeed, even by 1988,
the level of investment spending had not reached the levels recorded
in the. late 1970s and early 1980s. The sharp drop in investment
spending during 1982-84 poses problems for long-run growth in the
sense that a rising capital to output ratio can imply productivity gains.
Further investment spending is necessary to provide infrastructure and
machines for efficient production.
On a sectoral basis, the manufacturing sector increased its
investment spending in real terms by 16 percent between 1986 and
1988. The government, however, continues to be a major contributor
to investment spending, particularly in the area of infrastructure. This
public infrastructure is particularly important in ensuring that
agriculture output is able to get to the market. Transportation
Kerlya 441
Table 10.15 Gross Fixed Capital Formation, 1974-89
(nillions of Ken yan pounds, constant 1982 prices)
Percentage Percentage
Year Level of GDP. clhange
1974 590.5 29.1 -
1975 605.2 29.0 2.5
1976 598.2 27.5 -1.2
1977 723.0 30.7 20.9
1978 852.2 33.6 17.9
1979 787.5 29.6 -7.6
1980 807.3 29.2 2.5
1981 844.0 28.8 4.5
1982 668.2 21.9 -20.8
1983 576.0 18.4 -13.8
1984 593.6 18.8 Z1
1985 597.2 18.-G 0.6
1986 668.1 19.1 11.9
1987 708.0 19.3 6.0
1988 769.3 19.9 8.7
1989 781.4 19.3 1.6
Note: GDP is measured at factor cost in 1982 prices.
Source: Kenya, Republic of (1989) and Economic Survey (1990).
infrastructure is also crucial in ensuring that Kenya remain self-
sufficient in food.
Women in the Labor Market
As noted in the report of the Woman's Bureau to the Conference
on Women held in N*uirobi in 1985 (Kenya 1985b), women's
traditional role has not been to head households. However, as women
have become more educated, the evidence indicates that womcn are
marrying later and becoming more independent financially. Further,
in rural areas, many women act as heads of households.
The Government of Kenya has significantly aided the education
process by eliminating all school fees for the primary grades and
442 William J. Milne and Mornica Nelizert
introducing a free milk program. However, for women to have more
opportunities, further incentives will' be required at the secondary
school level, but given the state of the Kenyan budget deficit, this may
not be possible.
The number of women in formal employment has been small, but
steadily increasing. In 1964, women held only 12.2 percent of total
formal sector jobs, but this figure increased to 14.8 percent in 1972,
18 percent in 1982.. and 21.3 by 1988. Of course, the sectoral
distribution of employment is also important. In general, women are
employed, primarily in the community, social, and personal services
sector.
Data for employment of women in the informal sector are difficult
to obtain. However, data from Livingstone (1981) based on a 1978
survey of the informal sector in Kenya are set out in table 10.16. As
these data indicate, women make up a significant proportion of
employment in the retail trade and restaurant part of the informal
sector. However, in most cases women do not own Lne businesses, and
while the growth of employment in the informal sector will absorb
-many female labor force participants in the future, remuneration in
the informal sector is significantly lower thani in the formal sector.
Ng'ethe and Wahome (1989) estimated the percentage of women in
the informal sector in 1989, and found that women made up 39.1
percent of employment in four districts of Kenya. This may indicate
that women are playing a greater role in the informal sector.
Changes in the returns to labor are apparent from an examination
of figures 10.2 and 10.3, which show the age-related hourly wage
profiles for men and women, respectively. Figure 10.3 gives the
predicted wage, in 1977 Kenyan shillings, for a self-employed,
unskilled woman with one year of education working in Nairobi. The
figure shows separate profiles for self-employed, nonprofessional
('selfnon") workers and for other employees ("rest") in 1977 and
1986. Figure 10.3 indicates that returns to self-employed,
nonprofessional women dropped dramatically between 1977 and
1986. While earnings from self-employment were higher than for
other wage employees in 1986, the gap had significantly narrowed
since 1977. In 1986, self-employment earnings for women also
peaked with age (in the early 1940s) much earlier than for other
Kenya 443
Table 10.16 Employment in the Informal Sector by Sex, 1978
(number of people)
Percentage
Sector Men Women of women
Retail trade 37,904 26,823 41.4
Rest&urants, bars 9,921 5,389 35.2
Manufacturing 15,192 1,824 10.7
Constmiction and transport 1,181 1 0.1
Other services 15,255 447 2.8
Total 79,453 34,484 30.3
Source: Livingstone (1981, table 6:17).
employees (for whom the peak is in the early 1960s). Consequently,
wages of other employees surpass those of the nonprofessional self-
employed beyond the age of 59. Thus female workers have all
suffered a loss, particularly the self-employed. The picture for men is
more or less. the same. Figure 10.2 shows that there has certainly been
an adjustment in wages between 1977 and 1986, and while all groups
have suffered losses, self-employed workers have been particularly
affected. The gap between male and female earnings does not appear
to have changed much during this period.
Another important aspect of the effect of adjustment on women is
through price increases that occur due to decontrol or other upward
adjustments in controlled prices to reflect market conditions. As, in
large measure, women undertake farming activities in rural areas, these
upward -price adjustments can result in a significant reduction in
purchasing power inasmuch as these increases are for goods and
services they must purchase. Further, with user fees established for
secondary school and for some health care activities, women must
spend a significant portion of their time ensuring they have adequate
funds to provide for their children.
444 Wfllia, J. Mittme and Monica Neitzeri
Figure 10.2 Male Wage Profiles, 1977 and 1986
9- 7
B6 -
i25-
0\
15 20 25-30 35 40 45 50 55 60 65
Age
.- -Rest86 -- -- SelfnonB6 - Rest77 - Selfnon77
Figure 10.3 Female Wage Profiles, 1977 and 1986
Z2-
0-'
15 20 25 30 35 40 45 50 55 60 65
Age
- - Rest86 - -- SelfonB6 - Rest77 Seltfon77
Note: In figures 10.2 and 10.3, the "sclfnon" columns show the results of multiplying
the parameter estimates from table 10.21 setting education = 1, Nairobi = 1, and
female = I (figure 10.3). Since unskilled was the omitted category of skills ftom the
regression, these estimates show predicted wages for unskilled workers in Nairobi
with one year of education. The "rest" category uses the parameter estimates from
regressing the same set of variables for all workers not included in the selfnon group,
that is, casual self-emnployed professionals and regular wage workers.
Kenya 445
The Development Plan, 1989-93 (Kenya, Republic of, 1989), also
notes the importance of women in the labor force. With continuing
higher levels of education, their paiticipation rate is likely to rise,
thereby creating more labor force participants. This is almost sure to
require further. adjustment in the labor market and the government
must put in place programs, such as training programs, so that women
-may become self-employed to address this expanding role of women
in the economy.
Income Distribution and Poverty
An analysis of income distribution is difficult in developing
countries. However, as suggested earlier in this chapter, the issue of
poverty as a result of the continuing decline in real wages must not be
overlooked. Collier. and Lal (1986) estimate that in 1974, 29.5 percent
of the population was below the poverty line. Smallholders form the
largest proportion of the poor, representing 71 percent of persons
below the poverty line, followed by those living in rural areas. Further,
as indicated in UNICEF (1989), the number of those below the
poverty line has probably increased as a result of rapid population
growth and poor econonic perfonnance during 1980-85.
Morrison (1973) indicates that in 1969, the bottom 10 percent of
Kenyans received only 1.8 percent of total income while the top 10
percent received 56.3 percent of income. According to UNICEF
(1989), by 1980 the percent of income going to the bottom 10
percent had declined. In Nairobi, the poorest 40 percent of the
population received 17.2 percent of income in 1969, 15.1 percent'in
1974, and 14.3 percent in 1985. This indicates that over time, the
income distribution is shifting toward the rich, an undesirable result.
Table 10.17, which displays the shares of total employment income-
from all jobs--for 1977 and 1986 by various subcategories, supplies
further evidence of this trend. The first notable fact emerging from
the table is that in the 1986 survey workers aged 15 to 64 eamed
almost all income. In the 1977 survey, individuals younger than' 15 or
over-64 earned 12.6 percent of total employment income for the
surveyed population. The table shows clearly that .the distribution of
employment earnings has changed somewhat between 1977 and 1986.
First, while a smaller portion of the working population was engaged
446 William J. Milne and Monica Neitzert
in the formal sector in 1986, it commanded a larger portion of total
income despite the decline in wages in that sector. This would confirm
earlier findings (table 10.11) demonstrating the more dramatic decline
in earnings in the informal than the formal sector. Thus, the. period
saw a substantial transfer of income from the informal sector to the
formal sector. The bulk of this transfer has gone to male heads of
households. Within the formal sector, employment has shifted from
Nairobi and Mombasa to other urban centers, coinciding with a rise in
income shares for all urban areas. Within the . informal sector
employment has expanded in all urban centers, while the greatest
income losses have occurred in urban areas other than Nairobi and
Mombasa, although atmost every group suffered a loss in its share of
income. A notable exception to this is households headled by women,
which accounted for a smaller share of informal employment in 1986
than in 1977, but a relatively larger share of income (at least in urban
areas other than the two largest cities).
The picture that emerges from table 10.17 is not one of a labor'
force undergoing adjustment. The formal sector accounts for a
smaller share of total employment, but nevertheless earned a larger
portion of employment income in 1986 than in 1977. Meanwhile the
informal sector, which may produce a larger portion .of nontradables,
provided a larger share of employment and a smaller share of
employment income in. 1986. An examination of the figures for
employment shares in 1986 shows that workers migrated to jobs in the
informal sector in Nairobi and Mombasa. The decline in the share of
income from informal employment in smaller towns is particularly
discouraging.
The information in table 10.18 provides the basis for an analysis of
household welfare. This table shows that the share of households with
per capita incomes less than the minimum wage fell between 1977 and
1986 (recall from table 10.9, however, that the real minimum wage fell
by more than 25 percent between 1977 and 1986). Nevertheless, it is'
disturbing that more than one-fifth of female headed households in
the informal sector had per capita incomes below 25 percent of the
minimum wage. Male headed households in the formal sector made
the greatest gains:.almost two-thirds of them had per capita incomes at
least 1.5 times the minimum wage in 1986. This reinforces the view
Kenya 447
that the income distribution was shifting toward those with more
resources, at least up until 1986.
Table 10.17 Share of Total Employment Income Urban Areas, 1977
and. 1986
(percent)
1977 shtare of. 1986 slhare, oft
Work Work
Category Income force Income force
Surveyed workers 15-64 years old 87.4 100.0 98.3 100.0
Fonnal sector 58.0 82.1 76.2 78.9
Residents of Nairobi and Mombasa 43.9 55.2 50.4 41.6
Female household heads 1.3 2.7 1.6 2.5
Male household heads 35.6 40.8 38.6 25.0
Other household members 7.1 11.7 10.2 14.1
* Residents of other urban areas 14.0 26.9 25.8 37.2
Female household heads 1.1 2.8 2.9 5.9
hMale household heads -11.6 20.5 18.9 23.0
Other household members 1.3 3.7 4.0 8.5
Informal sector 29.4 17.9 14.5 21.1
Residents of Nairobi and-Mombasa 18.3 8.8 8.4 8.6
Female household heads . 0.1 1.1 0.3 0.8
Male household heads 7.9 5.0 4.1 4.5
Other household members 10.3 2.8 4.0 3.3
Residents of other urban areas -11.1 9.1 6.1 12.5
Female household heads 1.0 2.4 1.2 0.3
Male household heads 9.5 4.4 3.0 5.4
Other household members 0.6 2.2 1.9 4.5
Workers in both sectors* n.a. n.a. 7.6 1.8
n.a. = not available
* The 1977 survey did not question'respondents about secondary jobs.
Notes: Data for 1986 are weighted, for 1977-78 unweighted. The 1986 data include
casual workers in the informal sector. This would tend to increase the share of income
relative to employment in the informal sector in 1986 as casual workers represented
5.8 percent of employment and 1.9 percent of employment income in 1986. Thus, the
growth of informal sector employment between 1977-78 and 1986 may be overstated
in the table.
Sources: 1977-78 and 1986 Urban Labor Force Surveys.
448 William J. Milne and Monica Neitzert
rable 10.18 Per Capita Household Income Relative to the Minimum
Wage, 1977 and 1986
(percentage of households)
Per Capita IncomeMne(mintum IWage
Category <0.25 0.25c>0.75 0.75c<>1.5 1.5<
1977
Female headed households
Formal sector 30.93 36.09 1.8.55 14.43
Informal sector 50,75 29.85 14.93 4.48
Male headed households
Formal sector 30.32 35.57 19.91 14.21
Infornal sector 33.71 33.15 13.48 19.66
1986
Fcmale headed households
Formal sector 3.14 21.28 29.95 45.63
Infbrmal sector 28.62 23.03 13.27 35.08
Male headed households
Formal sector 2.37 12.90 18.18 66.55
Informal sector 11.01 23.81 17.46 47.72
Note: 1986 data are weighted, 1977-78 data are not. The 1986 data include casual
workers in the informal sector. This would tend to affect only the lowest income
categories and would affect men more than women (more casual workers were men). In
addition, it is most likely that many of the casual workers are probably not household
heads, so the effect of this revision is probably not great.
Sources: 1977 and 1986 Urban Labour Forces Surveys.
An Analysis of Earnings Functions
An alternative method of considering labor market adjustment is
through an analysis of earnings functions. Table 10.19 presents some
regression results for earnings functions for workers undertaking wage
employment in urban areas. Through a comparison of the estimated
parameters some insight may: be gained into the changes that occurred
in the urban labor market between 1977-78 aad 1986.
The equations are* estimated in a semi-log functional form
following Rosen (1977). The dependent variable in the regressions is
Kenya 449
the logarithm of the hourly wage rate. The results are as expected and
are remarkahly similar across the two surveys. Education and
experience (as proxied by the age of the individual) enter positively.
The education variables enter as continu-ous variables, and one can
therefore determine the effect of an additional year of education on
earnings. Of course, this effect should depend on the educational
attainment of the individual as well as the individual's age. These
effects are captured through the introduction of the nonlinear term on
education and the interaction between education and age. Table 10.20
sets out the effect on hourly earnings of one more year of education
for an individual aged 30. As the figures in this table indicate, the
shape of the earnings function has changed somewhat across the
surveys, with less benefit accruing to lower levels of education and
more benefit to higher educational attainment. This probably follows
from the universal primary education program introduced in the mid-
1970s beginning to have an effect in the .1980s through the large
number of potential labor force entrants with at least a primary school
education. However, the costs and entrance requirements associated
with secondary school remain high, yet the demand for workers with
skills provided by secondary school or the polytechnics is increasing,
consequently, the returns to higher education are larger in the 1986
survey than the 1977-78 survey.
In both these surveys a working woman earns less than a working
man. In the 1977-78 survey, the difference amounted to
approximately 12 percent, but by the 1986 survey this had increased
to almost 18 percent. As indicated, earlier, this may reflect the
increasing number of women in the labor force (as measured by the
increase in their participation rate), with the result of a greater
differential between male and female earnings. However, since
educational and skills differences are controlled for, this implies that
women in Kenya are increasingly concentrated in lower paying jobs.
The dummy variables for different skills indicate that professionals
or administrative workers consistently earned about 46 percent more
than unskilled workers (the omitted group) in 1986. The two skills
groups that show the greatest change between the surveys are sales and
service workers and professional workers. The drop in premiums paid
to professional workers may be due to the increasing number of
450 William J. Milnc and Monica Nehlzert
Table 10l19 Estimates of Earnings Functions for Wage Workers
1977-78 srnrvey 1986 survey
Mean Parameter Mean Parameter
Variable (Sid. deviatOi) (1-sainlstle) (StI. deviation) (t-stusislic)
Intercept - -1.0480 - -0.5487
(3.16) (2.59)
Age 33.755 0.0748 32.681 0.0808
(9.48) (4.69) (9.23) (7.75)
-Age2 1,229.140 -0.0008 1,153.138 -0.0008
(711.13) (4.03) (670.58) (6.27)
Education 6.433 0.0224 8.518 -0.0130
(3.91) (0.79) (4.28j (0.79)
Education2 56.690 0.0024 90.856 0.0052
(51.57) (1.96) (69.31) (7.82)
Education x age 205.054 0.0008 268.762 0.0005
(135.93) (1.39) (153.82) (1.50)
Female 0.137 -0.1205 0.248 -0.1799
(0.34) (2.00) (0.43) .(5.52)
Skill classes
Professional 0.202 0.6291 0.248 0.4649
(0.40) (7.84) (0.43) (9.00)
Skilled labor 0.077 0.2635 0.071 0.2075
(0.27) (2.93) (0.26) (3.38)
Semi-skilled 0.116 0.1536 0.455 0.0071
(0.32) (1.93) (0.36) (0.14)
Clerical 0.178 0.4147 0. I 17 0.3758
(0.38) (5.33) (0.32) (6.52).
Sales & service 0.281 -0.0971 0.293 -0.3241
(0.45) (1.49) (0.46) (7.23)
Location 0.476 0.1414 0.462 0.3221
Nairobi (0.50) (3.07) (0.50) (11.6)
Mombasa 0.203 0.2969 0.125 0.3184
(0.40) (5.08) (0.33) (7.57)
Mean of dependent variable 1.275 1.853
Number of observations 1,106 2,416
R-squared 0.421 0.533
Sources: 1977-78 and 1986 Urban'Labour Force Surveys.
Kenya 451
Table 10.20 Effect of One More Year of Education on Hourly
Earnings, 1977-78 and 1986
(percent)
Educational
attainment 1977-78 1986
Primary school graduate 9.11 7.13
Secondary school graduate - 11.23 15.73
Note: Calculations assume the individual is aged 30.
Source: Calculated from the parameter estimates in table 10.21.
professionally qualified workers in the labor force, and may. also-
reflect some adjustment in relative wages. Since the majority of
professionals work in the public sector, an effective adjustment policy
would see their relative wages falling. It is in the sales and service
occupations where at least some of the informal sector is located. With
the rapid expansion of the informal sector, which typically has lower
wages than the modern sector, the narrowing of the wage differentials
between the surveys in these sectors is not surprising.
Residence also plays an important role. In both surveys, living in
Nairobi or Mombasa (the two largest cities) provides a large wage
differential compared to other urban areas. In the case of Nairobi, this
may reflect its status as the capital and the consequent demand for
goods and services by the government.
Table 10.21 shows the estimates of the earnings equations for self-
employed nonprofessionals for 1977-78 and 1986. This is a
subcategory of the informal sector, and is likely to be the relatively
wealthier portion of informal workers. The equations show the
differences in returns to human capital variables for self-employed
nonprofessionals between the two surveys. A striking feature of the
equations is the lack of explanatory power of the human capital model
for the self-employed sector, particularly in 1977. Apparently wages
are not determined on the basis of productive characteristics in that
sector. The results nevertheless confirm that age-related returns have
increased since 1977. The table also shows that the gap between male
and female returns remained more or less constant and not
significantly different from zero. Disparities in the returns between
452 William J. Milne and Monica Neitzert
Table 10.21 Estimates of Earnings Functions for the Self-Employed
Non-Professional, 1977-78 and 1986
1977-78 survey 1986 survey
Mean Pararneter Meani Parameter
Variable (Sid. deviation) (i-statistic) (SId. deviation) (i-statistic)
Intercept - -0.282 9 - -0.7390
(0.25) (0.96)
Age 37.117 0.0400 34.763 0.1021
(10.89) (0.74) (11.04) (2.73)
Age2 1,495.672 -0.0004 1,330.199 -0.0011
(862.81) (0.67) (860.77) (2.46)
Education 3.564 -0.0025 5.574 0.0233
(3.60) (0.02) (3.95) (0.34)
Education2 25.601 0.0016 46.647 0.0050
(33.93) (0.20) (47.48) (1.40)
Education x age 123.453 0.0006 178.533 -0.0003
(133.82) (0.27) (141.72) (0.26)
Female 0.328 -0.2073 0.490 -0.2091
(0.47) (1.08) (0.50) (1.71)
Skil classes
Skilled labor 0.111 0.0191 0.116 -0.2247
(0.32) (0.05) (0.32) (1.06)
Semi-skilled 0.172 -0.1917 0.066 0.0471
(0.38) (0.55) (0.25) (0.18)
Sales & service 0.622 0.0575 0.602 -0.1010
(0.49) (0.19) (0.49) (0.68)
Location
Nairobi 0.161 -0.2418 0.272 0.5167
(0.37) (1.01) (0.45) (3.89)
Mombasa 0.311 0.0801 0.147 0.5047
(0.46) (0.39) (0.35) (2.97)
Mean of dependent variable 1.148 1.665
Number of observations 180 482
R-squared 0.053 0.114
Kenya 453
skills groups are not significantly different from zero. This implies
that factors other than skill variations are more important in
determining wages in that sector, and mnaybe that variations in the
technology of production are not wide. It looks as if location and age
are much more important in determining wages in the informal sector,
and that the importance of these variables has increased since 1977.
The wage premiums earned by workers in Nairobi and Mombasa
increased between 1977 and 1986. As long as workers can earn a
premium in Nairobi's informal sector, attracting them to jobs in the
tradables sector will be difficult. At the same time, with such rapid
growth in the labor force, envisaging a policy that would significantly
alter relative wages without further impoverishing those already at the
bottom end of the income distribution would be difficult.
Table 10.21 thus provides additional evidence that relative wages
had not changed all that much up to 1986, but that all returns had
fallen. This suggests that the Kenyan labor force had not yet reached
the stage of adjustment by 1986, but rather remained in a phase of
stabilization where all incomes were declining. Nevertheless, certain
groups were bearing a heavier cost than others.
All the estimated wage equations are subject to the criticism of
selection bias, as no variable has been introduced that would correct
for any unobserved systematic heterogeneity between self-employed,
nonprofessional w,orkers and others. There is thus the, possibility that
some of the parameter estimates are biased, although theory does not
provide any guide as to the direction of this bias. On the basis of
recent findings for the Cote d'Ivoire, however (Collier and Horsnell
1989), the gender gap may be overstated.
Conclusion
This chapter has attempted to assess the adjustment of the labor
market to macroeconomic sho&ks and the structural adjustment
programs of the 1980s. Of particular interest is the period from 1973
to the present. Through this period, there -were two significant
increases in oil prices, the worldwide recession of 1981-82, a major
drought in 1984, and policies aimed at changing the economy's
structure. The adjustment of the labor market through wage changes,
454 William J. Milne and Monica Neitzert
sectoral changes in employment, increased unemployment, and
increased poverty has been the focus of this chapter.
The deterioration of the economy was particularly evident after the
second oil price shock in 1979-S0. The adjustment to this macro
shock took the form of increased interest rates (a tighter mon&tary
policy), adjustments to the value: of the Kenyan shilling (with its
ensuing depreciation), fiscal restraint (although unsuccessful due to
lack of control on the expenditure sid), policies to influence the
rural-urban balance of the population, and sectoral changes. in the
labor mnarket. Following stabilization effects during 1979-84, the
government instigated policy changes aimed at structural adjustment
from 1985 onward.
The public sector continues to be a large source of modern sector
employment, accounting for over 50 percent of employment, and
while the government is committed to slowing employment growth in
the sector,. to date this has not materialized. Through the adjustment
phase, real wages in all modern sectors fell, although the drop in the
public sector was more pronounced. Indeed, real wage rates seem to
have provided the major part of the adjustment as there do not appear
to have been major changes in the urban unemployment rate between
1977-78 and 1986. This means, however, that poverty has increased,
which may cause problems in providing basic needs.
During the period the informal sector seems to have grown
substantially, although this is somewhat difficult to assess given the
chaniging coverage of the survey of this sector. Furthermore, relative
wages in the informal sector also seem to have fallen since 1979, and
the informal sector's share of earnings has fallen. This again points to
lack of adjustment and resumed growth.
The rate of capital formation slowed during 198144. However, it
is difficult to attribute this drop to the economic crises that occurred.
Nevertheless, in terms of structural adjustment, investment is critical.
Women in the labor market do not seem to have been any more
affected by the economic downturn than men. In terms of the
percentage employed in the formal sector, the.share of women has
increased, although they are more likely to be employed at low paying
jobs. For women wage workers, earnings fell relative to men's, but
Kenrya 455
there was n1o change in women's relative earnings. ih the self-
employed category.
In sum, the labor market has adjusted through the period, but the
government still faces serious problems, both with regard to balance of
payments considerations and the budget deficit. While some structural
policies have been put in place, it is somewhlat too early to assess their
effects, and significant further changes must be made.
References
Barber, G. M., and W. J. Milne. 1988. "Modelling Interregional
Migration in Kenya." Environ1ment and Planini-ng A(20):
1185-1196.
Collier, P., and D. Lal. 1986. Labour and Poverty in Kenya, 1900-
1980. Oxford: Oxford University Press.
Collier, P., and P. Horsnell. 1989. "The Mobility of Women's
Labour and Structural Adjustment: Issues and Some Evidence
from the Cote d'Ivoire." Processed.
Fallon, P. R. 1985. "The Labour Market in Kenya: Recent
Evidence." Processed.
ILO (International Labour Organisation). 1989. African Employntent
Report, 1989. Addis Ababa, Ethiopia: World Employment
Program.
Kenya, Republic of, Central Bureau of Statistics. 1981. Employmertt
and Earnings. Nairobi: Government Printer.
v Various issues, 1990. Economnic 1989 Suirvey.
Nairobi.
. Various issues. Statistical Abstract. Nairobi.
Kenya, Republic of. 1985a. Small Scale Enterprises ill Rural and
Ulrban Areas of Kenzya. Nairobi: Central Bureau of Statistics.
-_________ .1985b. Women of Kenya: Review and Evaluation of
Progress. Nairobi: Kenya Literature Bureau..
--_ .1986a. Labour Force Survey, 1977-78: Basic
Report. Nairobi: Central Bureau of Statistics.
456 William J. Milne atid Monica Neitzert
. 1986b. Sessional Paper No. 1 of 1986: Economic
Management for Renewed Growth. Nairobi: Government
Printer.
-. 1988. Urban Labour Force Survey, 1986. Nairobi:
Long Range Planning Unit and Central Bureau of Statistics.
. 1989. Development Plan,. 1989-93. Nairobi:
Government Printer.
Livingstone, I. 1981. Rural Development, Employment and Incomes
in Kenya. Geneva: International Labour Organisation.
Mills, M. 1988. "Kenya Public Expenditure Review: Government
Employment and Personnel Expenditures." Nairobi: World
Bank, East Africa Department. Processed.
Mimne, W. 1986. A Nationtal and Urban-Rural Population Model for
Kenya. Technical Paper 86-01. Nairobi, Kenya: Ministry of
Planning and National Development, Long Range Planning
Unit.
Morrison, C. 1973. "Income Distributioni in Kenya." Washington,
D.C.: World Bank. Processed.
Ng'ethe, N., and J. G. Wahome. 1989. The Rural Non-Farm Sector in
Kenya: A Study of Micro-Enterprises in Nyeri, Meru, Uasin
Gishu and Siaya Districts. IDS Occasional Paper No. 54.
Nairobi: University of Nairobi, Institute for Develo'pment
Studies.
O'Connell, S. A. 1987. "Fiscal Policy in Low-Income Africa."
Processed.
Ogundo, 0. 0. 1977. "Data Collection in the Informal Sector." In
F. C. Child, ed., Employment Technology and Growth and the
Role of the Intermediate Sector in Kenya. Occasional Paper
No. 19. Nairobi: University of Nairobi, Institute for
Development Studies.
Rosen, S. 1977. "Human Capital: A Survey of Empirical Research."
In Research in Labour Economics. Greenwich, Connecticut:
JAI Press.
Kenya 457
Tumbham, D., B. SaIom6, and A. Schwarz. 1990. The Itformal Sector
Revisited. Paris: OECD.
UNICEF. 1989. Situationi Analysis of Clhildren and Women in Kenya.
Nairobi.
van der Hoeven, R., and J. Vandemoortele. 1987. Stabilization and
Adjtustment Policies and Programmes: Kenya. Helsinki,
Finland: World Institute for Development Economics Research
of the United Nations University.
Vieira da Cunha, P. 1987. "Trends in Kenyan Manufacturing Wages
and Their Impact on Trade Liberalization." Washington,
D.C.: World Bank. Processed.
World Bank. 1988. "Employment and Growth in Kenya." Report
No. 7393-KE. Washington, D.C.: World Bank.
-_ _ .1990. World Tables. Baltimore, Md.: Johns Hopkins
University Press.
MALAYSIA
Dipak Mazumdar
Malaysia has maintained sustained growth for much of the 1970s. As
an oil exporter the cousntry received windfall gains during the two oil
price hikes, and in the 1970s, a commodity boom also helped growth.
Difficulties emerged with the decline, of oil prices in the 1980s, which
were accompanied by a fall in commodity prices. Malaysia bungled into
a rather severe depression in 1985-86, but to the surprise of some
observers, the economy recovered very quickly and growth resumed in
1987 and 1988.
The events that led to the recessioni and its quick turnaround are of
interest because Malaysia represents a Southeast Asian prototype. The
Republic of Korea and Thailand-although they are not oil exporters-
seemed to have experienced much the same type of severe but short
depression in the 1980s, followed by a strong recovery. This chapter
examines macroeconomic policies that led to the depression and the
recovery of the 1980s and-analyzes the behavior of labor markets in this
context.
The Basic Structure of the Economy
We will discuss in this section the more salient aspects of Malaysia's
economic structure, which had a significant effect on its growth and
adjustment.
This chapter is a revised version of a document produced for the research project on
the Human Resources Development Plan, Module 1. Other papers in the module on which
this. chapter has drawn are those by Gan Wee Beng and L. Krause, R. Richardson and
Soon Lee Yin, and to a lesser extent, S. Bhalla. I have also benefited from reading the
interim report on the whole project by R. Lucas and D. Verry. l am grateful to Homi
Kharas, Sue Horton, and members of the Economic Planning Unit in Malaysia for their
helpful comments. Special thanks are due to Homi Kharas for supplying me with clean
data tapes of the household surveys.
459
460 DipakMazumdar
Openness
Malaysia is a very open economy. Exports as a percentage of GDP
hovered just below 50 percent in the 1970s. This share started to increase
in 1983, rose more sharply after 1985, and reached a record level of 72
percent in 1988.
Table 11.1 gives the composition of exports by the most important
commodity groups for various years. The table shows that Malaysia's
economy has been based on natural resources, but that the proportion of
manufactures in total exports increased dramatically in the 1980s.
Originally rubber dominated the export scene, but over the years a
diversified group of commodities-tin, palm oil, sawlogs, timber-have
become more important. Petroleum became important after the oil price
increases of the 1970s. The growth rate of the export of manufactured
goods has been nothing short of spectacular in the last decade, but unlike
some of the newly industrializing countries, exports manufacture has
Table 11.1 Composition of Exports in Malaysia, Selected Years
(percent)
Export 1970 .1975 1980 1984 1987
Food, live animals, beverages,
tobacco 5.92 6.72 3.70 3.90 5.56
Crude materials, inedible 53.79 35.00 32.32 21.10 23.46
Mineral fuels 7.08 1OA8 24.49 29.59 19.76
Animal and vegetable
oils and fats 6.00 1634 11.11 15.18 9.15
Chemicals, manufactured goods,
machinery and transportation
equipment, miscellaneous
manufactures 26.08 30A0 27.82 29.76 41.58
Other exports 1.12 1.05 0.55 0.55 0.48
Plote: Mineral fuels include coke, coal, petroleum, petroleum products, and gas.
Sources: Bank Negara (1986,. table VII.6); Ministry of Finance Economic Report 1988-
89.
Malaysia 461
been highly concentrated in Malaysia, with electronics and electrical
machinery accounting for over half of total exports in this category.
Instability of exports earnings is a feature of the Malaysian economy,
and the high ratio of exports means that this instability is liable to
produce large swings in GDP. Although commodity exports are
diversified, large fluctuations in earnings from this group of exports are
not at all uncommon, for example, the 1985-86 crisis was caused by a
simultaneous fall in prices for all five commodity groups. A World Bank
study (1988, vol. I, appendix 1) calculated that the standard deviation of
the rate of change of export earnings from commodities due to price
changes was 15.8 percent. "This implies that one-third of the time, export
earnings are likely to deviate from their expected value by more than
M$2 billion (one standard deviation), and one-sixth of the tinme, there will
be a shortfall of this magnitude'" (p. 125).
The heavy concentration of manufacturing exports in an industry that
is distinguished for its volatile world mark'et adds to the instability. The
recession of 1985-86 was in no small measuire due to the shakeout in the
world semiconductor market.
Importance of the Public Sector
The public sector's role in the economy has increased throughout the
period of Malaysia's recent economic growth. The new economic policy
initiated in 1973 had as its objective a restructuring of the economy with
a view to giving the native Malays (Bhumiputras) a greater share of the
economic pie than they had traditionally enjoyed. A major instrument in
this transformation was the expansion of employment in the public
sector, in which Malays had a favored role.
At the same time, the boom in commodity and oil prices during the
1970s vastly increased the government's resources, which enabled it to
sustain its policy of expansion. The expansion took the form of a high
rate of growth of the government's wage bill, as well as a massive
increase in public investment through nonfinancial public enterprises and
other statutory authorities. During 1981-88, public consumption and
investnent together accoun'ted for an average of 37 percent of GNP, and
public investment was 41 percent of gross capital formation in the
economy (Malaysia Ministry of Finance 1988, table 2.1, p. x).
462 DipakMazumdar
The large size of the public sector has meant that it has had an
important influence on business cycles in the economy, supplementing
the impact of the external sector. As we shall see, public spending has
fluctuated sharply in response to external shocks, and not always in a
countercyclical way.
The Role of Food
An important variable that determines the impact of external shocks
on the economy is whether food is largely a tradable or a nontradable
good. The behavior of the consumer price index (of which food is an
important component) relative to the index of producer prices depends on
the "tradableness" of food.1 If, for instance, an extemal shock causes the
exchange rate to appreciate, the price of tradables falls relative to the
price of nontradables.2 If food is a tradable commodity, its price also falls
relatively, and a downward pressure is exerted on the consumer price
index. If real wages are sticky, money wages will still decline relative to
producer prices and the pressure on profitability in the traded goods
sector will be eased somewhat. However, if food belongs to the
nontradable category, product wages are likely to increase, thereby
adding to the burden of the producers of tradables unless real wages fall.
Rice is an important component of the basket of goods consumed by
Malaysian workers. Ordinarily it would be a commodity traded in the
world market, but Malaysia has a system of administered prices for rice
for both producers and consumers. The system is administered by the
National Paddy and Rice Authority; which is responsible for marketing
and pricing rice. The object is to maintain a level of prices for domestic
paddy farmers that is higher than the border price .of rice. The authority
imports rice as necessary to supplement the amount it can procure from
domestic producers, and sells the rice to consumers through its retail
outlets at predeternined prices that are not very different from prices paid
to farmers. In 1974, the cabinet set rice prices at parity with border prices
near the peak of the external market, and they have remained virtually
unchanged despite a halving of import prices. In nominal terms, the
1. The degree of 'tradablencss' oE food varies inversely with the extent to which
domestic consumers are insulated from movements in world prices.
2. This is the typical "Duteb disease" case of an oil exporter when the price of oil
increases.
Malaysia 463
constancy of rice prices meant that the price had fallen 39 percent in real
terms by 1984.
Thus, one component of the conisumption basket is virtually insulated
from external events, including exchange rate fluctuations. The consumer
price index in Malaysia tends to move with the commodities and services
in the basket whose prices are free to vary, which tend to be largely
nontradables. Of course, the extent of the variation is less than the price
of nontradables because of the constant nominal price of one component
of the index.
Thus, Malaysia belongs to the group of countries in which during
periods of increase in the ratio of prices of nontradables to tradables, the
squeeze on profitability in thei tradable sector tends to be accentuated by a
rise in the product wage, even if the real (consumer) wage is constant.
Growth and Cycles in the Malaysian Economy
Figure 11.1 plots the yearly rate of growth of GDP in real terms for.
1970-88. Becaluse Malaysia is such an open economy, the same figure
also shows the index of the terms of trade. The figure shows the high rate
of growth of 7 percent or more that Malaysia has been able to sustain for
most of the period. In the 1970s, growth was interrupted in a major way
only in 1975. More difficulties emerged in the 1980s, culminating in the
deep recession of 1.985 and 1986. However, the recession did not last
long, and by 1988 the economy had bounded back to a growth rate higher
than 7 percent.
The figure illustrates the close relationship between the GDP growth
rate and the terms of trade. A fall in the terms of trade in 1975 was
associated with the sharp downturn in 1975, while the sustained
improvement in the terms of trade for the next five years seems to have
been reflected in the recovery and maintenance of a generally high GDP
growth rate in the second half of the 1970s. In the 1980s, the relationship
between terms of trade changes and yearly variations in the growth rate is
particularly close, but the deep recession of 1985-86 was much larger
than the percentage fall in the terms of trade. Similarly, the recovery in
1987-88 is much stronger than would be warranted by a mere terms of
trade improvement. Although led by terms of trade movements, other
factors were cIearly involved in the rather strong cycle of the 1980s. We
shall now discuss the four phases of the cycle in more detail.
464 DipakMazurndar
Figure 11.1 Terms of Trade and GDP, 1970-88
130- 12
120 1-
-110 DI1
! 100 ., -.
, 10 . g I _1
:. 90~~~~~~~~~~~~~~
90- __ - O
-10-~~~~~~~ I I . I 1 -1-
80 - - Y -ar
Note: Index fo terms of trae: 1978 = 100
740 -
Phase~~~~I 1:TeUsin.17-9
: 30.
420 =~~~~~~~~~~~~~~
40 -~~~~~~~~~~~
30 -~~~~-I
20 -~~~~ 2 I 1
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988
Year
-Ternms of Trade -- -- GDP
Note: Index for terms of trade: 1978 = 100.
Sources: Krause (1989, table 1); Ministry of Finance (various years).
Phase L, The Upswing (1975-79)
As mentioned earlier, the upswing was fueled by a sharp increase in
the terms of trade, which were 51 percent higher In 1979 than in 1975.
'The prices of crude oil, rubber, tin, and palm oil all rose simultaneously.
rme volume response was also positive, but was particularly strong for
palm oil. The long-term investment in palm oil was now paying off and it
emerged as the leading agricultural export in this period, almost catching
up with the share of rubber in total exports. In value terms, however, the
most dramatic increase in export share was in petroleum, up from 9 to 24
percent during the period.
Malaysia 465
The large increase in the value of exports increased domestic income
through the multiplier process. Insofar as this type of economv depends a
great deal for its revenue on taxes on the external sector, the resources for
public spending were augmented, but during this phase of the cycle, real
incomes in Malaysia rose much more rapidly than expenditures. Savings
rose to record heights, averaging 29.6 percent of GNP in 1976-79.
Investment, although growing through the period at. a substantial rate,
averaged less, at 25.3 percent, so that a current account surplus in the
balance of payments was maintained throughout this period (figure 11.2).
The respective roles of the private and public sectors in the favorable
resource position can also be inferred from figure 11.2. The federal
Figure 11.2 Balance in the Federal Budget and in the Current Account of
the Balance of Payments, 1971-88
12
10 /
6-
4_- . - - ' -
-o -
0-6
- 1- 1 ' '., .'
-2 ~ ~ ~ ~ 1-- - .1I I I I
1971 1973 1975 1977 179 1981 1983 1985 1987
* M:ar 'ear
-- Federal Budget - --- CurrentAcxount
Source: Lucas and Verry (1989).
466 DipakMazumdar
budget deficit increased in absolute terms during this period, but not in
any marked way. As a proportion of GDP it was actually fairly constant.
This ratio is plotted in figure 11.3, together with the private sector
saving/investment imbalance for the 1970s and 1980s. It shows the
contribution of the positive private savings balance in sustaining the
external accounts surplus during the upswing of the late 1970s.3
Figure 11.3 Consolidated Public Sector Fiscal Deficit and the Private
Sector Saving/Investment Imbalance, 1971-88
(percentage of GDP)
14 -_
12 -- ,,
10 ,
8 -
4 -~~~~~~~
6 .* * ' .
}4~~~~~ .1 'I'",-'
2 -~ . I
-0
-120
-16-
-18 ~ ~ ~ ~ ~ ~ -
1971 1973 1975 1977 1979 1981 1983 1985 1987
Year
- Fiscal Deficit - - Printe Sector Savlng4nvestment Imbalance
Source: Lucas and Verry (1989).
3. The proportion of consumption goods in total imports fell from 22.2 percent in
1975 to 18.4 percent in 1980, while the proportion of intermediate goods went up from
41.3 to 49.9 percent. Although growth in Malaysia is dependent on imported inputs, the
relatively low marginal propensity to import consumer goods helped to prevent
deterioration of the balance of payments during this period.
Malaysia 467
Phase 11: Thfe Aiteinpt to Sustain the Upswing (1980-84)
Malaysia's external terms of trade peaked around the inid-1980s.
They then turned down until 1986 as commodity prices broke, with a
temporary respite in 1984. The decline in the terms of trade during 1980-
84 was around 20 percent. At the same time, the slowdown in the
economies of the OECD threatened to reduce the growth rate of
Malaysian exports.
The government's response to the downswing in the external sector
was to adopt a vigorous countercyclical fiscal policy in the expectation
that the recession affecting external trade was only temporary. Many
people were predicting that petroleum prices would continue to rise and
that OECD growth would quickly recover. The expansionary fiscal
policy took the form of a very sharp increase in federal government
expenditures, which was not compensated by an increase in revenues.
The swelling budget deficit, showed in f:gure 11.2, closely follows the
pattern of increase in government expenditures. It peaked in 1982, and
although it declined in successive years, the ratio of public deficit to GDP
in 1984 was 12.3 percent, nearly double the level of the mid-1970s.
There are only three ways to finance a budget deficit: inflationary
finance through an accommodating monetary policy; a reduction in
pnvate domestic absorption through a rise in domestic savings and/or a
fall in investment; or an increase in foreign borrowing. In Malaysia, the
monetary authorities forestalled the first option. The government adopted
a policy of monetary restraint explicitly to prevent the budget deficit from
spilling over into inflationary pressures and curTent account deficits in the
balance of payments. From 1980 onward, the annual rate of growth of
money supply (MI) declined continuously, registering a negative growth
rate of -0.6 percent in 1984.4 As a consequence, the rate of inflation
(measured by the GDP deflator) was at a lower level in phase II, with an
average of 4.2 percent compared to phase I, when it averaged 8.9 percent.
As far as private domestic absorption was concerned, unlike in phase
I, private savings fell sharply in phase 11 of the-cycle. This is, of course,
4. This policy was in sharp contrast to the general practice of Latin American
monetary authorities who seem to follow passively the needs of the fiscal authorities for
monetary accommodation. The difference in institutions involved in economic
decisionmaking in Malaysia (and other East Asian countries) and Latin American
countries is an interesting topic for invesligation.
468 DipakMazunmdar
to be expected if consumers behave rationally in attempting to smooth
out fluctuations over time. Savings out of transitory income gains
increased during an upswing of the ternts of trade and decreased when
the downswing brought unanticipated losses in income. Private
investment did not fall until the recession years (phase III of the cycle);
although its composition might have changed. Thus, the saving/
investment balance in the private sector moved in a way opposite to what
was required to offset the government budget deficit (figure 11.3).
Thus, only one way remained to finance the deficit: borrowing from
external sources. As a result of external debt financing, the debt: GNP
ratio increased dramatically from 9.4 in 1980 to 39.0 in 1984.
Furthermore, a great deal of the borrowing was done through commercial
banks at variable interest rates.5 Loans of this type increased from 45
percent of total debt in 1980 to 70 percent in 1984, and a high of nearly
80 percent in 1985. The high international interest rates of the early
1980s increased the average interest cost of external debt from 8.1
percent in 1979 to a high of 13.1 percent in 1981 before it fell gradually
to 9.7 percent in 1984.
Throughout this period, the Malaysian authorities took a passive
attitude to the exchange rate. The capital inflow triggered by the budget
deficit was instrumental in causing a significant appreciation of the real
exchange rate. This appreciation, together with adverse; movements in the
labor market, reduced Maleysia's competitiveness in the world market
and threatened to create an unsustainable deficit in the current account of
the balance of payments..
Phase 1II: The Period ofAdjustment and Recession (1985-86)
The management of economic policy in Malaysia became sensitive to
the emerging economic problem soon after the explosive budget defizit
of 1981-82. Measures to cut govemment expenditure were initiated in
1983. The ratio of consolidated public deficit to GDP was drastically
reduced from 18 percent in 1982 to 7 percent in 1985. The improvement
S. "A substantial proportion of the foreign borrowing (57 percent in 198344) was
undertaken by public enterprises. This recourse to extemal funds helped tlhese agencies
escape the surveillance and discipline that could have been imposed by the Federal.
Government had there been a greater reliance on the Treasury as source of funds." (World
Bank 1988, vol. I, p. 15).
Malaysia 469
of the terms of trade in 1984 proved to be temporary, and Malaysia was
hit by a further, drop in this key variable in 1985-86 (figure 11.1).
Without an offsetting rise in nublic expenditures, Malaysia sustained a
severe recession with the rate of growth of GDP actually turning negative
for the first time in 1985 and barely positive in 1986.
The recession, short as it was, managed to correct the basic
imbalances in the economy fairly quickly. As figure 11.3 shows, the
fiscal deficit continued its dramatic improvement in 1.985, but failed to
sustain it in 1986. Cutting budget deficits in a year of deep recession
when revenues are generally falling off is difficult. The private sector
surplus of savings over investment, however, continued to increase
strongly. Private savings no doubt declined with falling income during
the recession, but private investment fell faster. Taking the public and
private accounts together, the excess of spending over income finally
disappeared in 1986. At the same time, these two years of recession saw
for the first time the phenomenon of the current account of the balance of
payments -improving when the terms of trade were falling. This is
because the value of imports fell due to the recession while exports
registered a modest increase.
Phase [V: The Recovery (1987-)
Somewhat to -the surprise of observers, the Malaysian economy
registered a turnaround and a rate of growth of GDP of 5.2 percent in
1987. The performance in 1988 was even better at 7.9 percent, signifying
that the recovery was well underway. The upturn was again fueled by the
external sector with the terms of trade improving by 18 percent; the
prices of the major nonoil commodities once more moving up in unison.
At the same time, exports vol]me, which had already started to grow in
1986, surged forward in 1987-88.
A major development in the behavior of exports was the leading role
taken by manufactured goods, whose share in total exports climbed to 48
percent by the end of 1988. However, it remained heavily concentrated in
electronics. "The robust expansion in electronics demand in turn follows
from the economic growth of the leading industrial countries like Japan
and the United States which are currently in their sixth year of growth as
well as from the relocation of Japanese investments overseas"
(Government of Malaysia 1988, p. 114). At the same time the continued
470 Dipak MazwnJdar
depreciation of the ringgit helped ttie competitiveness of such exports in
the world market. The combined effect of price and volume increase was
that the current account of the balance of payments showed a sizeable
surplus for the first time since 1979 (figure 11.2).
During this period of recovery; the government restrained the growth
of public expenditure. With the public sector deficit holding steady in
proportion to GDP and the private savings/investment gap still remaining
positive, there was no need for borrowing from abroad. In fact, the
accumulation of reserves through the surplus in the current account
during these years enabled the government to prepay a substantial
amount of its outstanding external debt. The gross debt:GNP ratio fell
from a highl of 52 percent in 1985 to 37 percent in 1987, and was as low
as 30 percent in 1988.
Gan and Krause (1990) uniderline the point that "the prepayment
exercise provides an example of the judicious use of reserves in time of a
.primary commodity boom. The reserve inflows during the 1987-88
commodity boom would have resulted in the temporary appreciation of
the ringgit above its long-run equilibrium value, thereby delaying the
adjustment process in transferring resources out of the nontrarlable to the
tradable sector" (p. 23).
Recent information released by the government shows the recovery
that started in 1989 has continued despite a softening of commodity
prices. At an estimated 7.6 percent, it is only slightly below the strong
growth of 1988 (Government of Malaysia Economic Report 1990, p. 17).
A major factor in the sustained growth is the continued erl.ansion of
manufactured exports that grew by more than 30 percent for the third
year in a row. Malaysia has been able to sustain its international
competitiveness and so achieve a high rate of growth without external
imbalance.
The Short-Run Problems of the Macroeconomy
In this section we turn to a detailed analysis of the short-term cyclical
problems of the economy.
The Determinants of Exchange Rate Movements
The short-run problems for the Malaysian economy generated by the
cycles just described can be understood in terms of the standard three-
Malaysia 471
good model of an open economy. The three goods are commodities
whose world prices fluctuate sharply; other tradables with -a more stable
price determined in the international market; and nontraded goods whose
prices are determined in the domestic market. When an upswing. occurs
in the terms of trade due to a rise in commodity prices, there is a net flow
of resources into the economy only a part of which is spent on tradables.
Depending on the proportion that is spent on nontradables, in a freely
floating exchange rate regime the currency will appreciate.
At this point we must distinguish between three concepts of the
exchange rate. The first is the nominal effective exchange rate (NEER),
which is the price of the currency in terms of some weighted average of
the currencies of the trading partners. The second is the real effective
exchaiige rate (REER), which corrects the nominal rate for differences in
inflation rates between the country and its trading partners. This rate is
particularly important to the economy's international competitiveness.
The third is the real domestic exchange rate (RDER), which is the ratio of
the price of nontradables to tradables in the domestic market. This
determines the relative profitability in the two sectors, and, therefore,
affects the supply function of exports. The REER and RDER will
generally move together, but there is no reason (except under very severe
assumptions) why the magnitude of the change will be the same.
The "Dutch disease" class of models has stressed the appreciation of
the exchange rate becaluse of a terms of trade improvement due to an
upswing of the export prices of key commodities. In the Malaysian case
the story is somewhat different. Figure 11.4 shows movements in the
terms of trade, the REER, and the RDER. Note that while the terms of
trade increased in 1975-79, both the REER and RDER declined
substantially, with the opposite situation prevailing in 1980-85. The
REER reached its highest point in 1984, but the RDER continued to rise
untii 1986. Then in 1987 and 1988, the RDER declined while the terms
of trade improved appreciably. Thus, contrary to the predictions of the
standard model, the exchange rate indices and the terms of trade are
inversely related in the Malaysian case.
The "spending effect" caused by the terms of trade movements has
been dampened by the behavior of private savings, which have moved
directly with the terms of trade, and overshadowed by the much stronger
countercyclical behavior of public expenditure. The sign and magnitude
472 DipakMazjntidur
Figure 11.4 Terms of Trade and Exchange Rate Variability, 1970-88
160 - _ ____.
150 - .
140 -1
130 5 *
~120-
~11o
. . -0
100
90
1970 1972 .1974 1976 1978 1980 1982 1984 1986 1988
Year
- Terms of Trade --- BEER M -- EER B e lIDER
Notes: The terms of trade are the export unit divided by the import unit in U.S. dollars (1980 - 100).
REER M is the; REER with multilateral trading partner weights. REER B is the REER with bilateral
trading partner weights. RDER is the service price index divided by the weighted manufactured
export price index (the service price index includes rent, domnestic services, transport, and
coMmunication).
Sources: Department of Statistics Monthly Statistics Bulletin (various issues); IMPFInternaoional
Financial Statistics (various issues); Direction o,f tIternaiaonal Trade (various issues); Krause (1989,
table 1); United Nations Yearbook of International Statistics (various years).
of the capital inflow, -generated by the resultant excess of spending over
incomes (whether positive or negative) has been a more dominant effect
on the exchange rate than the terms of trade, both in the upswing and the
downswing of the latter.
In the upswing of the 1970s, both government and private expenditure
increased, but kept pace with the increase in GDP. As already mentioned,
the increase 'in private savings helped to balance the deficit in the
Malaysia 473
government budget, so that the current account of the balance of
payments was in surplus during this period. This surplus would have put
ax upward pressure on the exchange rate if government or private savers
or both were not willing to hold foreign assets. As it happened, they did.
The net international resources of Bank Negara (valued in U.S. dollars)
increased at a substantial rate: the total in. 1980 was three times the value
in 1975.
As the terms of trade declined in the 1980s, the govemment attempt to
sustain a large countercyclical expenditure through massive foreign
borrowing led to a large inflow of capital. It was this inflow that led to an
appreciation of the Malaysian ringgit even though the terms of trade were
declining.6 In other words, financial flows in the capital account were the
dominant influence in the exchange rate rather than flows generated by
the current account of the balance of payments.
* Real Exchange Rate Movements and the Short-Run Crisis
As we have seen earlier, the government had to indulge in massive
external borrowing to finance its countercyclical budget deficit. This
borrowing becomes unsustainable if the attendant appreciation of the
REER and. RDER dampens the growth of exports so that the current
account deficits fuel the soaring debt:GDP ratio.
This section will show that the exchange rate appreciation was indeed
a factor in the crisis of the 1980s that led up to the recession, and that the
subsequent depreciation was a necessary condition for the recovery.
The performance of the manufacturing sector, especially its export
capability, has increasingly become importantt to Malaysia's growth. As
noted earlier, the composition of exports has S JA. I spectacularly, with
manufacturing climbing to nearly half the share of 2-J ol_exports by 1988.
There is prima facie evidence about the slowing down of the
manufactiuring sector during the real exchange rate appreciation of 1980-
84.7 Similarly, the recovery of 1987-88 was accompanied by a strong
6. This interpretation differs from that given in C;an and Krause (1990), which tells a
standard Dutch disease story. The govemment expenditure boom of the 1980s is best
viewed as a deliberate countercyclical policy rather than a lagged response to th.e terms of
trade increase.
7. The average annual growth in manufacturing output declined from 13 percent in.
1973-77 to 7.4 percent in 1980-84. The rate of growth of exports of manufactured goods
474 Dipak Mazumtidar
revival in the growth of exports from the manufacturing sectors aided by
exchange rate depreciation. We need to know; however, if these.
fluctuations were due to changing economic conditions in the world
market rather than movements in the exchange rate. Gan (1988)
estimated a reduced form export function as follows:
I 1:
logXt =fo+ lboi logREERt._ + XlogWyt-L + YUi logY.t-l
where WY - real GDP of all OECD countries, Y = domestic real GDP,
and X = quantum index of manufactured exports.
The equation was formulated in this way to allow for adjustment lags.
It was estimated with quarterly data from 1974:1 to 1985:4. The real
income of OECD countries was used because those countries accounted
for 70 percent of Malaysian manufactured exports in 1983 (Gan 1988,
table 5). OECD real income was clearly significant, along with REER
(with different lags), but not domestic real income. However, the long-
run elasticity of manufactured exports (the sum of lagged coefficients)
was much higher for REER (4.70 percent) than for WY (0.06 percent)
(Gan 1988, table 8).
Turning to the supply side of the market, considerable evidence
demonstrates the squeezing of profitability in the tradable goods sector
due to the appreciation of RIDER. The initial expansion of the early 1980s
was sustained not only by government spending on services, but also by a
construction boom in the private sector that it.triggered (World Bank
1989, vol. I, p. 9). Data gathered by CGan and Krause (1990) showed that
during the first half of the t980s, the tradable goods sector in Malaysia
(including manufacturing) suffered a steady decline in the pre-tax return
on equity and on fixed capital from 1980, while the rates of return in
construction and retailing were well above the levels of the 1970s (Gan
and Krause 1990, table 16).
declined from an annual rate of 56 percent during 1973-79 to 25 percent during 1980-84
(Gan and Kiause 1990, tables 17 and 18).
Malaysia 475
Thze Cycle and Labor Markets
So far the labor market's responses to the cyclical movements of the
economy have not been discussed. It is now necessary to integrate the
labor market story into the story of exchange rate movements.
RDER AND THE WAGE RAM In the simple model of the open economy, the,
product wage in the tradable sector will rise puri passu with the RDER.
Under the assumption of full employment, a rise in the price of
nontradables is assumed to be fully reflected in an increase in money
wages, which rise to the same extent in both sectors. However, with a
complex labor market like in Malaysia, wc have to examine wage
behavior in different segments of the market in a more detailed way.
Unfortuniately, wage series on a quarterly and/or annual basis are
available only for parts of the formal sector of the market, principally
manufacturing and plantations. This is, however, not entirely unhelpful,
because a substantial part of the tradable sector coincides with the formal
sector of the labor market (in both manufacturing and plantation
agriculture). Food, which is produced in the self-employed (or informal)
labor market is a nontradable for the purposes of this analysis. Some of
the cash crops that are exported, notably rubber, are, however, produced
in the inforrnal sector by smallholders.
Thus, we will concentrate here on wage inovements in manufacturing
and plantations to throw light on the factors that affect changing labor
costs in the short run.
MOVEMENTS IN CONSUMPrON (REAL) AND PRODUCT WAGES IN THE ForuAL SECOR.
The series for real wage movements 1968-88 are plotted in figurd 11.5
for four subsectors of the format labor market. The figure shows that in
the manufacturing sector, real wages have been increasing steadily since
the low points, reached in 1973, but the trend rate of growth, which
continued to 1985, has not changed very much since the late 1970s. The
absence of variations in real wages with the phases of the cycle (and GDP
growth) discussed earlier needs explanation and is discussed below, but
before discussing this topic an irmportant point about the behavior of the
product wage should be noted.
REAL WAGE AND PRODUCT WAGE. From the point of view of producers, the
ratio of wages to producer prices, (W/Pp) or the product wage, is the
-476 DipakMaztmdar
Figure 11.5 Real Annual Earnings by Industry, 1968-88
7 -
2- --- I- '| - - - - et *. ae .eat
'6
4 - ~ ~ ~ r
'I
19 68 1970 1972 1974 1976 1978 1980 1982 1984 1986
Year
- Manufacturing - -- Construction --Rubber - Palm oil
Source: Richardson and Soon (1990, table 5).
critical variable determining costs. The real wage is related to the product
wage through the identity
w W Pc
p Pc p
where Pc and Pp are indices of consumer and producer goods
respectively.
Malaysia 477
Now PO/PJ will be related to the RDER (PN/PT) in a way that depends
crucially on whether or not food is a tradable good. As mentioned earlier,
in Malaysia, because of the policy of maintaining rice prices, food is
virtually a nontradable. Thus, when PN/PT increased with real exchange
rate appreciation, PJPp would have increased with it. It follows that as
the real wage W/Pc increased during this period, the producer wage
increased even more, putting pressure on costs in the tradable sector.
Figure 11.6 plots the indices of product wages for manufacturing and
the two estate sectors, palm oil and rubber. This graph demonstrates
Figure 11.6 Product Wages, 1970-86
300 - - _
280-
260- ..
240 -
220-
200 -
u 180
160-
140-
120
100
so�
1970 1972 1974 1976 1978 1980 1982 1984 1986
Year
'--Ruzbber' -1,anufacturin-, Patm oil
Note: The product wage is the index of the uominal wapge rate divided by the index of product price.
a. Index of rubber estate wage rate divided by index uf Malaysian RSS1 rubber price index.
b. Index of manufacturing wages and salaries per paid employees divided by manufacturing
weighted import and export unit value index.
c. Index of palm oil nominal average earnings per employee divided by index of Malaysian RSS1
palm oil price per metric tonne index. From 1970-74, price used was ci.f. London $/tonne.
From 197-88, price used was f.o.b. Kuala Lumpur S/wonne.
Sources. Ministry of Finance, Economic Report (variois issues), Department of Statistics Irlustrial
Survvs Oil Palm Stadstics; Richardson and Soon (1990-).
478 Dipak Mazumdair
vividly the remarkable increase in the indices during 1980-f6 after
comparative stability in the 1970s. This increase was of a magnitude that
the trend rate of increase in productivity could not offset. Figure 11.7
illustrates the resulting increase in unit labor costs for manufacturing. The
increase in the real (consumption) wage, as well as the relative fall in the
price of tradables, contributed to this adverse shock.
DETERMINANTS OF REAL WA1E BEHAVIOR. What determines changes in real
wages in the Malaysian economy? The point made above that real wage
behavior seemed to bear little relation to the phases of the cycle suggests
Figure 11.7 Real Wage and Labor Productivity, Manufacturing, 1971-87
(2980 = 000)
180-
170
io - - -,, " .
150 -
140 -
130 -
120
90
170
90 -
6� - -I I I I I I I I' I - 1- '- 1 1 1
1971 1973 1975 1977 1979 1981 1983 1985 1987
Year
Labor Productivity Real Wage
Source: World Bank (1988, vol. 1, figure 2).
Malaysia 479
that wages did not respond significantly or quickly to market forces.
Figure 11.8 reinforces this point. Between 1969 and 1973, real wages in
manufacturing declined while employment, both in manufacturing and
the formal sector as a whole, increased at a significant rate every year.
Between 1973 and 1981, the relationship between employment'and real
wage growth was normal: both increasing. During these years the rate of
unemployment was also falling, but during 1981-85, real wages
continued to increase almost as fast as in the 1970s, while the rate of
unemployment increased every year, and the rate of growth of total
employment fell, and was, indeed, stagnant in the manufacturing sector.
Figure 11.8 Manufacturing: Real Wages and Empioyment, 1968-88
(1980 =100)
140 -
130-
120
110 . C
100
�80-
60- - -
50 - -h%
40
30 I 1 1 II I I I I I
1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988
Year
- Real Wage - - - - Employment In Manufacturing --Total Employed
Total Unemployed
Sources: Real wages: Richardson and Soon '1990); employmerm and unemployment:
Ministry of Finance Economic Report (variot s <rcars).
480 Dipak lazrtntdar
Richardson and Yin (1989) tried to fit an augmented Phillips curve to
nominal wage rate changes in manufacturing, using quarterly data, for
1976-87. The equation is of the form:
AWt = a + biAEt + 73ilt-1
where W is nominal earnings in mnanufacturing, E is the full-time
manufacturing wages, and I is the rate of inflation. All tlhe numbers are in
logarithms. The last term is an autoregressive proxy for the
(unobservable) rate of inflation expected by actors in the labor market.8
The results are given in table 11.2 for the period as a whole and the
two subperiods before and after 1980. The ernployment change variable
is negative in both periods, but significant only in the second subperiod.
This result underscores the puzzle already noted: even after allowing for
the effect of inflation expectations, the relationlship between wage
increase and employment increase was the opposite of what would be
expected from labor market conditions. The puzzle becomes deeper when
we look at the divergence in the, trend in wages in manufacturing and
construction from the trend in the earnings in the plantation sector after
1980. Figure 11.5 shows that during the boom of the late 1970s, wages in
both rubber and palm oil estates rose along with manufacturing wages.
The absolute gap in earnings in favor of manufacturing was reduced for
palm oil and remained the same for rubber, so that the relative gap was
squeezed as manufacturing wages were higher. However, after 1980,
while wages in the plantation sector were stagnant, manufacturing and
construction wages bounded ahead until 1985.
What explains this odd behavior of wages in manufacturing and
construction in Malaysia? Part of the explanation could indeed be
economic. The wage series available is of average annual earnings.
During the downswing, retrenchment of wc'rkers will affect those at the
8. Note that because of the absence of quarterly data on total formal sector
employment, employment in manufacturing is used. The graphs in figure 11.8 show that
the correspondence between the two series of annual data is quite close. The wage and
employment data are quarterly, run from 1976 to 1987, and are taken from Afonthly
Industrial Slatistics published by the Department of Statistics. The change in employment
variable, Et, was instrumented to remove a possible inconsistency being imparted from
the joint determination of wages and employment. The instruments used were lagged
changes in employment and inflation.
Malaysia 481
Table 11.2 Nominal Wage Changes in Malaysian Manufacturing,
1976-87
Period (qiarter) Nominal wage chaanges
1976(3)-1987(4) AWt 3.385-- 0.134AEt + 2ai1 i where. a; 1.369
(4.536) (-2.213) (8.233)
R2=0.689 SEE 2.404.
1976(3)-1979(4) AWt - 13.682 - 0.179AEt + ailti where aj -0.985
(4.789) (-1.242) (-1.877)
R-40.314 SEE = 1.732
1976(3)-1979(4) AWt 2.920 -.0.233AEt + I ailtlwhere, ai 1.497
(3.912) (-3.723) (-9.149)
R2=0.799 SEE 2338
Note: Figures in parentheses are t-values.
Sources: Richardson and Soon (1990, table 24).
bottom of the wage ladder proportionately 'More, and this in itself will
tend to push up average earnings. Lucas and Verry (1989) examined the
characteristics of a sample of workers who were retrenched at some point
between 1984 and 1988. Their results confirmed tihat
It is the young and the oldest, the. less well-educated employees from the
private sector and (to a weaker extent statistically) thome outside of unionized
plants, who are most likely to have been retrenchewL. . However, it was
certainly not the only factor, for we know that pay of given individuals
continued to rise also (table D.2 and p. 12).
We turn now to institutional factors that might have been important.
Unionism is not a very powerful -factor in the Malaysian labor market. In
manufacturing by 1985, less than a quarter of the workers had been
unionized, and in some subsectors, for example, electronics, unionism
was forbidden. Paradoxically; the plantation sector, which saw the
482 Dipak Mazurndar
stagnation of real wages in the 1980s, had the highest proportion of union
members. Observers also generally agree that collective bargaining has
traditionally been pursued most vigorously in the plantation sector.
We have seen that the public sector has played a dominant role in the
Malaysian labor market. Could it be that this sector-in which wages are
set adnministratively-played a wage leadership role in the 1980s? Lucas
and Verry (1989) cite evidence to show that for 1980 through 1987,
average wages in the public sector increased less rapidly than those for
manufacturing.
On these grounds, it would seem difficult to make a case that public service pay
has led private pay over the entire period, though in some intervals, such as.
1983-84, it may have done so (figure D.14 and p. 21).
The contractual forms of wage agreement-both formal and
informal-help explain the rising wages of the 1.980s. Two different
practices in Malaysian wage setting seem to be particularly relevant here.
First, many collective bargaining agreements provide for two- to three-
year coverage. Even when the plant is not unionized, formal sector
employers are keen to follow the going practice of wage setting. Clearly,
with contracts fixed for a long period of time, employers cannot cut the
wages of those employees who are not retrenched. Note in this
conmection that agreements in the plantation sector include provisions for
tying workers' wages to the prices of their products through complicated
formulas that ensure that, to some extent, wages iluctuate with product
prices when severe shocks occur. The stagnation of real wages in the
1980s in the plantations, after a period of rapid increase, may partly
reflect this effect of the agreement working itself out after the decline in,
commodity prices. However, in the manufacturing or construction
sectors, where the practice of tying wages to product market conditions
does not exict, long contracts mean a substantial lag before wages start to
adjust. (This type of delayed adjustment of wages to economic shocks is
not uncommon in much of Asia.) In fact, when the rate of inflation is
falling, as it did in Malaysia in the 1980s, the length of time that elapses
before real rather than money wages begin to fall may be considerable.
The second wage setting practice that is pertinent to the problenm of
wage flexibility in the Malaysian labor market is that of automatic
Malaysia 483
seniority increments. Malaysian employers, at least in the formal sector,
follow the Japanese system of granting pay increases based on years of
service in the firm (McCarthy 1988). While the system is expected to
increase productivity by securing workers' loyalty, it does not help the
rapid adjustment of wages to business conditions, particularly when
external shocks tend to be as large as they do for Malaysia. The Japanese
wage system provides a safeguard against the seniority system by using
bonuses, which are geared to the firm's profitability, as a large
component of workers' earnings. In Malaysia, such payments, including
fringe benefits, constituted 15 percent of total earnings for male workers
in 1984. This, however, was relatively low compared to the Republic of
Korea, where the share of bonuses and overtime payments in total
compensation in 1982 was as high as 30 percent, split evenly between the
two (World Bank 1989, p. 155).
The World Bank (1989) has pointed out that while wages of senior
workers are relatively rigid, reduction in entry level salaries is a major
element in the downward flexibility of wages, but this particular
mechanism could be working strongly only when employment starts to
recover and the opportunity to hire a significant number of new workers
arises. This is, indeed, what seems to have happened in 1986 and 1987
when the economy started to recover. The, wage for new entrants.had
started to fall in 1985 at- the bottom. of the depression, but average
payments to all employees continued to rise through 1986 (by 7.2 percent
that year), reflecting built-in escalators in old contracts, but in 1987
average eamings finally fell, coinciding with the upturn (World Bank
1989, vol. I, p. 26).
LABOR MARKETS AND CHANGES IN COMPErITIVENESS. Malaysia's problem of
adjustment in the 1980s stemmed from the same sources as those for
other developing countries, a fall in commodity prices, and as an oil
exporter, the fall in oil prices. Unlike many other countries, particularly
in Latin America, Mallaysia had not overborrowed during the upswing of
the late 1970s. To the contrary, its strong reserve positions enabled
Malaysia to go in for a strong center cyclical policy, with substantial
foreign borrowing, when the downturn in the external sector hit.
As an emerging exporter of manufactured goods, the country was
heavily dependent on maintaining its international competitiveness,
484 Dipak Mazmndar
especially when the price of commodities collapsed. One could argue that
if wages had shown more flexibility, particularly in the manufacturing
sector, when employment growth slowed dlown in the early 1980s, the
recession of the mid-1980s might have been less severe. As it was, the
economy nceded a sharp, albeit short recession, with a negative growth
rate, to avert the increase in wages, Note, howevar, that it is the dollar
cost of labor that is important for competitiveness, not just the-doniestic
unit cost of labor. Thus, an important part of the story is the behavior of
the exchange rate.
The World Bank (1989) report commented that other Asian countries
(Hong Kong, Korea, Singapore, Taiwan, and Thailand) also suffered
from the phenomenon of domestic wages rising faster than pmductivity.
However, because of differences in exchange rate policies in "Korea,
Thailand and Taiwan, unit labor costs in manufacturing denominated in
US dollars were roughly level with their 1980 values in 1984 and 1985.
In Hong Kong they were over 20 percent lower. Only in Singapore and
Malaysia is the trend sharply upwards, with a rise of 40-50 percent in
just four years" (World Bank 1989, vol. I, p. 1 and table 2.6).9
Another point is that labor costs are only part of total costs, with the
share of wages in value added in manufacturing being some 31 percent
during 1983-86 (Lucas and Verry 1989, p. 57). In the t980s, tight
monetary policy in Malaysia pushed up real borrowing costs: the real
prime lending rate rose from 5.9 percent in 1983 to 7.1 percent in 1984
and to 12.3 percent by 1985. At the same time firms in the tradable sector
had been severely rationed. as concerned credit as progressively more
credit was channeled to the construction sector (see Gan and Krause
1990, appendix. A, for further detalls).
Thus, the entire package of fiscal, monetary, and exchange rate
policies, acting together with labor market behavior, led to developments
that culminated in the deep recession of 1985-86. Similarly, the
simultaneous downward movement of interest rates, wage costs, and
exchange rates, along with favorable movements in the world market,
fueled the recovery.
9. The different exchange rate experiences in Korea and Malaysia arise because Korea
has a closed capital account and effective central bank control over the nominal parity of
tlie wan. With a floating exchange rate syitem like Malaysia, and the large inflow of
capital, the only way to keep down the dollar costs of labor would be to depress wages.
Malaysia 485
Compared to the real exchange rate of over 30. percent and the drop in
the base lending rate from a peak of 12.25 percent in 1984 to 7.0 percent
in 1988, the drop in average earnings per worker was, indeed, marginal.
The massive wage costs witnessed in Latin America did not occur, but
clearly, unit labor costs fell as productivity increased faster. Second, and.
more important, the glut in the labor market changed the cost calculations
for the employer as far as expanding employment was concerned. As a
local economist wrote late in 1987:
The burden of adjustment has fallen primarily on new entrants into the labor-
maiket or otherwise on those who have been unfortunate enough to lose their
jobs, either through business failures or retrenchments. This is readily apparent
from a comparison of salaries of existing and new employees. In fact, in the
case of graduates, the difference in starting salaries now and before can be as
much as 50 percent or even more. The difference is somewhat less in the case
of non-graduates.
The Long-Run Aspects of Adjustment and Labor Markets
We turn now to a discussion of selected aspects of the longer-run
problems of econromic adjustment and the labor market problems
associated with them.
Tife Initial Conditions and Objectives of Structural Adjustment
At the time the government initiated the new economic policy (NEP)
in 1971, Malaysia's economy was still heavily dependent on agriculture.
Over 50 percent of the labor force was engaged in agriculture, and
agriculture produced 30 percent of the GDP. The manufacturing sector,
though expanding, contributed only 13.4 percent of the GDP and
employed just over 10 percent of the labor force.
The incidence of poverty was high-officially 49.3 percent in 1970-
but in absolute terms the Malaysian poverty line might have been high by
international standards. How could an economy whose relatively high per
capita income put it in the middle income category and whose growth
rate in the 1960s was a respectable 6 percent per annum have, such a high
incidence of poverty? Apart from a high population growth rate of 3
percent per annum, observers, including official planners, agreed that the
economy contained large pockets of low-income populations. One such
pocket could have been in the small-scale tertiary sector activities, which
486 DfpakMazlmdar
were an important source of employment. Another such pocket existed in.
agriculture because of its dualistic development. A large proportion of the
work force was involved in cultivating smallholdings, either in paddy, in
cash crops, or often in both. The estates, principally growing rubber and
oil, palm, produced a substantial proportion of the cash crops with a high
land/labor ratio and employed hired labor. The problem of dualism was
exacerbated by the racial cnncentration in economic activity. Malays
dominated the low-income activities, such as paddy, smallholdings, and
services. The Chinese were found in the more dynamic industrial and
commercial sectors. Thus, the incidence of poverty was much higher
among the Malays.
The NEP's objective was to bring about a restructuring of the
economy with a view to reducing racial disparities in incomes, and, as a
by-product, reduce. the incidence of poverty, particularly; among the
Malays (see Government of Malaysia 1973, 1976). A selective list of the
major policy instruments follows:
* price maintenance (at a level higher than world prices) and
subsidies on inputs for paddy farmers;
* land development, which took the form of resettling Malay
families in newly cleared land at govemment expense and setting
them up as viable smallholders cultivating cash crops, particularly
palm oil;
* massive expansion in education, especially at the postprimary
level, accompanied by a rapid increase in employment in
government services, for which the Malays were given
preference;
* application of a racial quota on new employment in industry arid
commerce.
Clearly the NEP encouraged a massive restructuring of employment,
and it was sustained in this effort by the very satisfactory rate of growth
of the GDP maintained for the last two decades despite the fluctuations
analyzed earlier. We shall now examine the long-term results of this.
restructuring as it affected the behavior of labor markets and the pattern
of earnings.
Malaysia 487
TheAggregate Supply of aid Demand for Labor-
Table 11.3 summarizes the trends in the rates of growth of population,
the labor force, and employment relative to GDP growth during the last
three decades. The rate of growth of the labor force has been higher than
the rate of growth of population throughout, partly because of a shift in
the age structure toward the working age group, and partly because of an
increase in participation rates, especially by women. In the 1960s,
employment growth failed to keep up with the growth in the labor force.
so that unemployment increased in the latter half of the decade, peaking
at nearly 10 percent by the end of the decade. The labor market picture
was much better in the.1970s, when despite. an accelerated growth in the
labor force, the'employment growth rate stayed ahead, responding to the
higher GDP growth stimulated by the NEP in the first half of the decade
and by the commodity boom in the second. The unemployment rate fell 5
to 6 percent by the end of the decade. The figures in table 11.3 for the
1980s reflect the short-run crisis and adjustment already discussed. By
the second half of the 1980s, the employment rate was back to where it
was at the end of the 1960s. As already mentioned, the apparent lack of.
sensitivity: of the unemployment rate to the recovery of 1987-88 has
raised the specter of a more long-term structural problem of
unemployment in the labor market -
Table 11.3 Growth Rates of Population, the Labor Force, and. Employ-
ment Relative to GDP in Malaysia, 1961-85
(percent per annum)
Category 1961-70: 1971-75. 197680 1980-85
GDP 5.3 7.3 85 4.5
Population . 2.8 2.7 -2.6 2.6
Labor force 3.1 . 3.6f 325 .8
. Emnployment 2.8 . 4.6 3.7. 2.8
Note: For years prior to 1980, the growth rate of GDP relates to all Malaysia, but the other
statistics are for Peninsular Malaysia only.
Source: Until 1980, Wong (1983, table 10); 1980-85: Government of Malaysia data.:
488 DipakMazumdar
Trends in Employment and Earnings in the Formal andInformal
Sectors
Real average earnings in manufacturing, after growing rather slowly
in the first half of the 1970s, started to accelerate around 1977-78, and
went on increasing through the 1980s despite slackening employment
growth and rising unemployment. However, these wage data refer
entirely to the formal sector. How do these wage movements compare
with trends in earnings in the informal sector? We need to know if
economic growth in Malaysia has been accompanied by a widening gap
in earnings between the formal and the informal sector, thereby
accentuating the economric dualism that the NEP was expected to correct.
If the evidence suggests that no such disparity exists, this implies that the
accelerating wage increase in the formal sector since the late 1970s is
symptomatic of Malaysia's transition from a labor surplus to tighter labor
market conditions, which the short-run downturn in employment in the
1980s did not affect significantly.
DEFININIONS OF TE FORMAL AND INFoRMAL SECTORS. The informal sector is
generally defined to include the self-employed who are outside the wage
system, but should exclude those self-employed who work in professions
with entry restrictions such as lawyers, doctors, and so on (see Mazumdar
1989, chapter 3, for an extensive discussion of the problems of defining
and measuring the informal sector). Generally, a reasonable way to
exclude the professionals is to apply an educational cut-off to the self-
employed. A further problem is the distinction between self-employed
and own-account workers. In the terminology followed in this chapter,
the self-employed group includes employers, that is, the owners of small
businesses who might also work in their establishments. Their earnings
are a mixture of wages, profits, and rents. Own-account workers are paid
family or autonomous workers and have no wage eamers working for
them.
The informal sector should also include small firms employing wage
labor that are outside the legall and institutional framework covering
larger firms, and that use methods of structuring and deploying labor that
are less bureaucratic, and hence more susceptible to the free, play of
forces of supply and demand. Although arbitrary to some extent,
researchers could usefully differentiate between the small and large
Malaysia 489
sectors of a particular economy given, sufficient information.
Unfortunately, the statistical data collected in Malaysia do not permit this
separation for wage employment.
However, a large proportion of the wage earners in Malaysia's small-
scale sector are employed in the tertiary sector, which has expanded very
fast. Thus, this section will examine the trends in earnings differences
between the tertiary and other sectors to establish whether or not there is
* evidence of pockets of l6w-income labor developing in the tertiary sector.
- TRENDS IN THE DISTRIBUTOrl OF THE WORK FORCE BY MODE OF EMPLOYMENT. In
conunon with most other developing economies, wage employn-3nt in
Malaysia has been growing relative to the number of self-employed and
own-account workers. However, this growth seems to have taken place
only since 1975. Wong's data (Wong 1985, table 12) show that the
proportion of employees in total employment in Peninsular Malaysia was
about the same in 1975 as in 1957, hovering round 58 percent. Table 11.4
provides data from the labor force sirvey for more recent years. The table
shows that the wage workers' share of employment increased
significantly between 1975 and 1984, about 10 percent, but that the
recession checked this' trend. Between 1984 and 1987, the share of
employees in the total actually fell by about 3 percent.
When the information in table 11.4 is broken down by sector (not
shown), the resilience of the self-employed in the agricultural sector
stands out. Unlike in many developing econo'mies, economic growth has
Table 11A Distribution of Employed Labor Force in Peninsular Malaysia
by Employment Status, Selected Years
(percent)
Employment status 1975 1980 1984 1987
Employer 2.46 2.74 266 296
Employee . 57.98 64.66 , 68.57 65.63
Own-account worker 24.50 21.A8 19.79 20.72.
* Unpaid family worker 15.05 11.12 8.98 91068
Source: Malaysian labor force survey. -
490 DipakMazumdar
not led to a transformation of agriculture from a family-based to a
commercial wage economy. Although such a trend is apparent in
particular regions, notably the Muda region, the region is not large
enough to affect trends in Malaysiti's agricultural sector as a whole. Also
the major growth in agricultural output has come from the cash crop
sector, where the proportion of wage workers in the total employed is
much higher than in the food subsector. However, for several reasons,
including the land development efforts of agencies like FELDA, this
proportion has been declining in the last two decades at a significant rate
(see Wong 1985, table 13).
A second point of interest in the Malaysian case is that the proportion
of nonwage workers in manufacturing is relatively small compared to
many. other developing countries. Further, this proportion has been
declining over time:. when wage workers reached their peak in 1984, the
self-employed made up barely 12 percent of the manufacturing work
force. Thus, the data imply that the informal manufacturing sector is
relatively unimportant.
Outside agriculture, the self-employed are important only in
"distribution," but here too they were losing ground between 1975 and
1984. Nevertheless, they continued to account for just under half of the
work force in the distributive trades.
TRENDS IN THE DISTRIBUTION OF THE WORK FORCE BY INDUSrRY. Through the
1970s, the share of agriculture in total employment declined fairly
continuously. The decline was arrested in the more difficult years of the
1980s as far as the noncash crop sector was concerned, but continued at
much the same rate in the rubber, palm oil, and coconut subsectors. Both
manufacturing and the 'tertiary sectors absorbed labor released by
industry, but after 1980 the share of manufacturing stabilized, as did
employment in government services. The private tertiary sector thus grew
at the expense of the declining cash crop sector in the years leadinig to
and during the recession.
Table 11.5 gives a better idea of where the growing labor force went.
During the first period of rapid growth, 1970-80, agriculture created very
few additional jobs. The manufacturing and the tertiary sectors had to
absorb the bulk of the increase in the labor force. The data show that
manufacturing played a leading role in providing new jobs to the tune.
Malaysia 491
Table 15 Marginal Changes in Employment by Sector in Peninsular
Malaysia, Selected Periods
(percentage of total change)
Sector 1970-80 1980-87 198084 1984-87
1. Agriculture, hunting, foresty, fiEl.ing 6.2 -0.4 -19.9 32.1
2. Mining, quarrying -0.9 -i.4 -OA -3.1
3I Manufacturing 28.9 16.4 16.0 17.1
4. mlecuicdty, gas, water 1.6 -1.8 -3A 0.9
5. Construcdon 10.5 6.2 24.3 -24.1
6. Wholesale/retail trade, restaurants, hotels - 30.6 27.4 36.2
7. Financing, insurance, real estate, business > 35.1
services .J 7.1 5.2 10.2
B. Transport, storage, commuuication 5.1 3.9 5.7 0.8
9. Comnunity, social, and personal s6rvices 13.5 39.4 45.0 29.9
Total . 100.0 100.0 100.0 100.0
n.a. = not available
Note: The major discrepancy between 1970 census data and 1980 survey data is the
increase in the agricultural labor force. 'rhe comparison between 1975 and 1980 gives a.
slight dlcrease.
a. Sectors 6 and 7 combined.
Source; 1970 census; Malaysian labor surveys.
of nearly 30 percent of the total increase. Nearly half of the new labor
force was absorbed in the private tertiary sector, in which distributive and
business services dominated, but construction also played a role. The
category "community, social, and personal services" includes both public
(government) and privati, employment. It is sure that its importance in
incremental employment is relatively small.
Rather dramatic changes are seen in the 'second period, 1980-87,
which consists of two distinct subperiods in terms of Malaysia's
economic cycle. The first, 198044, was tbe period leading to the crisis in
which government spending slowed down, but the boom was kept going
by the upswing in the private sector, based largely on construction. The'
second subperiod, 1984-87, was the years of recession. Taking the two
492 Dipak Mazumndar
subperiods together, the leading role of manufacturing and government in
labor absorption falls significantly. Thus, thf private tertiary sector was
called upon to absorb no less than 72 percent of the increase in new
employment during this period as a whole, compared to less than 50
percent in the 1970s. A significant aspect of the.changed role of the
private tertiary* sector was the importance of the private services
subsector in the. 1980s. We have already drawn .attention to the
remarkable fact that private services actually lost labor absolutely in the
1970s. However, in both the 1980s subperiods this sector provided more
than a quarter of the new jobs.
The recession after 1984 did not help employment in manufacturing to
pick up much. The government continued to increase its work force, but
at an even lower rate than in the earlier subperiod. At the same time,
employment in construction fell precipitously.-The lack of employment
opportunities in the hitherto booming sectors led to an increase in the rate
of unemployment, but table: 11.5 reveals another striking aspect of
adjustment in the labor market. Employment in agriculture, which had
been decreasing at a slow rate in the 1970s and much faster during 1980-
84, reversed its trend rather dramaticaly. New employment in agriculture
in the recession years of 1984-87 was nearly a third of the total increase.
DIFFERENCES IN AVERAGE EARNINNGS BY INDUSTRY AND MODE OF EMPLOYMENT.
Table 11.6 gives the means and first quartile values of the earnings of
different groups of workers, with each value expressed relative to the
earnings of paddy own-account workers, which are set.at 100.
PADDYANDDOTHER SECTORS The first point that stands out from table 11.6 is'
the substantial increase in the difference between mean incomes of paddy
cultivators, on the one hand, and workers in most other sectors on the
other. However, while this is true of mean incomes at the different dates,
the first quartile value of earnings does not seem to have increased in
other sectors relative to paddy, with the exception of own-account
workers in smallholdings. What this means is that the earnings
distribution in the paddy sector in [973 was flatter. than in the other
sectors, so that the difference, in first quartile earings was larger than the
difference in means, but in the 1980s, the earnings distribution in paddy
became as skewed to the right as in the other sectors.
Malaysia 493
Table 11.6 Relative Mean and First Quartile Earnings by Usual Employ-
ment Status for Males in Peninsular Malaysia for 1973, 1984, and 1987
(paddy.= 100)
1973 1984 1987
First First First
Usual employment svats Mean quartile Mean quartile Mean quartile
Rural sector
Paddy, own-account 100 l00 100 100 100 100
Smaliholdings, own-account 87 98 168 149 141 155
Estates, employees 119 194 214 191 158 195
Production, employees 132 202 190 224 184 216
Production, own-account 114; 141 189 216 227 208
Sales, employees 174 304 208 281 226 271
Services, own-account 196 180 277 255 257 249
Urban sector .
. Production, employees 148 247 211 272 215 240
Production, own-account 147 .196 226 269 213 245
* Production, self-employed 181 206 270 287 258 277
Sales, employees 217 210 230 283 273 233
Sales, nwn-account 214 271 329 315 273 283
Sales. self-employed 311 314 400 359 320 311
Services, employees 186 239 319 280 241 290
Services, own-account 157 245 335 345 284 280
Services, self-employed 273 275 404 383 -370 335
Notes: The self-employed include cmployers and own-account workers.
Sources: The Malaysian 1987. household income survey and the 1973 household
expenditure survey.
The Malaysian paddy sector has traditionally been a pocket of low
incomes, and in the 1970s was identified as one of the target groups for
poverty reduction. Despite the price support policies mentioned earlier,
labor that has remained in this sector has continued to fall behind
494 DipakM'zzumdar
incomes in other sectors.10 This has occurred despite the fact that the
proportion of the labor force engaged in this sector was halved between
1970 and 1987.
PADDY AND THE CASH CROP SECroR. Another striking point is the relative
improvement in earnings in the cash crop subsector of Malaysian
agriculture relative to paddy, both for own-account smallholders and
employees on estates. Between 1973 and 1984, mean earnings for both
the latter groups increased substantially- relative to. those of paddy
farmers. They fell back somewhat during the downswing of the 1980s,
but in 1987 were well above the relative levels of 1973. The point made
earlier about the differentials in first quartile earnings holds for estate
employees as compared to paddy farmers, but not for smallholders..
Evidently, productivity in the cash crop sector responded more positively
to the packagt. of policies to help this sector (including land
development) than those aimed at paddy farmers.
THEAGAIcuL TumL-NONAGPJcuLTuRAL DWFFERENCvz Table 11.6 suggests that the
strong economic growth between 1973 and 1984 reduced the differential
between mean earnings in nonagriculture and the earnings of both
smallholders and estate laborers. However, the impact of the downswing
on earnings of the cash crop sector in agriculture was larger than that of
the nonagricultural sector, so that in 1987, earnings differentials between
the two were back to the same level as in 1973.
THE DIFFERENCE N EARNINCS BETWEEN THE TERTIARY AND PRODUCTION SECTORS IN
NOIAOPJCULTURE. As we have seen, Malaysia has depended a great deal on
the tertiary sector to absorb its growing Jabor force. In the 1970s,
government services played a significant role in providing new jobs, but
in the 1980s, the burden shifted more to the private tertiary sector. In
1987, the tertiary sector as a whole-public and private together-
accounted for one-half of total employment:in Malaysia.
An important question to ask is whether this shift to a service
economy in Malaysia. has meant that large numbers of people are
10. Bhalla's data (Bhalla 1989, table 6.4) suggest that the incomes of paddy farmers
have improved in absolute terms, but the incidence of absolute poverty in 1987 was still
high at 57 percent, having fallen from 88 percent in 1970.
Malaysia 495
employed at relatively low incomes in the nonproduction sectors. The
figures of earnings for male workers given in table 11.6 do not suggest
that this has been the case. There is no evidence of any significant trend
in the differential in the 15-year period to 1987. In the service sector
evarnings seem to have. increased relative to production during the
upswing of the early 1980s and fallen during the downswing ending in.
1987; but even at the end of the period,. mean earnings in services, as in
sales, wVere significantly higher than. in the corresponding groups in
produiction compared to 1973.
Some economists argue that the earnings distribution in the tertiary
sector is markedly skewed to the right, so the high earnings of a minority
pull up the inean. Table 11.6 includes the first quartile differentials to
take care of this point. We see; that generally for all the years the
difference in first quartile earnings between tertiary sector workers and
production workers is smaller than the difference in means. This supports
the hypothesis that the earnings distribution in the tertiary sectors is more
skewed to the right than that in production. But, with the possible
exception of employees in the sales sector, the value of first quartile
earnings in the tertiary groups was. higher than in production, and the
differentials in first quartile earnings, if anything,.increased over time in
favor of the tertiary groups.
The skill level of workers will affect their earnings, thus ideally, a
comparison of earnings across occupational groups should control for
skill differences. One way to deal with this problem is to use the standard
human capital model of earnings, which assumes that the major
determinants of earnings are education and experience. Anand (1983)
estimated a large number of earnings.functions for different groups of
urban. employees in a large data set.(Post-Enumeration Survey) for 1970
(table; 11.7). XVe have estimated similar functions from the household
income survey data sets of 1984 and 1987. The analysis is confined to
male workers.
Table 11.7 presents the results of the estimated equations for the three
sectors in which blue collar workers are found, namely, production, sales,
and services.11 The pattern of the coefficients of the equation are
11. Note that a large number of workers in government services are excluded as they
are likely to be in the clerical or administrative occupational categories.'
496 DipakMazumdar
Table 11,7 Coefficients of the Human Capital Model, Males, 1970, 1984,
1987
.~ ~~~~~~~~~~~~~~~~~~T R .
Yearloccupation Intercept S T AR
1970
Production 5.74 0.08 0.10 -0.0014 0.40
Services 5.35. 0.13 0.11 4.0014 0.32
Sales 5.21 0.12 0.12 -0.0017 0.47
1984
Production 7.19 0.07 0.09 -0.001 0.12
Services 6.97 0.09 0.09 -0.001 0.09.
Sales 6.36 0.15 0.11 -0.001 0.16
1987
Production 6.70 0.09 o0.10 -0.001 0.20
Services 6.58 0.12 0.10 -0.001 0.07
Sales 6.42 0.13 0.10 -0.001 0.21
S = years of schooling, T = years of experience
Note: All the variables are highly significant.
Sources: 1970:;Anand (1983); 1984 and 1987: estimated from HIS data.
strikingly similar in the three years, though the value of RB2 seems to have
been reduced drastically in the 1980s. The clear result is that although
eamings of raw labor (with no schooling and no experience) are lower in
sales and services than in production, the higher coefficients of
experience in the tertiary sectors ensure that the differential is reduced if
not eliminated, for some educationi and experience. This is seen in the
figures of predicted eamings given in table 11.8 for two types of labor.
Furthermore, the important conclusion emerges that there is no evidence
for earnings in the tertiary sectors to fall furtlher below the earnings in
production in later years as labor was reallocated more toward the sales
and services categories;
Malaysia 497
Table 11.8 Predicted Relative Earnings for Male Urban Employees in
1970, 1984, 1987
(production.workers = 100)
Labor wIt: S -0, T O Labor wit/t S *6, T 0
Occupatioln 1970 1984 1987 1970 1984 1987
Production 100 100 100 100 100 100
Services 68 80 89 92 91 106
Sales 59 44 76 76 86 117
Source: Calculated from the coefficients in table 11.7.
THE DIFFERENCE IN EARNINGS BETWEEN THE FoRMAL AND INFORMAL SECTORS. We
now compare labor in thie formal sector (defined as employees) with;
labor in the informal sector (the self-employed who include own-account
workers as well as employers).:
THE DIFERENcE jvAGE PRoFILEs. The literature of the developing countries
generally hypothesizes that the informal sector is the depository. of low-.
income workers, partly because it contains a disproportionate number of
nonprime age and poorly educated workers. In the Malaysian case we
have already noted that this stereotype does not fit the picture very well
as far as age is concerned. Blau (1986) produced longitudinal evidence
from the Malaysian family life -survey carried out in Peninsular Malaysia
in 1976-77 to show that for males, both in urban and rural areas, the
proportion working as. employees fell with age, and the proportion in
self-employment increased. The household income surveys do not
provide the longitudinal experience of workers, but we can use cross-
sectional data to throw some light on this issue. Figure 11.9 gives the age
distributions of -the:three categories of workers, employees, self-
employed, and family workers, by sex and rural-urban location. Blau's
(1986) point is confirmed for the 1980s. The self-employed are-
498 Dipak Mazumndar
Figure 11.9 Distribution of Workers by Age Group in Different Modes
of Employment, 1984
100-- - (a) Urban Males
100 ,
.90
80 - -
60 -
50 -
t 30 :
20
10-_, ,,.
0
15.19 25-29 35-39 40-49 55-59
Age group (years)
(b) Rural Males
100 -
950
740-
10-
t50 0 / --.... _
40 9 .-25-29 - 5- 39 4049 55-59
Age group (years)
-+ S.Ele.moe -. . .Wk
--Self Employed Emaployee F amlIly Worker:
Malaysia 499
Figure 11.9 (continued)
(c) Urban Females
100-
390- ;
780 --
67o -
50 -
40'
q30
- 20-
* 15-19 25-29 35.39 40-49 55-59
Age group (ye=r)
(d) Rural Females
100 1
90 -
80
60 -
50
40
3 70 - _
- Self Employed Employee . Fauy Worker
500 DipakMazumdar
contrary to the popular hypothesis-a rising proportion of the work force
at successively higher age groups.12
D,PrFF ENcEsiJNOreaCRc rERmC sncs. Table 11.9 shows the distribution of
employees and the self-employed by various characteristics for 1984 for
both sexes. Apart from the marked difference in age profiles already
Table 11.9 Proportions of Employees and Self-Employed in Various
Categories, 1984
(total for each category = 1.060)
Males : Females
0 - . Self- - . Self-
Category -Employees employed Employees Employed
Age
15-19 o.09 0.02 0.10 0.01
20-24 0.19 0.05 0.27 0.06
25-29 0.20 0.10 0.23 0.10
30-39 0.28 0.30 0.24 030
40-49 0.16 0.28 0.12 0.29
50+ 0.08 0.25 0.04 0.24
Education
No schooling 0.05 0.14 .0.09 037
Comple&ud primary 0.54 0.72 0.41 0.55
LCE or equivalent . 0.13 0.07 0.12 0.03
MCE or equivalent 0.20 0.06 0.30 0.04
HSC or equivalent 0.02 0.00 0.02 0.01
College diploma 0.02 0.00 0.04 0.00
University degree 0.04 0.01 0.02 0.00
Race
Malay 0.44 0.46 0.40 0.66
Chinese 0.39 0.45 0.43 0.37
Indian 0.17 0.09 0.17 0.07
Sector
Rural 030 0.48 0.25 0.49
Urban 0.70 0.52 0.75 0.51
Note: LCE, MCE, and HSC represent the certificates of completion of lower, middle, and
high secondary schools, respectively.
12. Note, however, that this cross-sectional picture may be due partly to the shift from
employee to self-employed status with age, and partly because the number of employees
has been expanding faster than the number of self-employed over time.
Malaysia 50)
noted, very large differences are seen in educational attainment. Only a
very small proportion of the self-employed have art education beyond the
primary level, and of the females a significant proportion have no
schooling-in sharp contrast to the employees. Significant racial
difference exists for the females:.more of the self-employed are Malays.
As expected, while the majority of employees are in the urban sector, the
self-employed are equally divided between rural and urban areas.
EAvRNiGs We attempt to estimate earnings functions separately for the
self-employed and the employees for all three dates. When the whole
sample of individual observations is partitioned like this between two
groups, the OLS regression estimates of the determinants of earnings will
be biased..We examined estimates from a model that permits some
selectivity bias, The coeffi-cients of the regressors in the OLS estimates in
tables 11.10 and 11.11 are generally very different from those in the
models allowing for selectivity bias. But for our present purposes, we
want to look at what the pattern of eamings has been .for different groups
with different modes of employment rather than the expected level of
earnings if someone were to relocate from one mode to amother. The
relevant equations for this are the OLS equations of tables 11.10 and
*11.11.-
The OLS regressions are used to "predict" the earnings of employees
and the self-employed for 1973, 1984, and 1987. Table 11.12 provides
three sets of estimates. The first line gives the employee/self-employed
earnings difference on the basis of the characteristics actually observed
for the two categories of labor at the relevant dates. The. next two. lines
give the hypothetical, earnings differences on the assumption that both
groups had identical. characteristics, first, those shared by employees, and
second, those shared by the self-employed. These hypothetical estimates
are meant to assess the extent of the earnings difference due to the two
groups' different characteristics. The table shows that in most cases the
differential is drastically. reduced, and sometimes even reversed in favor
of the self-employed, when we control for the .difference in
characteristics.
The differential in. favor of the employees is generally much larger for
the females. The important point to note for our present purposes is that
the trends in the differentials over time are different for the two sexes. As
502 DipakAMazumdar
Table 11.10 Regression Analysis of Male IndWvidual- Earnings:
Employees and Self-Empinyed Workers, 1973, 1984, 1987
(dependent variable: logarithm of total earnings)
1973 1984 1987
* Characteristic Self Self- Self-
Characteriasic Employee employed Emnployee emplayed Employee employed
Intercept 3-515 2.932 7.175 7391 7.046 6.945
Age (base= 15-19)
20-24 03526 0.668 0.545 0X033 0.578 0-568.
2S 29 - 0.876 1.224 0.891 0362 1.033 C.952
30-39 :1151 1.376 1.180 0608 1.324 1.139
40-49 1328 1.531 1.286 0.676 1.445 1.248
50+ 1.023 i.233 1.151 0.564 1.368 1.084
Education
(base = no schooling)
Completed primary 0.489 0.378 0.277 0357 0.283 0.249
LCE or equivalent 0.857 0.726 0.483 0568 0.515 0.451
MCE or equivalent 1.354 0.876 0.791 0.859 0.771 0.602
HSC or equivalent 1.606 0.897 1.095 1380 0.966 0.710
College diploma 1.857 0.898 1.229 1.145 1.153 0.898
Universily degree 2.387 2.112 1.728 1.736 1.705 1.409
Race (base = Malay)
Chinese 0.188 0.807 0.249 0.547 0.183 0.605
Indian 0.065 0.292 0.025 0.158 0.013 0.257
Urban (base = rral) 0.183 0.243 0.164 0342 0.140 0.137
Number of observations 4,703 2,769 9,616 3,174 9,300 3,535
R2 0.456 0.307 0.477 0.306 0.472 0.293
Notes: All the coefficients are highly significant.
Sources: Malaysian expenditure survey 1973; Malaysian household survey 1984 and
1987.
far as male workers are concerned, the self-em.ployed increased their
earnings relative to employees not only in the boom period ending in
1984, but probably also during-the recession of 1984-87. This result is.
consistent with our earlier finding that male employees in the tertiary
sector did not lag behind in earnings relative to earnings in the production
sector.
Malaysia 503
Table 11.11 Regression Analysis of Female Individual Earniiigs:
Employees and Self-Employed Workers, 1973, 1984, 1987
(dependent variable: logarithm of total earnings)
1973 1984 1987
Self- Self- Self-
Characteristic Employee cmployed Employee employd Ernployee employed.
Intercept 3.539 2.670 6.828 6.540 6.717 6.835
Age (base =15-19)
20-24 0.141 0.332 0.442 0.580 0.443 0.079
25-29 0383 U0378 0.735 0.665 0.787 0.158
30-39 0.530 0.780 0.854 .707 0.945 0.346
40-49 0540 1.086 0.870 0.811 1.005 0.664
50+ 0293 0.900 0.856 0.724 0.967 0.597
Education
(base = no schooling)
Completed primary 0.288 0.278 0392 0.108 0.422 0.041
LCE or equivalent 0.926 0.854 0.483 0.073 0.769 0.155
MCE or equivalent 1.469 0.813 1.020 0.682 1.056 0.600
HSC or equivalent . 1.621 --- 1.151 1.025- 1.202 0.845
College diploma 2.042 1.749 .1.586 - - - 1.599 -0.269
University degree 2.576 1.812 1.972 1.032 ..2009 1.719
Race (base =Malay)
Chinese 0.360 0.951 0.151 0.566 . 0.164 0.641
Indian 0.400 0.226 0.027 0.143 0.012 0.128
Urban (base = rural) 0.070 0.037 0.203 0.238 0.132 0.128
Number of obscrvations 2,275 1,076 4,907 941 5,020 1,042
R2 00392 0.225 0.393 0.134 0.373 0.151
Nate: All the coefficients are highly significant except those marked with dotted lines.
Sources: Malaysian expenditure survey 1973; Malaysian household survey 1984 and
1987.
504 Dipak Mazurndar
The behavior of the female labor market has been quiie different.
Even though employees earned considerably more than self-employed
wormen in 1973 (even though a substantial part of this difference could be
attributed to differences in characteristics), the differential increased
significantly during the boom ending in 1984. Thus; the popular
expectation that pockets of labor in the informal sector do not share in the
upsurge of earnings in the formal sector is borne out for the female labor
market. Equally interesting is the phenomenon revealed by the data in
table 11.12 that female employees lost relative to the self-employed
during the economy's downswing.
This cyclical behavior of the earnings differential may be due to
institutional factors operating in the wage labor market, but if it were so,
the institutional influences, working to establish a premium for
employees, seem to be highly responsive to economic conditions.
However, the observed cycle is consistent with a purely economic
hypothesis that the supply of female labor to the informal (self-
employed) sector is much more elastic than the supply to the formal
sector.
Table 11.12 Predicted Incomes for Males and Females, Employees
Relative to Self-Employed (=1)
Males Females
Caregory 1973 7984 1987 1973 1984 1987
With own characteristics 1.58 1.03 1.01 2.04 2.37 1.90
With characteristics of 1.11 1.00 0.82 1.11 1.63 1.28
employees
With characteristics of 1.12 0.89 1.07 1.20 1.48 1.08
self-employed
..Source: Calculated from estimated regression coefficients given in tables 11.10 and 11.11
and using the mean values of the explanatory variables in the equations for each year.
Malaysia 505
Educational Expartsion and Chlange in the Occupational Structure
As indicated earlier, the Malaysian authorities pushed forward with a
policy of educational expansion, With one of its objects being to reduce
the racial imbalance between Malays and Chinese in educational
attainment. By 1984 this had generally been achieved.
The expansion of education in Malaysia has been very rapid. Wong
(1985) Tioted that "the proportion of the labor force with no schooling
was reduced by two-thirds (from 43 percent in 1962 to 15 percent in
1979) while the proportion with secondary education and above tripled
(from 13 percent to 39 percent)." Note that Wong's figures on the
proportion with secondary education are clearly based on the years of
schooling recorded, and would thus include drop-outs from the secondary
schools.
The structure of demand for labor has changed over time to
accommodate the changing skill composition of the work force. As table
11.13 shows, the occupational distribution of the employed has changed
significantly over time, with the white collar proportion nearly tripling
between 1957 and 1984 until the recession stopped the'continuous
increase. The question arises whether this upgrading of the labor force
proceeded smoothly without creating adjustment problems in the labor
market.
Let us first consider the rate of return to education. The gross rates are
found from the coefficients of schooling in the basic human capital modei
reported. in table 11.14. As the table shows, despite the expansion in
education, the gross rates did not decline in the 1980s compared to 1970.
On the contrary, they seem to have increased somewhat,'particularly for
the Malays, and were higher in 1987 after the downswing than in 1984.
Remember, however, that the gross rates do not take account of
unemployment or of wages in the length of time taken to find the first
job. Nevertheless, the generally higher coefficients of schooling for those.
who were in wage employment in the 1980s, after more than two decades
of expansion in education, is an important point..
The simple human capital earnings function of the type represented in
table' 11.14 does not take account of another significant issue: differential
returns to education at different levels; Earnings functions were estimated
for male workers for 1984 and 1987 from the survey data of the
506 DipakMUazumdar
Table 11.13 Employment Trends in Peninsular Malaysia by Occupation
(percent)
Occupation 1957 1975 1980 1984 1987
Professional 3.10 5.52 6.85 7.58 7.88
Administrative 1.20. 1.30 1.90 2.19 2.10
-Clerical workers 2.90 7.10 8.46 10.21 -9.92
Sales workers 8.60 10.36 10.42 11.38 12.65
Service workers. 8.60 8.16 9.42 12.05 12.37
Agricultural warkers 5621 41.86 32.85 26.61 26.94
Production workers 18.90 25.69 30.10 29.99 28.15
Source: Malaysian labor force survey.
Table 11.14 Estimates of -the Basic Human Capital Model: Peninsular
Malaysia, Urban Males, All Occupations, 1970, 1984, and 1987
Years of
Race and year Constant schooling E 2 A2
Malays
1970 5.42 0.14 0.09 -0.001 0.45
1984 6.20 0.16 0.10 -0.001. 0.45
1987 . 5.93 0.17 0.11 -0.001 0.44
Chinese
1970 5.32 0.14 0.11 -0.001 0.52
1984 6.66 0.14 0.10 -0.001 0.62
1987 6.34 0.15 0.10 -0.001 OA44
All
1970 5.42 0.14 0.10 -0.001 0.49
1984 6.51 0.14- 0.10 -0.001 0.41
1987 6.21 0.16 0.10 -0.001 0.44
E - years of experience (age minus E'-
Sources: 1970: Anand (1983, table 7.1); 1984, 1987: estimated from the household
income surveys data.
Malaysia 507
household income surveys, which replicated as closely as possible the
functions Mazumdar estimated from the PES data for 1970 (Mazumdar
1981, table 8-2). The incremental rates of return for different levels of.
education are derived from the estimated equations by calculating the
differences in the coefficients of the dummies for successive levels of
education. They are set out for the years 1970, 1984, and 1987 in table
11.15. Three main points emerge.
1. A major change has occurred at the bottom end of the educational
spectrum, with the incremental return to some primary schooling
(relative to no schooling) falling drastically in the 1980s,
compared to 1970. At the other end of the, spectrum, the returns to
a high school certificate and tertiary education went up sharply'
even though the recession of the mid-1980s reduced the returns
somewhat between 1984 and 1987. This finding is consistent with
the nature of. the upswing, in which public sector job creation
played a leading role. Rates of return to completed lower and
middle secondary education fell over time, but only sligbtly.
Table 11.15 Incremental Returns to Education, Urban Male Emrployees
and Self-Employed, 1970,1984, and 1987
Education Level 1970 1984. 1987
SDme primary 033 0.16 0316
Completed primary 0.18 0.13 0.18
Forms I to III no certificate 0.W 0.04 0.11
LCE or equivalent 0.17. 0.13 0.10
Forms IV and , no certificate 0.14 0.00 0.02
MCE or equivalent 033 0.30 0.25
Form III, no certificate 0.18 -0.12 -.01
HSC or higher 0.44 0.76 0.58
Sources: 1970: calculated from Mazumdar (1981, table 8-2); 1984 and 1987: calculated
from estimated equations.
508 Dipak Mazu?ndar
2. Mazumdar (1981) observed that the 1970 data showed evidence of
strong increasing returns to education at levels higher than lower
secondary. This phenomenon was accentuated in the 1980s.
3. Mazumidar also drew attention to the importance of "credentialism"
in the labor market for the educated in Malaysia: at a particular
level, the incremental returns to having a certificate are much
higher. During 197047, this became more pronounced. Note the
striking case of those with a formn VI education having lowe.
earnings than the completed middle secondary certificate holders,
both in 1984 and 1987.
The last two points suggest the existence of significant elements of
"administered prices" in the Malaysian kIbr,r market. Imbalances between
the supply of and the demand for labor would, in this case, result in
quantity rather than price adjustments. Thus, the next section examines
unemployment.
Unemployment. Trends and Causes
In the 1960s and 1970s, Malaysia suffered from a serious
unemployment problem. A study by Mazumdar (1981) indicated that the
cause was a supply/demand imbalane. in the market for educated labor.
Between 1957 and 1967, the rate of growth of unemployment was several
times higher than the growth rate of the labor force with post-primary
education. The unemployment rate was highest for young first entrants
with 7 to 9 years (lower secondary) and 10 to 11 years (middle
secondary) of education (Mazumdar 1981, figure 14-1, table 144). The
imbalance was caused by a growth rate of secondary school leavers well
in excess of the growth rate of low-grade white collar occupations to
which they aspired, together with sluggish change in their occupational
preferences.
Although the Department of Statistics and the Economic Planning
Unit (using data from the Treasury) come up with different estimates of
the unemployment rate, the broad trends in the rate during the last two
decades are clear. Throughout the 1970s, the population grew at about
2.6 percent per annum. fb'he labor force grew at a much higher rate, partly
because a higher proportion of the population was in the working age
group, and partly because of higher participation rates by women.
Malaysia 509
Nevertheless, employment grew. at an even higher rate, estimated at 3.7
percent per annum. Unemployment fell from some 8 percent to 5 percent
through the decade.
As discussed earlier, the boom came to an end with the fall in
commodity prices in the early 1980s. Increased government deficit
spending and foreign borrowing sustained the GDP. growth rate for a
while, but employment growth decelerated significantly,.falling to at least'.
one-third of its 1970s level. The unemployment rate started to increase
after 1982, reaching the 1970 level by 1985. It continued to increase
through the depression year of 1986, and fell only slightly, if at all,
through the recovery years of 1987, 1988, and 1989.
UNEMPLOYMENT AND AGGREGATE DEMAND. During the upturn of the.
unemployment rate, Malaysian officials debated the desirability of re-
inflating the economy to solve the unemployment problem. However, the
consensus among both Malaysian policymakers and the economic staff of
international agencies like the World Bank was that a purely Keynesian.
approach to the unemployment problem was inappropriate.
Unemployment. could not. be related in a simple way to the level of
aggregate demand. Substantial evidence pointed to the reemergence of*
the problem of structural unemployment, which had occurred in the early
1970s.
Figure .11.10 suggested that the link between unemployment and
aggregate demand is tenuous. The figure shows no simple relationship
between the unemployment rate and the GDP growth rate. In particular,
unemployment began to rise well before the downturn in GDP growth in
1985, and increased significantly in 1983 and 1984, when the growth rate
of GDP also increased. More recently, the unemployment rate has shown
resistance to decline despite the recovery and the strong increase in GDP
of 5.2 percent in -1987 and 7.4 percent in 1988 (in real terms).
- The slackening of the labor market in several years of the 1980s even
though GDP growth was strong suggests that the structure of demand for
labor had altered in a way that was less favorable to absorption of labor.
The problems would generally arise both on the demand and supply side
of the labor market. The need for structural adjustment in a changed
external environment. alters the composition of demand. Because of
rigidities in the labor market, the supply of labor adjusts only slowly to
510 Dipak Mazjirndar
the shifting demand, so that unemployment increases rather more than is
warranted by the slowdown in GDP growth.
NATURE oFrEI UNEMPLOYED. Malaysia's unemployed are concentrated in
the 15 to 19 and 20 to 24 age groups, are mostly first time job seekers,
and a disproportionate number of them are secondary school leavers.
During the recession, the proportion of the unemployed who were older
than 25 increased significantly, from 23 percent in 1982 to 32 percent in.:
1986 (World Bank 1989, table 33).
Figure 11.10 Actual Unemployment Rates and Real GDP Growth Rates,
1967-87
10-
12-
196 199 197 197' 197 1977 197 19118.918
~~~ - ~~~ , '- I '-'L'l ,
Year
-Actual Uneniptoyment Rate Real GDP Growth Rate
SucsRerdcdfo S
Sources: Reproduced from S~ali anYog l8.
Malaisia 51S
Table 11.16 Total Employment by Age and Education for Males and
Femnales, Peninsular Malaysia, 1975 and 1987
Percentage of total cmployed
- .- Males FemrJales
Education
Age group (years) 1975 1987 )975 1987
15-19 0
.1-6 13.9 4.6 13.1 5.4
7-9 15,6 14 Z 14.9 13.9
10-11 11.8 10.0 15.3 16A4
12-13
14+
20-24 0 - - 1.8
1-6 8.2 4.6 9.8 3.1
7-9 . 9.9 11.2. 10.0 7.8
10-11 9.0 12.5 12.2 20.3
12-13 2.4 3. t 1.2 7.4
14+ - - 2.2 - 3.2
Total 70.6 62.7 78.3 77.5
Dashes indicale no observalion.
Souirce: Malaysian labor force survey.
Table 11.16 shows the distribution of the unemployed by age and
education for young job seekers for two years-1975 and 1987. Three
important changes are noticeable between the two dates:-
1. For both sexes and for both age groups the proportion of the
unemployed with primary education or less declined substantially.
This partly reflects the withdrawal of young people from the labor
force due to the spread of education.
2. The proportion of those with lower secondary education (seven to
nine years of schooling) declined slightly, particularly for females.
3. The decline in the proportion of those with a lower secondary
education is offset by a big jump in the percentage of the
unemployed with middle and higher secondary education and with
512 DIjak Muzumndar
post-secondarly education. This is true of both sexes in the 20 to
24 age group, but is more. marked for females. Females ages 20 to
24 anld with mtore thanl a lower secondary education now account
for nearly a tlhird of the total femnles unemployed,
TMI PRIoABnAILIrrY ail UNuMmLOYMENT: A MUrTiVARMAm ANALYSis, Table I 1 .17
shows the results of a profit analysis for the determinants of
unemployment for the 1987 sample. The rate of unemployment decreases
significantly for both males and feimnles for age groups older than 25.
The. coc;fficients are large, confirming -the importance of youth
unemployment.
The profit analysis brings out clearly the important po;nt that, holding
age and other factors constant, education is positively related to the rate
of unemployment, but that there is an important difference between males
and females. For males the rate of unemployment begins to increase only
with the middle secondary certificate, and increases strongly only for
upper secondary and college certificates. By contrast, the female
unemployment rate increases strongly and monotonically from the level
of lower secondary education and all the wa'y to college education.. The
(positive) coefficients' of the successive levels of education are large and
increasing, and are all significant. Nott also that the coefficients are
much larger for females. The important conclusion emerges that the
education-unemployment link is stronger and quantitatively much more
significant for females.'
Another important result is the highly significant negative coefficient
of YCAP (the household income per capita).. The small value of the
coefficient should not mislead, sirtce the variable is used as a continuous
one, and only shows the decrease in the rate of unemployment with each
ringgit increase in per capita income. The result has implications for the
hypothesis of "voluntary unemployment," which has been advanced by.
some commentators, recently by the World Bank (1989). The clearest test
of voluntary unemployment is that it is more important for higher income
groups who can afford to wait for the right job to,turn up. Our result
would seem to negate this hypothesis. While unemployment in Malaysia
increased with the level of education (niore strongly for females), holding
other factors constant, it decreases significantly as the income level of the
family goes up.
Malaysia 513
Table 11.17 Probability of Unemployment, 1987
Variable Males Females
Constant- 1-4.52 1-1.05
(-6.25) (-9.20)
Age (base = 1 -24)
25-34 -0.79 -0.80
(-15:9) (-12.06)
35-44 -0.97 -1.08
(-15.77) (-10.23)
45-54 -0.89 -1.00
(-12.20) (-7.29)
-55-64 -0.76 -0.89
(-7.62) (-4.91)
Education (base no certificate)
Primary -0.13 0.004
(-1.89) (0.04)
Lower secondary -0.35 0.2i
(-0-50) (2.03)
Middle secondary 0.08 0.37
(1.10). (3.63)
Upper secondary 0.27 0.63
(2.15) (4.56)
College 0.33 0.67
(2.89) (4.37)
Region (base = region 1)
Region 2 0.009 0.16
(0.20) (2.51)
Region 3 0.041 021
(0.69) (2.66)
Race (base = Malay)
Chinese 0.15 -0.17
(3.13) (-2.59).
Indian 0.08 -0.005
(1.26) (-0.06)
Sector (base = rural)
Urban 0.22 0.089
(4.95) (1.56)
YCAP (household income per capita) -0.00013 -0.00014
(-10.30) (-7.81)
514 DipakMazumdar
Unfortunately, this result is not as unambiguous a test of voluntary
unemployment as one would like it to be. One can always argue that the
income per capita of families from which the unemployed come was
relatively low because the unemployed were not voluntarily taking a job.
However, when we added the potential income of the unemployed to the
actual family income (using an earnings function and characteristics of
the unemployed), the sign of the income vanrable in the profit model was
reversed. Thus, we can. only conclude that the positive association.of
unemployment with actual household income per capita (which a
strong" version of the voluntary unemployment hypothesis would
suggest) is not observed.
As far as location of the unemployed is concerned, for both sexes the
rate of unemployment is higher in the urban areas. Another interesting
result is that,. contrary to the Harris-Todaro type of. hypothesis, the
incidence of unemployment is not higher in the high income region
(Region I). In-fact, the unemployment rate for females is significantly
higher in the poorer regions.
Apparently, despite the large-scale internal migration of females in
recent years, a large proportion of young, educated, female job seekers
continue to be "locked in" without employment in low-income labor
markets.
Finally, there is an in:.resting difference between males and females
as far as race is concerned. Other things being equal, the incidence of
* unemployment is higher among Chinese males (relative to Malays) but
lower among Chinese females. The locked in female unemployed in poor
regions could be expected to be disproportionately Malay.
CAUSES OF UNEMPWLOYMENT. The review of the characteristics of the
unemployed does not give.a precise answer to the question how far is,
unemployment a structural rather than a demand-related phenomenon?
The fact that unemployment increased during 1982-86 among the group
that typically has a relatively lower incidence of unemployment, that is,
males over 25, suggests that demand deflation had some part to play in
the emergence of the problem. Btut as mentioned earlier; aggregate
demand (represented by GDP growth) is not clearly related to the degrpe
of slack in the labor market.. Apolicy of demand expansion would not by
Malaysia 515
itself create a significant dent in the unemployment rate,, and may indeed
create problems of inflationary pressure.
The fact that secondary school leavers constitute the largest part of the
unemployed suggests a strdctural problem of absorbing the educated in
the employed labor force. The problem of jobs for the more educated had
clearly become worse in 1987, when we saw a significant increase in'the
proportion of uneinployed who were somewhat older (in the 20 to 24
group) and who had more than lower secondary education compared to
the 1970s.
:An examination of the occupational distribution of the unemployed
throws more light on the causes of unemployment. This information
* cannot be derived for the entire sample of the unemployed. A large
proportion of the unemployed were fresh job seekers. In any event, the
labor force survey did not record job seekers' previous occupation or
o-ccupational preferences. However, the Treasury's Annual Economic.
Report provides information about the registered unemployed by
occupational group (table 11.18).
The changing problem of. unemployment is revealed by the' marked
shift in the occupational, structure of those unemployed who were
registered. The slackening demand for labor in government services had a
major effect on the white collar labor market. The proportion of the
unemployed with unsatisfied demand for clerical jobs nearly doubled
compared to the. mid-1970s, and toward the . end of the depression. the
proportion aspiring to professional and technical jobs also increased
significantly.
The resultant unemployment is best viewed as a rationing problem
rather than a voluntary job search. Under conditions of an excess supply
of labor where job seekers have distinct occupational preferences, people
are absorbed into employment more slowly, but the data suggest that the
duration or unemployment for the majority has not become excessive
(less than six months). In this sense the basic unemployment problem that
re-emerged in the 1980s was no different from that observed in the late
1970s, except that the average unemployed person was more educated.
The structural problem in this. aspect of the unemployment story is a
combination of rigidity in -occupational preference and limited wage
flexibility. When supply runs ahead of demand, relative wages may fall
sufficiently to clear the market, and if this does not happen,' employers
516 DipakMazumdar
Table 11.18 The Unemployed by Occupational Group, Selected Years
(percent)
Production Professional(
Year workers Agriculture Services Clerical tecdnical Other-
1975 59.1 4.6 7.5 24.1 4.2 0.6
1983 37.6 1.9 6.2 47A 5.5 - 1.3
1986 35.7 1.6 5.2 42.8 12.0 2.7
1987 31.1 1.3 4.8 39.8 14.1 2.9
Source: Treasury Annual Economic Report (various years).
would tend to adjust by upgrading the educational requirements of the
labor force (bumping). Job seekers of a particular educational skill will
lower their expectations and accept jobs requiring lower skill levels.
However, the required change in occupational preference comes
gradually, and thie.speed with which the change occurs determines the
average period of unemployment that new entrants into the labor market
will experience.
Women in the Labor Market andAdjustment
A major feature in the long-term evolution of labor markets in Malaysia
has been the increase in participation rates of females. Overall male
participation rates have declined by a few percentage points because of
the fall in the rates of younger people working due to schooling and of
older people due to earlier retirement, but this decline has been more than
offset by the increase in female participation. Figures 11.11 and 11.12
show the participation rates by age group for the years 1970, 1980, and
1987 for rural and, urban areas. These figures reveal that the increase in
female participation rates during 1970-87 is much larger in the urban
than in the rural areas. Nevertheless, the basic difference in the
rural/urban patterns of female participation rates persists after 20 years of
change. In particular, the urban distribution is single-peaked, with the
highest participation rate reached at age 20 to 24. By contrast, the rural
distribution is currently double-peaked. In 1970, rural females had the
highest participation rate in the post-childbearing age group, 35 to 49.
Mataysa 517
Figure 11.11 Urban Female Participation Rates, 1970, 1980, 1987
70 -ls*'<
S"
40 -970 - --- 1980 -* l98
I-~~~~ .4
. . .~~~~~~~~~~~~~~~~~~~~~4
. .~~~~~~~4
20-
15-19 25-29 35-394095.9
Age grop (years)
- 1970 - -1980 - 1987
Developments since 1970 have added to the sharpness of this peak,
which is now found in the age group 40 to 49, but at the same time there
has been a remarkable increase in participation rates in the 20 to 24 age
group, resulting in two peaks.
Tables 11.19 and 11.20 show the relative importance of females in
total employment by industry and occupation for selected years. The
major shift in the employment pattem of females has been away from
agriculture.
518 Dipak Mazvaidar
Figure 11.12 Rural Female Participation Rates, 1970, 1980, 1987
58-
52B- - E | E-
. . . ~ ~~~~ ..
46-
340
15-19 25-29 35-39 40-49 55-59
Age group (yrear)
- 1970 -- -- 1980 - 1987.
Table 11.19 Percentage Share of Females in Total Employment by
Industry, Selected Years
industry 1975 1980 1984 1987
Agriculture 40.9 39.6 36.6 32.9
Mining 12.3 13.7 11.9 11.1
Manufacturing 39.3 40.7 42.9 - 46.1
Utilities 3.2 5.7 - . 4.8 2.9
Construction 6.4 . 5.7 5.5 4.7
Distribution 26.9 31.5 36.5 38.2
Transport 6.3 7.9 9.1 9.9
Services 37.9 34.1 36.0 39.3
Total. 34.5 33.5 34.2 35.4
Source: Malaysian labor force surveys.
. .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Malaysia 519
..Table 11.20 Percentage Share of Females in Total Employment, by
Occupation, Selected Years
Occupation 1975 1980 1984 1987
Professional 35.0 39.3 42.2 46.5.
Administrative 3.7 6.5 9.7 10.3
Clerical 35.8 44.2 49.7 51.6
Sales 18.6 28.4 31.2 33.8
Services 453 42.3 40.0 43.8
Agriculture 41.2 . 40.5 38.1 34.0
Production 243 24.3 23.8 27.0
Total 34.5 34.1 34.1 35.4
Source: Malaysian labor force surveys.
While the share of agriculture in total employment declined from 38
. to 22 percent between 1975 and 1987, the share of females employed in
this sector has also declined. Female workers have thus contributed
significan'tly to the growth in employment in the leading tertiary and
secondary sectors. By occupation, the professional, administrative,
clerical, and sales categories have seen the growth of the female share of
employment. Trends in production activities have held their share in the
period of fast growth in manufacturing. employment. The industrial
classification shows that the share of females in this sector is very high
and increasing slowly (table 11.19); much higher than in the Republic of
Korea, where the female share in manufacturing between 1974 and 1984
hovered around 35 to 37 percent (Grootaert 1987, table 11).
The aggregate figures do not reveal the full story of the role of female
labor in Malaysia's industrial growth. The electronics industry has
dominated the upsurge in industrialization in Malaysia in the last decade.
As in other Southeast Asian countries, transnational cooperations in this
industry were, lured into Malaysia-particularly into the free trade
zones-by the prospect of a plentiful supply of cheap female labor
possessing the manual dexterity and pliability needed. Research on the
520 DipakMazumdar
workers in this industry in the late 1970s and early 1980s brought out the
importance of female operatives and demonstrated the remarkable
uniformity in their characteristics. They were, by and large, unmarried
women aged 16 to 24 in their first' job, and what might be of some
surprise, mostly Malay girls from rural areas (see Young 1987 and the
many references cited therein).
We can conclude from this research that creating a newly emergent
class of industrial workers in Malaysia was not difficult. Despite the lack
of a tradition of industrial work, neither ethnic nor religious barriers
prevented young Malay girls from meeting the demands of the leading
industry. Note, however, that the state's mediating role was an important.
one. The objective of the new economic policy was to encourage greater
participation by Malays in. manufacturing activities. New ventures,
particularly transnational ones, cooperated in this process, but did not:
seem to have faced too many rigidities in tapping the desired source 'of
labor.
Table 11.21 shows the percentage of male and female workers found
in different industrial sectors by mode of employment. An interesting
point is that female workers in Malaysia are more likely than males to be
in wage employment rather than self-employed. This difference has
increased over time, and is. more pronounced in nonagriculture.
EARNINGS OF MALES AND FEMALES. To examine the relative earnings of
males and females over the last 20 years when the employment of women
expanded so much in the economy, we again depend .on the earnings
function analysis. The base estimates are available for 1970 from
Anand's (1983) work. We have estimated our own earnings functions
from the household income surveys data for 1984 ant 1987. These are
presented, along with Anand's equations, in table 11.22. The analysis is
confined to urban employees.
In 1970 returns to schooling were almost the same for males and
females, but returns to experience were higher for males, spectacularly so
for Chinese males. Over the years. the returns to both education and
experience have, increased absolutely for females, and relatively with
respect to the males, for both races. At the peak in 1984, retiurns to
schooling were higher for females (relative to the males) for both races,
more so for the Malays. Males and females had also reversed their
Malaysia 521
Table 11.21 Distribution of Employed Labor Force in Peninsular
Malaysia by Gender, Industry, and Employment Status
(percent)
1975 2980 1984 1987
Self Self Self Self
Gender/ikuslry Employed employed Employed empled Employed employed Emplayed -emaployed
Males
Agriculture 14.86 20.05 12,25 15.20 9.34 13.55 9.61 15.51
Mining 1.44 0.06 1.30 0.08 1.11 0.04 0.81 0.02
Manufacturing 11.78 1.83 14.18 2.05 13.89 1.37 13.55 1.17
Utilities 0.82 0.00 1.71 0.00 0.87 0.00 0.97 0.00
Construction 5.85 1.17 6.95 1.51 10.38 1.36 7.55 1.33
Distribution 7.41 722 9.18 7.28 9.77 7.17 10.20 7.69
Transport 4.78 1.59 5.04 1.53 5.23 1.39 4.95 1.44
Commerce 3.44 0.54 3.76 0.54 3.63 0.37 -4.01 0.58
Services 15.86 125 16.12 1.29 19.20 1.30 19.07 1.51
Total 66.24 33.71 70.49 29.48 73.42 26.55 70.72 29.25
Females
Agrculture 25.09 15.40 18.99 12.31 13.85 9.03 10.58 7.77
Mining 0.30 0.19 0.47 0.09 0.35 0.03 0.20 0.04
Manufacturing 16.81 5.06 22.09 3.75 21.63 4.33 22.52 4.71
Utilities 0.06 0.00 0.22 0.00 0.08 0.00 0.02 0.00
Construction 1.32 0.00 1.15 0.02 -1.54 0.00 0.94 0.01
Distribution 5.97 4.46 9.05 4.23 11.59 4.99 11.62 536
Transport 0.98 0.03 1.23 0.02 1.38 0.03 1.47 0.03
Commerce 3.32 0.08 3.85 0.06 4.87 0.05 ;5.28 0.11
Services 19.37 1.71 20.81 1.64 24.30 1.91 26.65 2.58
Total 73.22 26.93 77.86 22.12 79.59 20.37 79.28 20.67
Note: Unpaid family workers are excluded from this table.
Source: Malaysian labor force surveys.
relative positions on returns to experience as far as the Malays are
concerned. Chinese females had pu1led up their returns to experience
considerably, but not enough to close the gap with respect to male
employees.
522 DipakMazumdar
Table 11.22 Earnings Functions for Urban Males and Females by Race,
All Occupations Together, Selected Years
1970
Malay
lo"y(males)- 5A2 + 0.142S + 0.093T - R.0012T2 R2 = 0451
log y(females) = 5.20 + 0.147S + 0.071T 0.001IT R2 = 0.421
Chinese
log y(males) = 532 + 0.139S + 0.110T - 0.0014T2 R2 = 0.521
Io,y(females) = 5.46 + 0.133S + 0.068T - 0.0007T2 R2 = 0.437
1984
Malay
log y(nales) = 6.20 + 0.157S + 0.104T - 0.0014T2 R2 = 0.452
log y(remales) 5.73 + 0.173S + 0.124T 0.0022T2 R2 = OA12
Chinese
log ynmales) = 6.66 + 0.139S + 0.096T - 0.0012T2 R2 = 0410
log y(females) = 6.36 4- 0.148S + 0.077T -. D.0009T2 R2 = 0351
1987
Malay
, log y(males)= 5.94 + 0.171S + 0.111T - 0.0014T2 R2 = 0.439
lo, y(females)= 5.47 + 0.196S + 0.110T - 0.0016T2 R2 = 0.421
Chinese
log y(males) = 634 + 0.153S + 0.09ST - 0.0012T2 R2 = 0.437
log y(females) = 6.25 + 0.152S + 0.076T - 0.0009T2 R2 = 0326
S = number of years of formal schooling
T = number of years of labor force experience
Note: Years of labor market experience T, are assumed to be measured by age A, minus
schooling S, minus 5; that is, T = A - S - 5, where six is assumed to be the age at which
schooling starts.
Sources: Malaysian household survey for 1984 and 1987; Anand (1983, tables 7.1 and
*7.6).
In 1987, when the data reflect the effects of the recession of the mid-
1980s, some interesting differences in the overall trend emerge. Females
seemed to have lost ground somewhat with respect to the retums to both
education and experience. Returns to experience were now equal for
males and females as far as the Malays were concemed, and the higher
return to education for females no longer held for the Chinese. The
relative decline in returns to human capital factors for females are
consistent with labor market behavior in recessionis. When the demand
for labor slackens, labor with a relatively weak position in the market will
Malaysia 523
be reduced in numbers first. This means that not only workers with low
values of experience and education will lose their positions first, but
within these groups those with less ability or less attachment to the firms
(and therefore with lower earnings) will decline in numbers. Thus, the
differential in average earnings between workers with low and high
endowments of human capital is reduced. Evidently, this mechanism was
more pronounced for females than for m& M.s. This aspect could, of
course, be reinforced by a more discreet fall in thl wages of highly
skilled female labor relative to skilled male labor.
What is the evidence on the relative endowments of human capital
factors for the two sexes, rather than in the rates of return? Table 11.23
presents the relevant data. Rather surpnsingly, females had more years of
schooling for both races in 1970 and 1984, but the recession reversed this
situation. Thus, the numbers and wages of educated female workers fell
relative to those of males in the depression. -
The data on experience tells the well-known story that male
involvement in the labor force tends to be substantially longer than
female participation, but the trend is clear. For both races, the difference:
in the years of experience between males and females has been
substantially reduced over time. In this case, long-run trends in the labor
Table 11.23 Difference in Mean Years of Eduation and Labor Market
Experience for Urban Males and Females by Occupation and Race,
Selected Years
1970 1984 1987
Mean years Malay Chinese Malay Chinese Malay Chinese
S (-)0.465 (-)0.234 (-)0.514 (-)0.195 0.561 0.686
T 4.930 3.464 4.024 2.632 3.241 2.166.
S = number of years difference between males and females (M-E) of formal schooling
T = number of years difference between males and females (M-F) of labor force
experience
Note: Years of labor market experience T, are assumed to be measured by age A, minus
schooling S, minus 5; that is T = A - S - 5, where six is assumed to be the age at which
schooling starts.-
Sources: Malaysian household survey for 1984 and 1987; Anand (1983, tables 7.1 and
7.6).
524. DipokMazumdar
market and cyclical factors have worked together. During the depression,
the relative decline in the numbers of less experienced workers would
affect females more. This will reinforce the long-run effects of
increasingly stronger attachment of females to the market.
The relative earnings of females are the product of their relative
endowments of human capital and the relative returns to these factors.
During 1970-84, the trends on both counts would imply. an increase in
the relative earnings of females. Durinng the recession, the two effects pull
in opposite directions, and the outcome is uncertain.
Table 11.24 gives the male-female earnings ratios for the two races,
both actual and what is predicted by the human capital equations (given
the mean values of education and experience for each.sex). As expected,
the trend factors during 1970-84 rnsult in an improvement of females'
relative earnings, and in a particularly striking way for the Malays.
During the recession of 1984-87, the trend improvement was arrested for
the Malays. Note also the large difference between the actual and
predicted eamings ratio for 1987; a difference that is not nearly so large
for the other years. Evidently, factors other than the human capital ones
were holding up females' relative earnings during the recession.
Table 11.24 Ratio of Earnings of Urban Females to Those of Males,
Employees, by Occupation and Race, Predicted and Actual, Selected
Years
Actual Predicted
Year Malay . Chinese Malay Chinese
1970 0.51 . 0.57 0.52 0.56
1984 0.68 0.60 0.71 0.57
1987 0.65 0.63 052 0.54
Note: Predicted values were determined by the equations in table 11.22 and by the mean
values of education and experience in table 11.23.
Sources: Malaysian lousehold survey for 1984 and 1987; Anand (1983, tables 7.1 and
7.6).
Malaysia 525
Regional Effects of Labor Market Adjustment
A very important issue for countries like Malaysia with a rapid growth
rate is whether different regions of the country have shared in the benefits
of growth. Economic growth, by its nature, is concentrated in particular
regions or sectors, but regions lagging in the process could still share in
the prosperity if internal migration of.labor is sufficiently large and
sufficiently responsive to income differentials.
In the economic development literature, the movement of labor from
rural to urban areas typically receives prime attention. In the
moderiiization of economies, growth often occurs in the urban sector.
Income for workers generated in urban activities is generally much
higher than in the' rural traditional sector. The nature of urbanization
could, however, generate problems of unequal growth if, for example,
urban growth is concentrated in one or two very large cities, or. if the
rural-urban income difference widens significantly durin'g the economy's
development.
Malaysia has' been no' exception to the general experience of
* developing countries in having its urban population grow fast. Between
the census years of 1970 and 1980, the total urban population of
Peninsular Malaysia. grew. by 59 percent, nearly half of. which was
accounted for by net internal migration. The proportion of the urban'
population (living in towns of 10,000 people or more) increased from
28.7 to 37.2 percent (Hugo and others 1989 quoting Wee 1985). An
important aspect of the rural-urban migration is that the Malays seemed
to have participated increasingly in this process. This is in accordance
with one of the major objectives' of the new economic policy, namely,
"the elimination of the identification of race.with vocation as -well as
location" (Hugo and others 1989, p. 45).
Bhalla (1982, p. 25 and table 4.4) reports that trends in urban and rural
incomes are consistent with the general trend of improved income
distribution in Malaysia since 1970. Urban income per capita in 1973 was
twice the level of rural income, but at the peak of the boom in 1984, the
differential was' no higher. During the recession 'of :1984-87, urban
income per capita (in real terms) fell while rural income remained more
or less unchanged, leading to a fall in the differential by about 10 percent.
However, the emergence of a primary city 'is an important
phenomenon in Malaysia as in other Asian countries. The urban
526 DlpakMazumdar
concentration based on Kuala Lumpur and the surrounding areas in
Selangor during 1970-80 grew at almost twice the rate of the total urban
population. The problem of concentration is, of course, not as great as in
Thailand1, for example, where Bangkok is some 50 times larger than the
seconDc city, Chingmai, but a "major change is occurring in the urban
system towards a more primate city-size distribution" (HIugo and others
1989, pp. 46-47), as indicated by several indices of concentration in the
last two decades.
Related to some extent to the pattern of urbanization is the persistence
of interstate.differences in income. Internal migration flows in Malaysia
have been well documented in official census.volumes. A study of these
problems summarized by Hugo and others (1989) shows clearly that "the
states of Selangor and Patang-are the main centers of net irmmigration-
the former largely because of the attraction exerted by the major
metropolitan area ... and the latter which was the focus of r".-pansion of
rural settlement" (pp. 45, 55-57). The persistence of these states as
recipients of a net inflow- of migrants for three decades, with all other
states showing net outflow, is striking. The data actually suggest a
progressive exacerbation of an established pattern of movement despite
the emergence of Penang in the northwest as a new center of industrial
growth. Furthermore, the main origins of immigrants to the Selangor-
Kuala Lumpur urban complex are the more developed and urbanized
states of the west coast. This major migration stream would tend to
maintain, if not increase, interregional inequality. It is balanced to some
extent.by the migration.stream to the other major recipient. area, the
agricultural region of Panang, which attracts migrants mostly from the
least developed agricultural areas of the rast coast,.but also from the
agricultural areas of the west coast.
Table 11.25 reproduces Bhalla's (1989) data showing 'te mean
income per capita in 1976 and 1984 by individual states (urban and rural
combined), as well as the distribution of population by three regions
distinguished by income levels. The correlation in the ranking of the
states by income level is very high and is plotted in figure 13.13. The
simple regression equation is:
Malaysia 527
Table 11.25 Per Capita Monthly Income by State and Distribution of
Population between Regions, 1976 and 1984
Per capita monthly Percentage distribution
incom' 'inggits) ofpopulation
State 1976 1984 1976 1984
Region I (high income) 28.6 29.9
Kuala Lumpur . 214 430
Selangor 129 335
Pulau Pinang 103 255
Region II (middle income) 47.2. 45.4
Melaka 102 206
Johore 92 208
Perak 80 117
Pahang 96 228
Negri Sembilan 99 226
Region III (low income) 24.2 -24.7
Perlis 68 129
Kedah . 59 139.
Terenganu -75 153
Kelantan 58 . 121
Mean total 99 227 100 100
Source: Bhalla (1989, tables 4.5a and 4.5b).
Y1976.= 0.586 + 0.448Y1984
(Jt.065) (11.634) :
Adjusted R2: 0.924
where Y is per capita income, and t-statistics are in parenthesis.
In 1976, almost all the states had a mean per capita income of 45 percent
of that in 1984. We conclude that the high rate of growth during the
decade did not alter relative interstate differences in income in any
significant -way, but clearly absolute differences 'between per capita
income levels have increased markedly along with the increase in overall
income levels in the economy. Note finally- in table 11.25 that -the
proportion of population found in the three regions of high, middle, and
528 DipakMazumdar
Figure 11.13 Per Capita Increase in Income, States of Peninsular
Malaysia, 1976 and 1984
300 -
fi200 0
1100
100 200 300 . 400 500
Income in 1984 (Y84)
Note: The regression line uses Y84 as the independent variable to predict Y76 for each
state. The data in the scatter diagram show actual values of each state.
Source: Bhaulla (1989, table 45a).
low per capita income is almost exactly the same at the end of the period
as at the beginning.
Conclusion
This chapter has examined both the short-run problem's of adjustment
as they were affected by changes in labor costs, and the longer-run issues
of labor markets during the growth process.
The Short-Run Problems
Malaysia underwent some sharp fluctuations in economic activity in
the last two decades associated with the commodity and oil. price booms
Malaysia 529
and the subsequent fall in prices in the 1980s. Government expenditure
policies-particularly the attempt to pursue a countercyclical fiscal
policy to. keep the upswing going when commodity prices turned
downward in-'1980-was not entirely successful. The heavy foreign
borrowing needed to finance this policy led to a recession of some
severity. The behavior of wages in the tradable sectors-particularly
manufacturing-was also of a nature that led to Malaysia's loss' of
competitiveness in world markets and must have contributed to the sharp
recession.
The event in Malaysia differed in sorme important details from the
standard sequence described in the Dutch disease models. In particular,
the appreciation of the real exchange rate was not due to greater spending
induced by the upswing in commodity prices, but to the inflow of foreign
capital to support the government's budget deficit at a later period when
the terms of trade were falling. Similarly, the increase in average wages
during the period leading up to the recession was not corrected with the
rise in the domestic exchange rate (the ratio of the prices of nontradables
to tradables) in a fully employed economy. Wage increases in excess of
labor productivity increase occurred at a time when employment growth
had slowed down, and the rate of unemployment increased significantly.
Some labor market institutions common to most East Asian countries,
particularly the steep wage-seniority scales and the attachment of workers
to firms after a period of service, might have contributed to this perverse
behavior of average wages.-
However, rising labor costs were only part of the problem of rising
costs in Malaysia's tradable sector in the period leading .up to the
recession. The appreciation of the real exchange rate increased the dollar
cost of labor compared to the country's trading partners. A large increase
in interest costs contributed to the problem and was a direct consequence
of the monetary policies followed. Thus, the entire package of fiscal,
mnonetary, and exchange rate policies, acting together with labor market
behavior, led to developments culminating in the recession. However, the
recession was short-lived, lasting no more than two years. Factor markets
proved to be highly flexible downward, with wages, interest rates, and
exchange rates all drifting downward. This acollapse" of factor markets
was instrumental in fueling the recovery when favorable trends reasserted
themselves-in Malaysia's extemal markets.
530 DipakMazumdar
Lontg-Run Aspects
As the short-run problems of labor markets in Malaysia were of
limited importance, the longer-run aspects of adjustment in the labor
markets in response to rapid economic growth are probably of greater
interest.
The chapter showed that real wages in Malaysia's formal sector have
increased significantly in the last two decades. Plantations, particularly
rubber, lagged somewhat behind manufacturing, but the growth rate of
real wages was positive, at least before the slowdown of the 1980s. The
* question arises whether the nonformal sectors shared in the growth in
eamings?
.The availability of household surveys for 1973, 1984, and 1987, and
the post enumeration survey of 1970 provide some statistical information.
on this point, since earn'ings data for the nonformal sector are not
collected regularly. In the agricultural sector, paddy farmers and
smallholder cash crop growers had significantly lower earnings in 1973
not only than employees in the nonagricultural sector, both rural and
urban, but also relative to estate employees. Despite the policy of price
maintenance and subsidies the government pursued, paddy farmers have
not been able to improve their relative. earnings over time, and the
differential in eamings of paddy farmers and other workers actually
widened.significantly during the boom period. It narrowed somewhat in
the downswing of the 1980s, but in 1987 the differential was well above
that in 1973.
* The tertiary sector as a whole increased its share of total employment
from'36 percent in 1970 to 49 percent in 1980 and 55 percent in 1987..
Employment in government accounted for a growing part of the service
sector, at least during the 1970s, but even subtracting the share of
government, the private tertiary sector increased its share. The evidence.
suggests that at least as far as males are concerned, these new workers in
the tertiary sector are not a pocket of low-incbome labor who could not
break into the manufacturing sector.
Although Malaysia differs from many other developing countries in
that the self-employed in the manufacturing sector do not constitute a
very large proportion of the work force (around 15 percent), they
constitute rather more than a- third of the total in agriculture and the
Malaysia 531
distributive trades. The proportion of the self-employed in the economy
as a whole has been declining slowly. The evidence refutes the traditional
view of the self-employed, which holds that they enter young, then
"graduate" to the formal sector as employees. Both in rural and urban
areas and for both males and females, the proportion of the self-employed
increases with age.
An analysis of predicted earnings for the years 1973, 1984, and 1987.
showed that although employees earned more than.the self-employed in
all three years,. a great deal of the difference could be attributed to
differences in characteristics. Holding characteristics constant, the
differentials in favor of male employees were within 10 percentage
points, and in some cases the differential was actually reversed in favor
of the self-employed. Female earnings differences in favor of employees
were generally much higher than for males.
There were important differences between the trends in the male and
female labor markets. As far as male workers are concerned,.the self-
employed increased their earnings relative to employees not only in the
boom period, but probably also in the recession of 1984-87. The
behavior of the female labor market has been quite different. There was a
substantial increase in the differential in favor of employees in the boom
up to 1984, even though employees started with much higher relative
earnings (compared. to males) in 1973, but there was. a reduction in the,
differential in the downswing. Female labor fits the prediction of some
labor market models that the informal sector does not fully snare in the
upsurge of earnings in the formal sector in a boom.
Another aspect of structural transformation in the Malaysian labor.
market is the very large. educational upgrading of the labor force. The
evidence suggests that although the proportion of nonmanual jobs
increased significantly, the rate of increase of educated labor was higher
than the rate of growth of white.collar jobs. Thus, we see the usual
adjustment in the labor market, with a gradual movement of educated
labor to blue collar Jobs over time. The frictions involved in this process
of adjustment leads to a problem of unemployment of the more educated.
In Malaysia in the late 1960s, the high rate of unemployment (around 10
percent) was identified as being a problem of secondary school leavers.
This problem -was alleviated during the boom of the 1970s, but
apparently reemerged in the 1980s. This type of unemployment is of the
532 Dipak Mazunidar
structural kind, not responsive to changes in demand, unless, as in the
latter half of the 1970s, the boom is sustained and intense (also perhaps
favoring the white collar sector)..
Another important feature of Malaysian development has been the
increasing participation of women in the labor market, particularly of
Malay labor in urban activities. The analysis of the earnings of male and
female employees during 1970-87 showed that the trend factors in the
acquisition of human capital factors, as well as in the rates of return to
these factors, resulted in improved relative earnings by females, and in a
particularly striking way for Malays. During the recession of 1984-87,
the improvement trend was arrested, but there seems to have been an
increase over time in the importance of factors other than education and
experience in "explaining" the male-female differential.
Despite substantial internal movements of labor, the income per capita
of an individual state in 1976 could be exactly predicted by the per capita
income of that state in 1984: it was 44 percent of the 1984 level. The
almost bizarre constancy in the relative interstate differences in earnings
suggest a serious problem in the sharing of the fruits of economic growth,
through internal migration of the factors of production. With growth in
income over time of the magnitude that Malaysia has experienced,
constancy of relative differences produces rather large* widening of
absolute differences in income per capita. Moreover, the distribution of
population among the regions of high, medium, and low income per
capita seems to have remained unchanged.
References
Anand, Sudhir. 1983. Inequality and Poverty in Malaysia: Measurement
and Composition. New York: Oxford University Press.
Bank Negara. 1986. Quarterly Economic Bulletin (March/June).
Bhalla, Surjit. 1989. "Restructuring of the Malaysian Economy: An
Evaluation." Preliminary draft report prepared for the
UNDPJILO research project on the Malaysian Human Resources
Development Plan. Processed.
Blau, David M. 1986. "Self-Employment, Earnings, and Mobility in
Peninsular Malaysia." World Development 14(7): 839-852.
Malaysia 533
Gan, Wee Beng. 1988. "Industrialization and Manufacturing. Export,
Performance in Malaysia." Kuala Lumpur. Processed.
(Forthcoming in Brian Brogan, ed., Export Prernium as a Bonus
to Growth. Cambridge University Press.)
Gan, Wee Beng, and Lawrence B. Krause. 1990. "Issues of Macro
Adjustment Affecting Human Resource Development in.
Malaysia: Basis for a New Strategy." A report for the Malaysian
Huiman Resource Development Plan, Kuala Lumpur. Processed.
Government of Malaysia. 1973. Mid-Term Review of the Second
Malaysia Plan. Kuala Lumpur: Government Press.
. 1976. Third Malaysia Plan. 1976-80. Kuala Lumpur:
Government Press.
. 1986. Fifth Malaysia Plan. 1986-90. Kuala Lumpur:
Govemrment Press.
Grootaert, Christiaan. 1987. "The Labor Force Participation of Women
in the Republic of Korea: Evolution and Policy Issues." Report
No. IDP2. Washington, D.C.: World Bank..
Hugo, Sraeni, Lim Lean, Lean and Suresh Narayan. 1989. Labor
Mobility. Study No. 4, Module H.. Kuala ILumpur: Malaysian
Human Resource Development Planning Project. Processed.
Lucas, R., and D. Verry. 1989. "Human Resource Development
Project," 2 vols. Kuala Lumpur. Draft, processed.
Malaysia Ministry of Finance. Various years. Economic Report. Kuala
Lumpur..
Mazumdar, Dipak. 1981. The Urban Labor Market and Income
Distribution: A Study of Malaysia New York: Oxford University
Press.
-1989. Micro-Economic Issues of Labor Markets in
Developing Countries. EDI Seniinar Paper No. 40. Washington,
D.C.: World Bank.
McCarthy, Eugene. 1988. "4The Wage and Salary System in Malaysia."
Geneva: International Labour Organisation. Processed.
534 Dipak AMzaiendar
Richardson, R., and Lee Yin Soon. 1990. "Wage Trends and Structures
in Malaysia," Kuala Lumpur: Malaysian Human Resources
Development Plan Projects. Draft, processed.
Salih, Kamal, and Mei Ling Young. 1989. "Economic Recovery and
Employment Growth: Why is Unemployment so Persistent?"
Kuala Lumpur: Malaysia Institute of Economic Research.
Processed.
Wong, Po Kamn. 1985. "Economic Development and Labor Market
Changes in Peninsular Malaysia." Working Paper No. 12. Kuala
Lumpur Conference: Aseam-Australia Research Project.
World Bank. 1989. Malaysia: Matching Risks and Rewards in a Mixed
Econotny. Washington, D.C.
Young, *Mei Ling. 1987. "Women Workers in Malaysia." Discussion
paper. Malaysian Institute of Economic Research. Processed.
12
THE REPUBLIC OF KOREA
Dipak Mazumdar
The economy of the Republic of Korea provides a very interesting
case study from the point of view of both long-term and short-run
adjustments. Since 1965, the economy's growth rate has been very
high; and the economy has seen rapid and fundamental restructuring.
At the same time, as an open, export-oriented economy that had to
depend on the importation of oil and a wide array of intermediate
inputs, it has been fully exposed to the external shocks of oil price an.
interest rate hikes. 'rhe economy appears to be more vulnerable than
most. In a bid to keep up the rate of investment, Korea borrowed
heavily in the world market. It also has a sustained history of walking
a tightrope of inflationary pressures and balance of payments deficits.
Korea's ability to prevent the economy from going off the rails
during difficult periods is as remarkable as its achievement of high,
long-run rates of growth.
Cycles in the Korean Economy
Korea's recent economic history can be broken down into four
phases. The period 1965-73 was one of sustained growth in the GNP,
which although it varied from year to year, was at a generally high
level (figure 12.1).' Difficulties emerged after the first oil shock. It
led to a period when government economic policy leaned toward
fostering development in heavy industry so as to make the economy
less dependent on the vagaries of the world economy. This policy led
to a faster buildup of foreign debt, so that when the second oil price
1. The basic time series an which the graphs are based is given in table 12.A2 of
the annex.
535
536 DipakMazuindar
hike and interest rate hike struck, the economy went into a depression
in the early 1980s, the first time the average rate of growth of real
GNP fell below zero. However, the depression was extremely
shortlived. As in other Southeast Asian countries (with the exception
of the Philippines), the economy was able -to adjust very quickly to the
external shocks (which were aggravated by internal, shocks), and since
1982 the recovery has been rapice and sustained.
Phase 1: The Period of Export-Led Growtht (1965-73)
As figure 12.2 shows, during the period 1965-73 the barter terms.
of trade either increased or were constant (except for 1969), while the
Figure 12.1 Real GNP Growth and Current Account/GNP, 1963-88
25 -
20-
15 '
10
o~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1963 1969 1974 - 1979 19ff4 1988
- Year
Real GNP Growth- Current Acwunt: GNP
Thfe Repub/c of Korea 537
income terms of trade increased at a very high rate from year to year
(see annex table 12.A1 for the basic data). Throughout the period, the.
lowest annual rate of growth of the income terms of trade was 30
percent, and in most'years it was well above this. This was the period
when Korea's outward-looking strategy was getting established in a
spectacular way.
The current account was, however, in deficit throughout this period
(figure 12.1), and 'until 1971 the annual growth rate of the deficit
accelerated. It also went up sharply as a percentage of GNP from -3.7
percent in 1966-67 to -8.9 percent in 1971.
Figure 12.2 Growth of the Terms -of Trade; 196348
: 0
80-~
5 m -
so
70 ''
:10 a
*
'20
1'964 1969 - 1974 1L979 . -1984 1988
a - ' MYar
Bart--- Income
* - ~ ~ ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ .
538 DipakMazumdar
The reason for this deficit was the high rate of investment sustained
at a level higher than the domestic rate of savings. Foreign borrowing
was used to bridge the gap, as well as to take care of the diminishing
role of foreign aid. According to Collins and Park (1987, p. 6):
"Firms (especially exporters) were given strong incentives to borrow
abroad. A system of loan guarantees substantially reduced the risks
and the real cost of borrowing was negative." External debt as a
percentage of GNP grew to over 30 percent by the end of the period,
but because of the increase in the export: GNP ratio, the ratio of debt
to exports-which ultimately determined Korea's ability to finance
the debt-fell significantly toward the end of the period (figure 12.3)..
Figure 12.3 Korean Debt Ratios, 1970-88
-6o -
o ,J.
120 - , _ '_ -
U0 -
_0 ,' ' - . . .
140 - ' ,
.20~~~~II
120 - ', . 1979 - 1 1988
btIG .... xt
. X 80 - .: - - -
10 20-
'I~ ~ I - - 1 ,
690 .193 17 99 92 1B Y
Year
- Debt/GNP ---DebtEports
-im Republic of Korea 539.
The role of the public sector in maintaining the high rate of.
investment was limited. This, together with the virtual doubling of tax
revenue as a percentage of GNP (Dornbusch and Park 1987, figure 2,
p. 408), held the budget deficit at a relatively modest level. Except for
1972 when the deficit was 4.6 percent of GNP, the ratio was generally
2 percent or less.
Phase 2: Period of Directed Heavy Industrialization and the
Shadow of Crisis (1973-79)
Korea's difficulties in the 1970s started with the slowdown in the
world economy following the increase in the price of oil. As an oil
importer, Korea was hurt by the price hike itself The reduction in the
volume of exports aggravated the situation. As figure 12.2 shows, the
percentage change in the income terms of trade was negative for the
first time in 1974, and even when it recovered to positive levels it
was-with the exception of 1975-well below the levels reached in the
earlier period.
The government decided to counter the economic slowdown with a
big push in the investment program in the heavy and chemical
industries. Despite the fall in the domestic savings ratio. in the.
aftermath of the slowdown, Korea elected to borrow through the crisis
to keep -up its planned investment rate. In 1974 and. 1975, the
debt/GDP ratio and the budget deficit/GNP ratio reached their:highest
levels (although neither was excessive by, say, Latin American
standards). Government intervention in the form of greater direction:
of investment decisions increased, as did the chief instruments of
control: import restriction and credit rationing. In addition, the
exchange rate, which had been allowed to drift downward throughout
the previous period, was fixed during 1975-79, and the real exchange
rate was allowed to appreciate. While it helped importers of
intermediate goods and materials,. it clearly eroded Korea's
international competitiveness.
This phase of economic policy in Korea has been the subject of
controversy. The policy has been justified on the grounds that it laid
the basis for long-run diversification of the Korean economy (and its
540 DipakMazurndar
external trade) away from light industry. Although it mighlt have been
costly in terms of immediate reallocation, the policyl has been
commended for wisely anticipating long-rm-n changes in comparative
advantage.
In any event, the Korean economy recovered to some extent
following the recovery of the world economy in 1975 and 1976. It
also benefited from the export of skilled labor to the Middle East and
the subsequent flow of remittances. Nevertheless, the shadow of a crisis
that the events of this -period generated lingered, leading to the major
depression at the end of the decade.
Phase 3: The Crisis and Adjustment (1979-82)
The second oil price hike triggered a depression, with the barter
terms of trade registering negative percentage changes in 1979, 1980,
and 1981. The annual rate of change of the income terms of trade was
also negative in 1979 and 1980 (figure 12.1). The GDP growth rate
fell and for the first time was negative in 1980. As domestic savings
plunged, current account deficits mushroomed. The- government
resorted to external borrowing on a large scale. This was the period of
the most rapid accumulation of foreign debt in Korean history. The
debt/GNP ratio climbed from 32 to 53 percent during these years,
equaling that of some Latin American borrowers, such as Brazil.
Internal balance was also disrupted severely. with inflation rates
reaching levels well above those seen in the 1960s and 1970s, with the
exception of the years of the previous crisis, 197475 (figure 12.4).
Even while Korea stepped up its external borrowing to record levels.
in response to the crisis following the second oil shock, it had already
started to take steps to increase its competitiveness, particularly
through wage and eAichange rate policies. At the same time, the
government took further measures to adjust its fiscal, monetary, and
industrial policies. The package of policies immediately started to
restore internal and external balance. By the end of 1982, the rate of
inflation and the current account deficit had been reduced drastically.
Thte Republic of Korea 541
Figure 12.4 Rate of Urban and-Rural Inflation,-1968-88
35-
25 - , . - - - . - -
20 -
15~~~~~~~~~~~~~~~~~1i
25 I 3 p 1 -
. ,. .- *... * - . .
- . ,. - : ~~~~~~~I . I' . ' - '; -
.1966 -1968 1970 -1972 19l74 1976 1978 1980 1982. 1984 1986 1988
: ~~~Year-
20 . . .I I
- Urban Infation - - - - Run. Tn3adon-
- -- - . � ..-
Phase 4: Recovsery and Growth (1982-91):
.~~~~~~~~~~~~~~~~~~~~~~~~~
The- rate of gotofGPwsngatv only- in 1980,- but'it w'as
still low by Kore'an-standards in-1981 and 1982. The economic
measures takcen. in these year,s, however, pre-pared. Korea -for a strong
po sitive response when world trade rebounded in 1983-84. As figure
12.1 shows, the economy's growth rate increased substantially. This
also produced an increase in dom'estic savings,.which he'lped to.reduce
the, deficit ins the current account of Athe balance of payments. When a
*slowdown in the world economy thlreatened to produce another dip -in
"the growth rate in .1985, Korea countered by 'substantial real
depreciations of the exchange rate. Note, however', that devaluation-did
fuel. the- inflationary spira'l as i-app r-ntly did in 1979-80. On'the
contrary, the achievement of internal balance sustained the low rate of
'inflation attained at the, end -of 1982. .- . I .
542 DipakMuawidar
Since 1986, Korea has been experiencing a remarkable economic
boom, with an annual growth rate of 12 percent during 1986-88. As
in earlier periods, the boom has been fueled by a remarkable' rate of
growth of exports, which can, in turn, be traced to. a variety of external
factors, including the low and stable price of oil, the appreciation of
the yen, and the continued strong growth of the OECD economics.
The concern about foreign debt that had loomed large in the early
1980s has disappeared as the current account surpluses-generated by
the export boom have been used to prepay part of the liabilities.
Inflation had been kept under control until 1988, when it showed
signs of accelerating. The threat of incipient inflation is the product of
ne-w developments, particularly in the foreign exchange and labor
markets. The liquidity influx from the trade surplus .and capital
inflows threaten appreciation of the won beyond levels, that are
considered "safe" from the point of view of external'competitiveness.
At the same time, new developments in the labor market threaten to
create wage inflation of a kind not yet secn in Korea. Many-of these
new problems and concerns are outside the scope of this chapter, as it
is mainly concerned with Korea's success in adjustment policies after
the shocks of 1973-74 and 1979-80. We will, however, refer to the
labor market developments of the late 1980s insofar as it helps, explain
Korean wage movements during the earlier periods of adjustment.
The Characteristics of Korean Stabilization and
Adjustment Policies
We now know that the idea that Korea's development was fueled by
unregulated free markets is false. The government,- both during the
regime, of Park (who was assassinated in 19.79) and subsequently,.'
played a determined regulatory role. The package of policies involved
bringing about structural adjustments in the economy and pushing
through stabilization measures when the economy threatened to go off
the rails due to external and/or internal shocks.
The major strategy in Korea's industrialization has been the
* promotion of exports. Exports as a percentage of GNP rose from less
than 3 percent in the 1950s to 15 percent in 1969 and 35 percent in
the early 1980s. This, however, did not mean that the domestic market
was ignored, even for industres such as textiles, which were heavily
T.e IRepublic of Korea 543
involved in exports. Korea's tariff system was dualistic. Imported
intermediate inputs could be duty free, but industries targeted for
development were granted tariff protection. When the export growth
of textiles threatened to slow down, Korea embarked on. its "big'
push" policy after the first oil crisis with a shift, from light to heavy
industry.
The major instruments of targeted industrial development were
licensing and credit policy. The Economic Planning Board (EPB),
which was -responsible for.targeting, had control over licensing and
credit. If a proposal originated from the private sector, the EPB had to
approve it, and if it complemented the EPB's overall strategy, the
Ministry of Finance would arrange credit. lf, however, the government
* took the initiative, the EPB would typically find.a private firm to
undertake. the project rather than set up a public enterprise. As
Amsden (1987, p. 5) points out: "Government control 'of credit
differentiated Korean and Japanese development. The Japanese
zaibatsu owned their own banks whereas the Korean chaebof did not
. .--. . Direction' of the economy. was -more centralized because power
over the purse was more centralized.."
The state's.central role in credit for industry was possible'because
of financial repression. Although Korea has a less centrally controlled'.
nonbank financial sector as well as a curb market, the official banking
sector has been dominant, at least until the 1980s. Generally, deposit
rates were kept low, and.sometimes were even negative in real terms.
The implicit tax on depositors helped to channel resources into
investment in targeted areas, and to finance budget deficits in a
noninflationary way (for more details see Dornbusch and'Park 1987,
pp. 417-19, and the references cited therein).
Nowhere was control over financial flows more important than in
the external capital account. The government maintained tight control
ove.r foreigni borrowing. .Both short- and long-term borrowing
required government approval, "but the repayments of interest and
principal on loans (were) guaranteed by the banks owned or strictly-
controlled by the government or by the government itself' (Park in
Wong and Krause 1981, . p. 226). Park also makes the point that for
practical purposes "there is no point . in distinguishing private
borrowing from government indebtedness." The government used
544 DipakMazwnmdar
foreign borrowing for three purposes. First, it used foreign borrowing
to bridge the gap between domestic savings and investment, and thus.
to maintain a rate of investment higher than would. have been possible
from domestic savings alone. Second, foreign borrowing was used
along with, the control over domestic credit to support the priorities of
restructunrng the economy. Third, it was used to tide over balance of
payments difficulties originating from internal or external shocks.
Along with many other economies, especially, in Latin America,
Korea had a rising debt/GNP ratio:throughout the period 1965-82,
which increased strikingly in the years of crisis, 1974-75 and 1979-
80. However, Korea managed to avoid having the crises escalate into
prolonged difflculties that dampened long-run growth: in each of the
two cases of external shock that Korea experienced as an oil importer,
for the stabilization measures to succeed, the debt/GNP ratio to fall,
and sustained growth to resume took no more than two or three years.
The effective control over the external flow of funds clearly
helped. Unlike in many countries of Latin Amierica, capital flight did
not deepen. the crisis. The major difference with Latin America,
however, was Korea's.substantially lower debt/export ratio. In 1981-
-this was 76.6, while the major countries facing difficult problems in
the 1980s-Argentina, Brazil, Chile, and Mexico-had debt/export
ratios that were three to five times higher. Thus, in Korea during
1980-83, debt servicing was below the level of exports, but in the
Latin AAmerican countries it exceeded exports by anywhere between
30 and 100 percent (see Sachs 1985, p. 533, table 4, and pp. 53Z-35
for further discussion of the differences between Asia and Latin
America).
Maintaining export growth has thus been as important for Korea's
long-term economic development as it has for successful response to
the shocks. The factors affecting external competitiveness are
therefore of central importance in the analysis of Korea's policies of
adjustment and stabilization. The behavior of average wages,
particularly in the export-oriented manufacturing sector, together with
other factors affecting unit labor costs, are the relevant issues in this
connection.
The Republic oj'Korea 545
Determinants of Unit Labor Cost and Wage Behavior in
Korean Manufacturing
.This section focuses on manufacturing firms that employ ten or
more,workers. This is because detailed data are only available for this
sector; but.in any case, when it comes to the question of maintaining
competitiveness, this is the sector that needs to be singled out for
analysis as the exports of manufactured goods have driven Korean
growth.
When examining an. economy's external competitiveness, the key
statistic is .the unit labor cost of the exporting country in the
international market. The unit labor cost can be expressed as. shown in
equation 1.
Uc W/V 1/e (1)
where Uc = unit labor cost in dollars, W = wages per worker, V = value
added per worker, and e = the exchange rate (won per. dollar). The
three elements determining unit labor cost in world prices are (a) wage
behavior, (b) changes in labor productivity, and (c) the exchange rate.
Korea has always followed an ictive exchange,rate policy, together.
with the control over extemal capital flows -described earlier. As table
12.1 shows, the exchange rate depreciated continuously between 1968
and 1975. During the crisis periods of 1971-72 and 1975 the
depreciations were particularly large. The won was fixed to the dollar
between 1976 and 1979, but active devaluation of the currency
resumed following the second oil crisis. Another major devaluation
occurred in 1980 like after the first oil shock..
The. more or less continuous devaluation of the currency was
necessary because of the persistent double-digit inflation rates until
after 1982 (figure 12.4). As a result, there has been continuous
pressure for the real exchange' rate to increase, which had to be
countered by devaluation to maintain competitiveness.
-Devaluation has, however, not been always a successful way to
prevent the real exchange rate from increasing in open economies like
Korea, which have to import many materials and intermediate goods,
including oil. The higher.unit,cost of imports adds to the inflationa-ry
546 DipakMazumdar
Table 12.1 Annual Percentage Change in Unit. Labor Costs anid Its
Components, 1968-86
Wage-. Consumer- Nominal = () + f2 -(3).
productivity producerprice average Unit labor
Year gap differential exchange rate costs (USS)
- - (1) (2) - (3) (4)
1968 -6.14. . 3.88 2.27 -4.52
1969 -7.25 -3.69 4.16 -7.72
1970 -3.19 3.70 7.77 -7.27
1971 -16.80 8.65 . 11.78 -19.93
1972 3.63 -2.44 13.18 -11.98
1973 5.17 -8.78 1.38 -4.99
1974 9.47 -1.57 1.54 6.35
1975 -10.45 4.78 19.66 -25.34
1976 10.37 -a.95 0.00 -9.42
1977 7.43 -0.65 0.00 6.78
1978 3.59 1.27 0.00 4.86
1979 17.01 -1.00 0.00 16.00
1980 -9.34 1.99 25.50 -32.86
1981 -16.14 7.58 12.12 -20.68
1982 0.30 2.25 7.35 . -4.80
1983 -5.65 0.58 6.11 -11.18
1984 -1.59 1.56 3.90 -3.93
1985 . 2.97 0.05 7.95 -4.92
1986 -2.47 -1.39 1.31 -5.17
Averages
1967-73 -4.40 1.43 6.66 -9.63.
1973-79 6.58 0.29 . 3.30 3.57
1979-81 -12.60 4.90 18.62 . -26.32
1981-86 -1.21 0.60 5.28 -5.89
Sources: CPI and exchange rate: Bank of Korea (various years); deflatora IMF (various
years); wage bill and value added: U.N. (various years).
spiral. In the Korean case, devaluation could enhance the rate of
inflation through an additional route. Korean food policy has the dual
objective of supporting a high price for farmers and enabling
consumers to buy at a lower price (although still higher than world
.prices). The difference between the buying and selling prices creates a
T-he Republic of Korea 547.
deficit for the Grain Management Fund that is used to administer the
. policy. Apart from domestic procurement,. the government has had to
import a substantial amount of rice and barley to hold down selling
prices. Thus with devaluation, the Grain Management Fund's deficit
increases. Although food prices are not directly affected, the
. inflationa.ry impact of the devaluation through an increase in the fiscal
* deficit could be significant.
However, as figure 12.4 reveals, although Korea is walking an
* inflationary tightrope, it has never been faced with the problems of-
spiraling inflation. Inflation rates jumped to the rather high rates of 25
to 30 percent in both the periods of maxi-devaluation (associated with
the oil price shocks), but was brought down to moderate levels very
quickly, and rather spectacularly so in the 1980s. -The success story on
this point involves.two main policy. and economic responses. First, thb.
budget deficit (and the growth of money. supply) was controlled:
"The unified budget deficit, although swinging widely, never reached
5 percent of GNP and never stayed very high for more than two years
in a row.' (Dornbusch and Park 1987, p. 414). Second, a crucial issue
was the behavior of wages relative to labor productivity.
Determinants of Unit Labor Cost
We can use equation (1) to derive the following relationship (the.
dots represent proportionate rates of change):
UC- W-v - -el (w+ Pc)-(v + p )- e(w - v)+(Pc-Pp)- (2)
-~~~~~~~~~~C W. - . - + . c-P
- The additional variables are defined as follows: w = real wage (in
terms of consumer goods), v.= index of. physical productivity of labor,
P, = index of cost of living, Pp index of prices of manufactured
goods.
Equation (2) decomposes the percentage change in the unit labor
cost into three elements: the wage-productivity gap, the shift in the;
ratio of consumer to producer prices, and the change in the nominal
548 DipakMazumdar
exchange rate. Table 12.1 shows each factor's contribution to. the
change in unit labor cost (see annex table 12.A3 for the basic data).2
The following points in table 12.1 are worth emphasizing:
The continious depreciation of the exchange rate did not lead
to an increase in the price of tradables relative to the price of
nontradables (as approximated by the producers' price index
relative to the cost of living index).3 Thus, the domestic real
exchange rate generally moved against manufacturing and
increased, the unit labor cost in most years. This is because
devaluation did not fully compensate for inflation.
Nevertheless, it moderated the impact of inflation, and as the:
table shows, the magnitude of the upward pressure on unit
labor cost from this source was small.
* In the years of crisis and stabilization policies, large
devaluations and a substantial negative wage-productivity gap
helped to reduce unit labor costs. This happened in all three
stabilization periods: first, in 1971 when the government moved
to counteract a temporary slowdown-in exports; second, in
1975 following the first oil price shock; and third, during the
"comprehensive" stabilization plan of 1980-81.
* The average figures given for the three periods 1967-73,
1973-79, and 1980-86 show clearly the different trends in unit
labor costs associated with varying performances of the
economy. They also help us to quantify the relative importance
of the wage-productivity gap and the exchange rate movements
in accounting for movements in unit labor costs (in dollars).
During the first period of export expansion, unit labor costs
declined at a substantial annual rate of 9.6 percent per annum. The
depreciation of the nominal exchange rate contributed as much to the
decline of unit labor costs as the excess of productivity growth over
2. Note that the wage series is really one of average earnings per worker: the
annual wage bill divided by the number employed. The wage bill incltdes basic wages
as well as supplementary payments to labor.
3. Cereals, an important part of the CPI, although imported to some extent, are
really nontradables in Korea because of the administered price system orerated by the
Grain Management Fund. .
T/e Republic of Korea 549
wage growth, despite an adverse movement of the domestic real
exchange rate. The problem years after the first oil shock and the big
.push reversed the trend in unit labor cost. Its sharp increase in-the
period 1973-79 was largely due to the adverse wage-productivity gap.
Although the exchange rate was devalued sharply in 1975, Korea went
to a fixed rate for the rest of the 1970s. This policy was abandoned
following the. second. oil shock. The experience of the two years of
adjustment-1980 and 1981-shows the large contribution of.
devaluation, some 50 percent more than the negative wage-
productivity gap, to the reduction of unit labor costs. However, the
negative wage-productivity gap was substantial, so the unit labor cost
decline* was massive, offset only slightly by the increase in the
* domestic real exchange rate. The continued decline of unit labor costs.
until 1986-which was instrumental in the recovery-was again due
more to nominal devaluation than to the negative wage-productivity.
gap, although the latter contributed significantly to it.
-Wage-Produdtivity Trends
We conclude that the behavior of wages relative to productivity has
been of crucial importance .both. during the periods of Korean growth
and the short periods of stabilization. In developing countries with. a
large :farming sector, it is tempting to assume i la Lewis that the
negative wage-productivity gap is due to, an elastic supply of labor at a
constant real wage, while productivity growth in the modem sector is
significant due to exogenous technological progress, thereby leading
to a fall in unit labor costs over time. But the Korean story is different.
As figure 12.5 and annex table 12.A4 show, in most years during the
20-year period under consideraticn, real wage growth was more than 5
percent per annum. The exceptions were the. years of stabilization
policies:'1971-72, 1975, and 1980481. Of these, real wage'growth was
negative only in 1980-81, but fell by less than 5 percent.
The sustained and substantial rate of 'increase in labor productivity;
thus emerges as.a critical variable in the achievement of a continued
reduction in unit labor costs despite the continuous'devaluation of the
currency. The productivity growth was sufficient to counter the rising
import costs produced by the devaluation and to permit a significant
growth of. real wages. In the crisis years all or more' of the increase in
550 DipukMazumdar
Figure 12.5 Real Value Added and Wages, 196746
(annual figures per worker, 1980 = loo)
160 -
140-
120 - - .
o 100
60- -
" 0' S*-''.
90-
20 -1~~~~~~~~~~~~~~~~~~~~~~~~-
28 0 - - -I X -- X - I -- I I l
-1967 1969 1971 1973 1975 1977 1979 1981 1983 1985
Year
Value Added Wages
Sources: Bank of Korea (various years); U.N. (various years).
productivity went into reducing unit labor costs while real wage
growth was temporarily halted.
The importance of productivity growth for the economy's stability
is also rele"ant for another reason: preventing the emergency of
inflationary expectations. We have seen that until after the 1980-81
stabilization, the Korean inflation rate was at double-digit levels and
was very high in short bursts. Yet the economy never degenerated into.
a dangerous spiral of rising wages and prices. Stabilization efforts in
most countries generally need a period of stagnation, or even decline,
in real wages.- Often this wage effect is produced by an abnormal.
increase in the rate of inflation. Success depends on the subsequent
T te Republic of Korea 551
behavior of wages as the inflationary expectations of workers affect
them. In many developing countries, particularly in Latin America,
inflationary expectations have been explosive. Periods of real wage
stability or decline have been followed by spiraling increases in wages
and prices, leading to erosion of international competitiveness as
workers seek to defend their real wage unsuccessfully through
accelerating money wage increases. In an economy like Korea's,
productivity growth has sustained a significant rate of real wage
growth: over many years. Thus, the workers' confidence in the
economy's ability to improve their standard of living is coiitinuously
reinforced. It is easier for them to accept temporary real vtage restraint
(or even decline) without demanding money wage increases that feed
an explosive inflationary spiral.
While the investment rate in Korea was high, the sustained growth
in labor productivity was, to a large extent, due to the growth of total
factor productivity. Nishimuzu and Robinson (1984) showed that
during the 1960s and 1970s, Korea's total factor productivity growth
of 3.7 percent per annum was by far the .highest of the countries
investigated (1.3 percent per annum for Turkey, 0.5 percent for-
Yugoslavia, 2 percent for Japan). The increasiny efficiency in the use
of both capital and labor allowed Korea a safe margin for real wage
increases without eating into profitability.
Wage Determination in Korean Manufacturing
The increase in productivity made wage increases possible, but why
did this happen at the sustained rate that it did? What was the
mechanism of determining wages in industry?
Before the late 1980s, unions' influence on wage levels was
minimal: the right to strike was banned by presidential decree. in 1971.
Unions did exist in large firms, particularly in the textiles,
metalworking, and chemical sectors, but they needed prior permission
from the government for collective bargaining. Earnings function
studies have found no significant effect of unions on relative wages
(see, for example, Park 1980, part 3).
From time to time, the Federation of Korean Industries (FKI) and
the Federation of Korean Trade Unions specify wage guidelines for
both white and blue collar workers. The chaebol dominates the
552 Dipak Mazurndar
former, while the government's influence on the latter has been.
recognized for some time. The government's own influence was used
to support wage restraint, as during the stabilization period of 1980-
81, and to ensure that the workers received a share of productivity
gains in the years of sustained growth.4
. . .With or without government encouragement, Korean industry
showed strong predilections toward a profit-sharing system of
- remuneration. The basic wage constituted no more than 75 percent of
total monthly earnings in the early 1970s, and fell to 70 percent in the
1980s (Park and Castaneda 1987, table 17, p. 38). Overtime pay and
annual bonuses-both of which are related to business conditions and
profitability-constitute the rest.
The industrial firms, particularly the larger ones, seem to have
determined the structure of the internal ~abor market. Starting wages
are predominantly determined by a worker's formal schooling and
sex, regardless of job content (Park 1980, chapter V)..This basic wage
rises on an almost regular basis by certain fixed amounts, the so-called
'tannual base-up." This base-up. is directly related to the length of
service in the company, and is not inecessarily associated with
promotion. Promotion takes the form of skipping several. base-ups.
One econometric study (Amsden 1990, p. 88, quoting Lee 1983)
found that "in the case of male workers, one year, of 'inside'
experience (with the same employer) tended to raise wages on average
by about 10 percent, whereas one year of outside experience (with a
different employer) raised them by an average of only about 3.8
percent."
With a strong mechanism for rewarding firm-specific skills in place,
clearly a major incentive for efficiency would be to share the gains of
4. In 1988-89, the Korean government showed a new commitment to a less
iaterventionist policy toward labor markets. The impact on independent wage
bargaining was immediate. After two successive years of double-digit nominal wage
increases, the Federation of Korean Trade Unions was asking for a 27 percent wage
increase in the spring negotiations of 1989. The Federation of Korean Industries
counte.red with an offer of, an 8.9 to 12.9 percent increase depending on the sector.
The government suggested that nominal wage increases should be no higher than real
productivity gains, but as shown by continuing labor unrest, including large-scale
strikes, this informl incomes policy is experiencing implementation difficulties.
The Republic ofjXorea 553
productivity increases with the workers. An apt question is: what is the
* exact nature of firm-specific skills that are being rewarded?
Amsden (1987) makes the point, that Korea depended heavily on
* imported 'technology, and had little expeinc ith such technology
with the possible exception of textiles. According to Anmsden (1990, p
* ~~89): "Korean managers could never h'ope to manage in a tight,
'Taylorist' 'top-down fashion, at least not initially, bec'ause no one at
the top knew e nough about-the process. -(of production) to. do so.
Under these conditions,.it was imperativect rely upon, motivated
workers, even if they possessed little more than formnal schooling, to)
exercise. the most -fundamental skill of all-intelligence." This was
particularly 'so' because an export-oriented* strategy was quite
demanding on the m'aintenance of product* quality.
* A rofi-sharing Imodel of wage determination- could explain the
observed increase in -the real Wage at a rate a little below productivity
growth in the period before 1974 and again after the adjustment of
1980-81. The successful wage repression of 1971, 1975, and 1980-
8 1, whi ch contributed to the stabiliza'tion effort, haLs the hallma'rk of
state paternalism in wage setting.
*However, one might still ask why wages inc.reased significantly
faster than productivity during the big push of the second half of the
1970s. Part of the answer is probably the high optimism of -the state-
driven investmeints in diversification'. Another.'factor -was thie tightness
ofthe labor market caused not only by the big push, but also by the
rather sudden and substantial emigration of Kor'ean workers to the
Middle East to help in the latter's post-oil construction boom.
As annex table 12.A2 shows, the unemployment rate fell to a
historic low in. 1978 (note that the unemployment rates touched this'
low again in 1986, and fell even lower in 1987 and 1988). As already
*pointe out, the events of the last few -years have crea'ted a new
* ~sit'uation in the Korean laibor market. The 'wage explosion, which; is
still underway, is as much due to the~ tightn'ess of the labor market as to
the government's.less paternalistic role in wage determnination..*.
An attempt was made to test these points with *an econometric
model of wage determinatio'n. Our model was the usual augmented
Phillips curve, together- with an elemnent.- to' capture the, pfit-sharing
aspect.. It is hypothesized that workers. have a target -real wa'ge in any
554 DipakMazurndar
period thal is governed by the productivity growth of a previous
period. If the percentage increase in real wages falls short of the
percentage increase in productivity of the earlier period, then there is
additional upward pressure on money wages. Note that the mechanism
of the target wage could percolate through the decision of workers,
employers, or both. Thus, the model would be:
W-t a+b Pe + cU.t-x + d(V t- w -y) '(3
where W'I = the percentage change in money wage in the current
period, Pe = the expected rate of inflation, U = the unemployment
rate x periods before, v t-y = the percentage increase in productivity y
periods before, and w;t-y_ the percentage increase- in real wages y
periods before. The values of x and y are found by the best fit of the
model to the data.
The model was estimated with quarterly data for the period
1970.3-1988.3. The results are given in table 12.2. The expected:
inflation rate is approximated by the rate of increase in the CPI in the
previous period. In the first equation reported in table 12.2, we get a
reasonably good fit with all the variables having the right sign and
strong significance. The second equation increa.ses the R2
substantially without reducing the significance of the explanatory
variables, significantly. The extra term DNOMAW(-4) is the-
percentage change in money wages since four quarters before the.
present. The inclusion of the variable increases the R2 by so much
because there is a strong seasonal pattern in the money wage series, in.
particular, average earnings in the fourth quarter of each year are
bumped up as workers are paid their annual bonus.
The variables are defined under the table. The fitted equations
strongly support the hypothesis. Both the rate of unemployment and
the target real wage based on actual productivity increase enter the
process of wage determination.
T/je Republic of Korea 555
Table 12.2 Determinants of Percentagc Changes in Nominal Wages
1970-88, Quarterly Data Regression Analysis (OLS Estimates)
Durbis
DNOMA W Adjusted Watson
Constant DCPI(-). UEfR Target(-2) (-4) R-squared statis hc F-staristic
0.135 0.869 -0.028 0.214 - 0.519 2.24 .25.8
(5.870) (4.050) (-5.230). (4.510)
0.070 0.478 -0.016 0.097 0.59 0.695 2.24 40.3
(3.300) (2.620) (-3.390) (2.300) (6.25)
DCPI(-1) = iliflation rate lagged one quarter.
UER = unemployment rate.
Target (-2) = difference between growth in productivity and growth in real
wages. The variable is lagged two quarters.
DNOMAW(-4) = the dependent variable lagged four quarters.
DNONiAW - percentage change in nominal monthly earnings per regular
employee in manufacturing, averaged for each quarter.
Notes: The variable TARGET (-2) can be broken down into rates of growth of money
wages, prices, and productivity, all lagged two periods. When we tried productivity
only without the lagged wage and price indices, the estimated equation performed less
well, with a smaUller R2 and greatly reduced significance. of the TARGET variable.
Periods covered by both regressions go from the second quarter of 1971 to the third
quarter of 1988 (70 observations). Figures in parentheses are t-statislics.
The Structure of Korean Labor Markets and Wage
Differentials
The previous discussion on wages and productivity referred to the
formal manufacturing sector. The data was limited to wage employees
in firms employing more than ten workers. A great deal of
employment in Korea has always been in the informnal sector.
However, the lack of data means that we cannot provide a
comprehensive picture of the whole informal sector. Thus, we focus
on specific groups in the sector whose relative earnings appear
significant in Korea's process of structuiral change.
The size of the informal sector employment can be estimated by
comparing two sources of employment data: the Economically Active
Population Survey (Economic Planning Board), which estimates total
employment on the basis of a household survey, and the Actual Labor
Thc Repuiblic of Korea 557
Figure 12.6 Real Aniual. Income, per Earner, by Earner Type,
1966-85
4- :
3.5 - ,
2.51
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0
0.
3- -- -
* 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984
Year.
_-E Farming. - Manutacturudg. * Urban Salary ---- Urban Wage
wage earner's household income per earner was significantly higher.5
. However, over the period the rate of growth of household income was
substantially lower than that of farm income, so that the differential
was squeezed.
After the stabilization efforts following the oil crisis, urban earnings
stagnated for a couple of years. Manufacturing earnings had a very
low growth only in 1975, but urban wage earners' income fell before
this, both in 1973 and 1974. However, the slowdown of the economy
had no impact on the growth of income in the farm sector. The
rural/urban eamings difference fell during this period-of adjustment.:
S. This -may be due to (a) the -exclusion from the sample of single member urban
households who would presumably have low earnings, and (b) the inclusiDn in the
family income of supplementary income from other sources.
558 Dipak Mazumdar
Earnings in the nonfarm sector took a sharp upward turn during
the big push in the second half of the 1970s, when real wages in
manufacturing increased at a rapid rate higher than the rate of growth
of labor productivity. As figure 12.6.indicates, the incomes of urban
households went up even faster. Farm incomes per wiorker increased
by about the same absolute amount per year as in the previous period,
so that the rate of growth slowed down. Thus, the period of the big
push widened the rural/urban wage differential substantially. In 1979
manufacturing wages were 80 percent higher than farming earnings
and urban wage earners' income was 100 percent higher.
As during the first episode of stabilization, urban earnings fell after
the second oil shock, while farm earnings continued to grow at much
the same rate. The rural/urban differential'fell, but not by enough to
restore it to the levels of the early 1970s. Since 1982, rural and urban
earnings have increased at a somewhat similar rate. The widening of
the differential in favor of urban wage earners'that occurred in the late
1970s seems to have been a permanent one.
Some economists maintain that the comparison of average .incomes,
particularly after'the stabilization program of the 1980s, may be
giving too favorable a picture of the farm sector. The deficit in the
Grain Management Fund-which, as we have' seen, supported the
prices paid to farmers above the prices of cereals sold to consumers-
was drastically reduced as part of the post-1981 stabilization package.
The slow down in the inflation rate clearly helped the deficit reduction
process, but some believe that the terms of trade for the farming
population deteriorated. Amsden (1987, p. 36) suggests that such a
deterioration prompted the mass exodus out of agriculture between
1982 and 1985 "even larger than the migration associated with'the
1980 harvest failure." Moreover, "the last wave of migrants was
believed to consist of relatively older people, unequipped to enter the
labor force and unaccounted for in the unemployment statistics which,
therefore, were lower than otherwise" (Amsden 1987, p. 36).
The outmigration from agriculture might indeed have prevented
the rural-urban differential from increasing further in favor of the
urban sector in the post-1981 period. Also the earnings distribution
within the farn sector might have deteriorated (a point on which no
statistical information is available). In any case, taken in conjunction
Thle Republic of Korea 559
with the earnings differential in favor of urban workers being at a
higher level in the 1980s than in the early 1970s, the implication is
that the farming sector did suffer a relative deprivation after the boom
of the late 1970s and the subsequent adjustment of the economy.
Wage Difference by Size of Firms
The difference in labor earnings between the informal and the
formal sectors in the nonfarm economy is of major importance in the
history of Korean development. Unfortunately, the absence of
comparable household surveys does not permit examination of trends
in differentials for too many subsectors. Information does exist,
however, on wage difference by size of manufacturing firms. The
wage levels in small firms could be expected to approximate levels of
earnings in the informal sector.
The government policies thal4 led to Korea's export-oriented
industrialization also produced a dualism in the manufacturing sector.
aWhile government policy towards domestic market-oriented small-
scale industry has been characterized by benign neglect or active
discrimination, export/large scale sector has enjoyed considerable
advantages from the government through direct and indirect
subsidies" (Park 1980, p. 57).6f
Credit policy played a central role in this process of
differentiations. The preferential interest rate on export credit was
reduced to 6 percent in June 1967, while the ordinary bank rate was
set at 26 percent in 1965. In addition, exporting firms enjoyed a string
of preferences in import licenses, tax concessions, and favorable tariff
rates for imported inputs. "These subsidies were disproportionately
favorable to large-scale industries. In 1974, only 6 percent of small-
or medium-scale industries (less than 200 employees) were designated
by government as 'export Industries.' Government export subsidies
were also scaled according to export volume and performance" (Park
1980, p. 61).
The government also encouraged Korean industry to adopt state-
of-the-art technology developed in the high-income countries to
6. See chapter 11 of this work for an extended discussion of the issues summarized
here.
560 DipakMazuiKdar.
eniable it to cater to world markets. This led to the adoption of capital-
intensive technology, a trend abetted by the low cost of loanable
funds. Further, the recently designed technology of mass production
favors large-scale operations to reap the benefits of' machine
specialization. Thus, successive plants in Korea's export.industries
have been designed. for increasingly large-scale production.
The differences in technology and labor productivity between large
and small firms are associated with large differences in wages. To
some extent the observed difference in average earnings per worker
reflect differences in skill composition, but even. for a relatively
homogeneous group like semiskilled production workers, very large
wage differentials exist. An element of profit sharing, clearly enters
into the high wages paid by large firms with high labor productivity.
As already mentioned, the importance of labor unions in determining
wages is relatively srmall. 'However, employer paternalism, plus
incentives for efficiency and low turnover, is involved in setting wages
at high levels in large firms. As noted earlier, basic wages account for
only. a part of total earnings in Korea. Various allowances and annual
special earnings bonuses are a substantial component of earnings, and
this proportion increases sharply with firm size.
* Table 12.3 shows the evolution of employment, output, and labor
productivity by firm size between 1960 and 1982 (note that the
definitions of small, medium, and large firms are different in 1960
and 1963 from the definitions for later years). The data show the
enormous importance of firms with fewer than 100 employees (small
and medium firms) in the early 1960s. Small firms (those employing
fewer than 30 workers) employed 45 percent of total workers- and
produced a third of gross output. Value added per worker in the small
firms was half of that in the large firms, and in the medium firms it
was twa-thirds of that in the large firms.
A major change seems to have taken place between 1969 and
1975. The large firms expanded fast at the expense of the small, with
the latter's share dropping from 32 to 17 percent in terms of
employment, and from 17 to 8 percent in terms of gross output. At
the same time, the difference in value added per worker narrowed
markedly, from 14 percent of the level in large firms in 1969 to 40
percent in 1975.
The Republic of Korea 56)
Table 12.3 Technology, Size, and Productivity Differentials in
Manufacturing Establishments, Selected Years
Number of Value added Fired assets
employees Gross output per worker per worker
Year (% of total) (% of total) index index
1960
Small 45.2 36.9 59.4 n.a.
Medium 22.4 20.4 67.8 n.a.
Large 32.3 42.6 100.0 n.a.
1963
Small 42.0 31.5 46,3 n.a.
Medium 23.0 22.9 66.8 - n.a.
Large 34.9 45.5 100.0 n.a.
1969
Small 31.6 16.6 14.2 16.6
Medium 20.1 15.0 19.8 43.2
Large 48.2 68.3. 100.0 100.0
1975
Small. 17.4 . 8.3 40.5 18.6
Medium 20.2 15.7 69.6 35.4
Large 62.3- 75.8 100.0 100.0
1980
Small 18.3 8.1 42.5 19.8
Medium 22.7 . 15.6 61.1 42.4
Large 58.9 76.1 100.0 100.0
1982
Small 21.4 9.1 37.2. 22.7
Medium 23.3 17.1 . 57.1 39.5
Large 55.1 73.6 100.0 100.0
n.a. = not available.
Note: For the years 1960 and 1963, small = 5-29 employees, medium = 30-99
employees, and large = 100+ employees. For the years 1969, 1975, 1980, and 1982,
small = 5-49 employees, medium 50-199 employees, and large = 200+ employees.
Source: Report on Mining and Manufacturing Survey.
The shocks of the mid-1970s and the early 1980s arrested the fast
relative expansion of large firms despite the big push of. 1975-79.
The proportion of employment in large firms fell from 62 to 55
562 Dipak Marumdar
percent. The difference in value added per worker, however, widened
somewhat, perhaps reflecting rationalization and weeding out of less
efficient firms in the large-scale sector.
Thus, while the Korean experience supports the model of a
shrinkage of the large firm sector during the periods of adjustment
with the slack taken up by small firms, the changes are not nearly as
dramatic in the difficult period of 1975-82 compared to the
expansionary phase of 1969-75.
Turning to the differential in earnings by size of firm in
manufacturing, Table 12.4 gives the differentials in average earnings.
These data show' a substantial increase in the differential, particularly
with respect to small firms during the 1960s. The trend was reversed in
the 1970s, both, in the years leading up to the first oil crisis, and
subsequently during the big push. By the end of the 1970s the small-
large differential was at about the same level as in 1960, but the
second oil crisis and the adjustment of the 1980s again saw a widening.
of the differential, but to a smaller extent than-in the 1960s.
Table 12.4 Indices of Differentials in Average Remuneration by Firm
Size in Manufacturing, Selected Years.
Large/small Medium/small Large/medium
Year (small = 100) (small= 100) (medium = 100) .
1960 136.5 99.9 136.6
1967 155.7 126.9 122.7
1970 1o80.9 147.6 122.6
1974 152.4 129.3 117.8
1979 130.1 115.9 . 112.3
1983 147.6 120.3 122.7
1986 149.0 119.4 124.8
Note: Small firms are defined as having 5-49 workers, medium firms have 50-199
workers, and large firms have 200 or more workers.
Source: Korea-Statistical Yearbooks (1962, 1976, 1981, 1985) and Reports on
Mining and Manufacturing Surveys for 1967 and 1970.
Tle Republic of Korea 563
The data in table 12.4 do not control for skill, education, and skill
differences. When we do control for such differences as is done in
table 12.5, a significant. decline in the differentials over time is
confirmed for the 1970s, except for female university graduates.
Taken together. with table 12.4, we could conclude that dualism has
indeed been accentuated within Korea's manufacturing sector, butlthat
this process has taken the form of selecting workers with better human
capital .attributes for the large-scale sector rather than widening the
standardized wage differential by firm.size.
Male-Female Differences
Table 12.6 gives the labor force participation rates of males and
females for farm and nonfann households. The participation rate for
females in farm households does not show much of a trend, but the
rate in nonfarm households, although well below the rate in farm
households, has been slowly increasing since 1970 (except for a small
dip in 1981 and 1984).
The process of development is generally accompanied by.
substantial increases in female participation arising from both the
supply and demand sides. On the supply side, important factors
* helping the process are rising levels of education, reduced fertility, and
a general change . in attitudes toward market work by women. Korea's
educational expansion seems to have benefited women as muc.h as
men. The average number of years of schooling of women has. risen
fr-om 2.92 years in 1960 to 6.63 years in 1980 (as against the overall
average of 3.86 and 7.61; Park 1980, p. 6). At the same time fertility
levels have declined drastically (by more than half in the last. 20
years), reducing the number of small children at home. This would
tend to increase market activity. for married females. On the demand
* side, industrialization and the growth of urban services-social and
private-create opportunities for female employment. What makes the
Korean case unusual is that despite 'the presence of these factors at
levels above those for. other developing countries, after two decades of
development, the nonfarm participation rate for females is well below
that of other countries. Even the neighboring countries of Asia had
significantly higher rates: 49.7 percent for Hong Kong, 48.9 percent
for Japan, 45.8 percent for Singapore, and a high of 78.4 percent for
Table 12.5 Earnings Differentials by Size of Firm, Sex, and Educational Level of Workers, 1967 and 1980
(base - 100, firms withz 10-29 workers in eacs category)
University AMiddle or elementary
graduate school graduate
Male Fenale Male Femare
Firm size
(number of workers) 1967 1980 1967 1980 1967 1980 1967 1980
10-29 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
30-99 117.6 103.4 88.5 88.7 113.0 105.1 107.3 98.2
100-299 131.7 111.3 126.5 109.3 124.3 125.2 124.0 103.6
300-499 149.2 112.1 99.9 101.5 157.4 131.7 132.5 104.7
500+ .71.0 113.6 96.4 119.0 201.9 133.4 163.3 107-9
Notes: For 1967, firm size ranges from 100 to 199 (instead of the 1980 range of 100 to 299); and from 200 to 499 (instead of the
1980 range of 300 to 499). Therefore, comparisons between these two groups should take this into account. The category for males
and females with a middle or elementary school education for 1967 refers to production workers.
Sources: Administration of Labor Affairs (1980, tables 111.4 and 111.5, pp. 336-461); Bank of Korea (1967, table 2, pp. 50-65).
Tiie Republic of Korea 565
Table- 12.6 Labor Force Participation Rates oy Sex, 1970 and
1975-85
Farm households Nonfarm houselolds
Year Male Female -All Male Female All
1970 75.2 48.2 60.9 75.1 29.8 51.5
1975 - 73.8 51.8 62.7 75.1 31.2 52.5
1976 74.5 55.3 64.8 74.7 33.7 53.3
1977 74.3 52.5 63.3 76.9 33.5 54.0
-1978 74.5 54.0 63.9 75.3 35.6 54.6
1979 73.5 54.2 63.6 74.4 35.9 54.4
-1980 72.4 53.0 62.5 74.2 36.1 54.4
1981 72.1.- 53.4 62.6 73.7 35.4 53.8
1982 70.4 53.6 61.9 73.4 37.5- 54.7
1983 68.7 51.3 59.8 71.8 37.9 54.2
1984 68.8 50.1 59.3 - 69.6 36.1 52.2
1985 68.9 50.7 59.7 -69.8 37.7 53.1
Source: Grootaert (1987, table 2, p. 5).
Thailand. The United States and Canada have rates of around 52.0
percent (Grootaert 1987, p. 6, quoting the LO Yearbook for 1984).
Institutional changes facilitating greater participation of women in
the nonfarm work force have been slow in coming. Grootaert (1987)
points out that part-time work is not very common for Korean women;
only .about 7 to 8 percent work 35 hours or less per week. The
distributioni of workers by hours worked showed little difference
between men and women, except that men do more overtime work
(more than 54 hours a week) (Grootaert 1987, table 10, p. 15).
Evidently, Korean employers have not. taken the initiative in
developing the market for jobs in clerical, sales, and assembly line
production work that can. easily be split into two part-time jobs.
Grootaert also points to the government's limited efforts to establish
566 DipakMazumdar
public day care centers and the various restrictions on private sector
initiatives.
Turning to the composition of female employment in the nonfarm
sector, tables 12.7 and 12.8 show time series for the proportion of
female employment by industry and by occupation, respectively. As
Table 12.7 Evolution of Female Employment by Industry, 1963-84
(female workers as a percentage of the total labor force)
Mining and
Year . Agriculture manufacturing Construction Services
1963 37.98 27.89 8.81 32.26
1964 37.97 29.57 3.83 32.08
1965 38.32 28.03 4.62 34.30
1966 38.84 29.46 3.35 31.79
1967 39.28 30.76 3.86 32.35
1968 40.22 31.75 6.01 34.22
1969 39.01 32.76 8.01 33.59
1970 41.62 31.18 1.76 34.67
1971 41.82 33.40 2.59 34.27
1972 42.95 33.09 230 - 32.90
1973 42.13 37.78 2.96 34.67
1974 41.51 35.321 5.11 35.59
1975 41.53 33.55 4.89 -35.9].
1976 42.64 37.59 4.91 37.33
1977 41.59 37.99 7.84 35.12
1978 44.10 38.39 7.67 36.56
1979 44.57 38.52 7.66 37.54
1980 43.77 38.05 8.56 38.23
1981 43.66 37.72- 7.89 38.43
1982 43.74 37.30 6.98 - 41.01
1983 43.16 37.10 6.99 42.13
1984 42.72 36.49 7.42 41.43
Source: Economic Planning Boird (various years).
Ther "b/ic of Korea 567
Table 12.8 Evolution of Female Employment by Occupation, 1963-
84 (female workers as a percentage of the total labor force)
Professional Produiction
Year and managerial Clerical- Sales Service Agriacture operations.
1963 21.46 11.28 44.81 19.83 38.08 19.83
1964 19.18 10.20 45.52 20.64 38.00 20.64
1965 18.03 10.03 46.21 20.06 38.38 20.06
1966 16.73 9.85 43.97 22.21 38.67 22.21
1967 16.14 11.63 42.77 23.61 39.37 23.61
1968 16.25 17.00 :44.25 24.38 40.35 24.38
1969 15.57 13.81 42.95 25.09 39.30 25.09
1970 18.40 13.54 42.70 57.70 2.!36- 23.35
1971 19.43 16.55 41.25 23.72 41.92 23.72
1972 16.45 16.79 42.10 22.81 42.93 22.81
1973 19.11 17.83 42.08 27.88 41.99 27.88
1974s 19.90 19.39 .41.67 26.92 41.43 26.92
1975 20.86 20.88 40.72 57.38 41.41 25.84
1976. 20.56 23.08 43.03 58.44 42.57 29.96
1977 22.53 24.47 41.60 54.38 41.61 29.92
1978 25.08 27.67 42.21 56.14 44.17 29.16
1979 26.34 30.60 43.41 56.27 44.62 29.02
1980 25.34 32.75- 43.72 58.12 43.83 27.73
1981 23.54 33.62 44.21 57.72 43.82 26.68
1982 26.64 34.12 45.96 58.18 44.00 27.10
1983 26.67 34.21 47.35 59.99 43.38 47.89
1984 27.16 33.58 46.82 60.72 42.99 26.99
Note: Thir survey includes all women age 14+ except those in the armed forces,
foreigners, and prisoners.
Source: Economic Planning Board (various years).
far as mining and manufacturing are concerned, the percentage of
women workers increased at a modest but steady rate until 1980, and
declined somewhat thereafter. There was, however, a decline in the
crisis years following the first oil shock in 1974 and 1975. Thus, the
568 Di pakMazuitndar
evidence suggests that the proportion of women in manufacturing
employment responds significantly to cyclical demand factors.
The increase of less than.10 percentage points in the share of
female employment in industry must be considered rather marginal
compared to the large shifts in the economy's industrial structure in
the 20-year period. Much more pronounced growth of female
employment is seen in the occupations categories in table 12.8,
especially in the clerical.and service categories, where the proportion
of women in total employment has doubled. However, the increase in
female employment in these white collar categories has been confined
to narrow low-income groups. "Clearly clerical work has undergone a
major and rapid image shift from a male to a female occupation ..
More than 80 percent of the clerical work force below 25 (in i984)
are women; the clerical work force above age 35 is 95 percent male"
(Grootaert 1987, p. 20). In the seivice categories women appear to be
crowded into two subgroups, namely, teachers and medical, dental, and.
veterinary personnel. In the administrative and managerial category,-
only 2.9 percent are women, far below the proportion in other East
Asian countries (13.1 percent in Hong Kong, 6.1 percent in Japan,
17.4 percent in Singapore, and 19.8 percent in Thailand) (Grootaert
1987, table 14; p. 22).
Another aspect of the differential conditions of employment by sex
is revealed by looking at the. changes in employment by work status.
The employed labor force is classified into self-employed.workers,
family. workers, and employees. The distinction between the first two is
important for work status.. A self-employed person could be a small
entrepreneur and often -earns more than an employee. A family
worker, by contrast, is an unpaid working, member of the -household.
One of the most striking developments in Korea is.that in 1984, nearly
80 percent of family workers in the mining and manufacturing sector
were women. Betwveen 1.976 and 1984 the proportion of women
among employees remained unchanged, but the proportion of women.
among family workers rose from 65 to 80 percent, while women's
share in the self-employed category was halved to 26 percent
*(Grootaert 1987, p. 27 and table 18). Evidently, during the big push
and the subsequent adjustment, the number of vomen entrepreneurs
dropped sharply.
Thie Republic of Korea 569
l,mployees in Korean labor markets can be categorized as regular,
temporary, or daily workers. The security of tenure and access to
bonuses and other benefits associated with internal labor markets lead
Korean employers to hire a large proportion of their work force on
temporary or daily contracts. This allows them to vary the size of the
work force with changing business conditions and holds down labor
costs. Table 12.9 breaks down the three categories of workers by sex
for 1963'85. The table shows that the proportion of regulars among
male workers has increased steadily during the period except for the
Table 12.9 Distribution of Total Employment by Sex and Status,
1963+85
(perctnt, nonfarrn hIouseholds)
Men Womeon
Regular Temporary Daily Regular Teniporary Daily
Year employees enmployees vorkers employees employecs workers
* 1963 49.6 .17.6 32.6 34.6 26.6 38.6
1964 48.6 20.4. 30.8 32.0 34.0 33.8
1965 52.4 20.7 27.7 35.2 36.4. 28
1966 55.3 17.4 27.2 39.3 30.0 30.6
1967 57.5 17.4 25.0 45.7 23.2 30.9
1968 62.9 14.5 22.4 . 49.0 22.5 28.4
1969 65.7 1 1.3 22.8 55.3 20.1 . 24.5
1970 68.6 11.8 19.2 57.9 22.1 19.8
1971 69.2 -11.3.. 19.4 55.2, .23.9 20.8
1972 62.4 12.6 24.8 52.0 26.7 21.1
1973 56.5 19.7 23.7 44.7 34.5 20.7
1974 62.3 .17.0 20.5 46.7 33. 3 19.8
1975 60.9 19.5 .19.4 . 48.3 33.9 17.6
1977 64.9 15.6 19.3 54.9 26.1 18.8
.1979 - 68.3 13.8 17.8 57.7 24.9 . 17.2
1980 71.5 12.1 16.2 58.5 24.6 16.7
1982 73.7 . 10.8 15.3 62.3 21.5 16.0
1984 71.9 14.7 13.2 50.6 30.5 18.7
1985 72.3 14.9 12.8 48.8 30.4 20.8
Note: Tl,e. survey is based on the population aged 14 and over and not in the army,
imprisoned, or foreigners.
Source: Ecbnomic Planning Board (various years).
570 DlpakMazurdar
period 1972-75, but for women the proportion of regular employees
in 1985 was about the same as that in 1968. The percentage of regular
female employees fell significantly both after the first oil shock in
1973-75 and following the second period of stabilization in the early
1980s. Evidently, women workers have been used in a more
"marginal" way in the last two decades.
Turning to earnings,, the trend in the female/male differential for
different educational groups is plotted in figure. 12.7. This shows that
Figure 12.7 Female/Male Earnings Ratio by Level of Education,
1970-84
0.74 - - ,
0.72 -
0.70 -
0.68 . . .'
o0.66-
0.64 K
0.62 -1.
o.6o
0.58 /
0.56 l . II.
0.54/ /
0.52-
0.50 /
1970 1974 1975 1977 1980 1981 1982 1983 1984
Year
unive- t -* - Junior College * igh School
- - Middle School
Note: The survey is based on a random sample of establishments employing ten or
mdre regular employees, excluding agriculture, forestry, fishing, government
administration, public education, the army, and the police force.
Source: Occupational Wage Survey.
The Republic of Korea 571
while university and college educated.women have improved their
relative itarnings since 1970 (although still earning a little rnore than
70 percent of the male average), the bulk of female workers with hi,,:s
school or middle school educations have more or less-the same relative
earnings as in the early 1970s. A trend toward reducing the gender
differential. for this group. (particularly the high school graduates) was
apparent during the big push of the late. 1970s, but this gain was lost
in the 1980s.
Conclusion
This chapter investigated the successful adjustment to external
shocks 'of an open economy heavily dependent on key imports. In the
Korean case, a high investment rate maintained throughout the growth
process of the last -two decades accentuated the country's
macroeconomic problems. The rate of investment continuously
outrun the rate of domestic savings, which- kept the economy on a
tightrope of external deficits on the current account and internal
inflationary pressures for much of the period. Maintaining its external
competitiveness was -of central importance to this export-oriented
economy.
The evidence shows that an active exchange rate policy (continuous
devaluation of the won except for a few years in the late 1970s and
maxi devaluations during shocks) has been central to the mechanism'
that kept unit labor costs falling.throughout the period. However, the
devaluation's success.in producing the desired result.depended on
policies affecting both the capital and labor markets. In the capital,
markets, the maintenance of cheap. credit for the, large-scale sector
cushioned the exporting firms from the rising costs of interest.
payments and rimported inputs that devaluation induced. Equally
important was the tight government control of external capital flows
that prevented destabilizing speculative movements of capital.
On the labor market front, the evidence shows the importance of
state paternalism in wage negotiations in the formal sector in- keeping
real wage increases in line with productivity growth, but somewhat
below it in most periods (again with the exception of the big push
period). It was also eminently successful in drastically slowing down,
or even halting, real wage.. growth during the short-run periods of
572 DipakMazlmdar
crisis. However, the wage-productivity relationship behaved in the
healthy way it did because real wages increased significantly in most
years. This experience must have been instrumental both in securing
worker acquiescence to the temporary stagnation of wages and in
preventing destabilizing inflationary expectations from developing.
The critical factor here was the strongly positive time trend in total
factor productivity growth. It is this that kept the rate of real wage
growth high at the same time that the unit labor cost in industry was
falling.
Real wages in Korea's large-scale manufacturing sector have
"risen faster, possibly than in any previous or contemporary industrial.
revolution" . (Amsden 1990, p. 79). Our analysis suggests that'the
major factor behind this real wage growth was probably profit sharing
as an incentive' scheme in (tie process of wage determination.' The
internal labor market structure of the' large manufacturing firms
encouraged this process, as did state paternalism. We would expect that
in large segments of the labor market outside the large 'firms, the
mechanisms of the internal labor markets would be, weak, and wages
would be lower in these segments and would rise less fast. This is,
indeed, what earlier writers have suggested. For example, Amsden
writes: "By world standards, Korea has the highest inter-'
manufacturing industry wage dispersion and-the widest gap in gross
wages between the sexes (Krueger and Summers 1986; Lee 1983).
Underlying the rapid rise in real' wages beginning in 1965 was the
preening of a labor aristocracy: male, employed by the chaebol, in the
new heavy industries. At the opposite end of the spectrum is the
economically active population-in the informal sector" (1987, p. 4).
The absence of comprehensive household surveys in Korea
precluded the investigation of formal-informal earnings differentials
in large parts of the labor market, but we were able to examine a
limited range of wage differentials. Earnings in the farm sector did not
perform all'that badly compared to average earnings in the nonfarm
or urban sectors. Up to the first oil shock of 1973 annual farm
incomes per worker increased'at a rate only slightly lower, than the,
average earnings i, manufacturing. However, during the big-push of
the late 1970s, the differential in favor of manufacturing wages
increased from 50 to 80 percent, and although it fell somewhat after
Thte Republic of Korea 573
the adjustment that followed the second oil shock, in the mnid-1980s, it
was well above earlier levels.
Analysis of wage differentials by firm size within the formal sector
(excluding very small firms employing fewer than ten workers) slpws
very large differences in average earnings that increase significantly
over time, but to a large extent these differences are due to differences
in skill composition. A comparison of wage differences between more
homogeneous categories of labor reveals that the differential, which
had stood at about 100 percent in 1967 (for male- middle or
elementary school graduates), fell to 33 percent in 1980. Taken
together, the evidence suggests that the accentuation of labor market
dualism in Korea has taken the form of selecting workers with better
human capital, attributes for the large-scale sector rather than widening
the standardized wage differential by firm size. However, the large-
scale. sector seems to have taken the main brunt of short-run
adjustment following the oil shocks. Wages stagnated in this sector for
short periods after the adjustment, and the relative expansion of the
large firms at the expense of the small, which was occurring rapidly
until 1973, halted in the next decade.
Amsden suggested that "not only Korea set. world records with its
growth rate in wages, it also has outcompeted other countries. in its
discrimination against women workers" (1990, p. 85). We found
evidence of surprisingly small increases in participation rates of
women in the nonfarm sectors; severe occupational crowding, a larger
proportion of women in lower status jobs like family workers or in
temporary positions, and a more or less constant wage differential in
favor-of male workers over the years. The average earnings of fermale
workers with a high or middle school eduication were 52 percent of the
earnings of men with a similar education. As we expected, womeen
were disproportionately affected during the postshock periods. of
sharp adjustment.
While Korea's record in solving some of the structural problems in
the. labor market has not been very good, its astonishing success in
managing the economy-'s short-run macroeconomic balance may also
be threatened in the future. A full. analysis of contemporary
developments in the labor market is beyond the scope of this chapter,
but we should, in concluding, draw attention to the explosive increases
4~~.
574 DipakMazumdar
in wages in manufacturing since 1987. This type of wage push,
emanating from a breakdown of the traditional relationship between
the Korean Federation of Trade Unions and. that of employers,
threatens to upset the wage-productivity balance that has been central
to the success of Korea's macroeconomic stability. If wages go
soaring above productivity. growth, Korea's share of the export market
relative to its close competitors will undoubtedly be threatened, and in
addition, the country might have to deal with an inflationary spiral .and
the need for much more painful adjustments when external shocks
develop in the future. According to the World Bank: "A recent study
of manufacturing unit labor costs found that between 1980 and 1986
Taiwan's ULC rose 56 percent relative to Korea's. In 1987 and 1988
the two economies' ULCs increased at the same rate. It was only in the
first-quarter of 1989 that Korea's ULC began to increase relative to
Taiwan" (1989, p. 7). The concern of the coming years is how much'
and in what way can Korea contain these new developments in the
labor market. The other significant question is Korea's ability to
sustain-if not to increase-the record rate of total factor productivity
growth that it has achieved in the last two decades or more.
APPENDIX
Table 12.AI Net Barter and Income Terms of Trade, 1963-88
(index, 1980 = 100)
Percent cliange
Net barter Income terms
Year terms of trade Barter income
1963 - 111.48 121 -
1964 112.83 1-88 1.21 55.60
1965 114.87 3.36 1.81 78.39
1966 128.15 4.91 11.57 -46.41
1967 132.85 6.23 3.66 26.80
1968 138.06 -8.93 3.92 43.31
1969 133.27 12.40 -3.47 38.89
1970 134.19 16.06 0.69 29.45
1971 133.08 20.57 -0.82 28.09
1972 132 35 30.76 -0.55 49.56
1973 125.62 45.82 -5.08 48.97
1974 101.31 40.73 -19.35 -11.12
1975 . 92.10 45.10 -9.09 10.74
1976 105.10 69.90 -14.12 54.99
1977 112.40 89.00 6.95 .27.32
1978 117.80 106.70 4.80 19.89
1979 115.30 . 103.40 -2.12 -3.09
1980 100.00 100.00 -13.27 -3.29
1981 97.90 115.00 -2.10 15.00
1982 102.20 - 127.90 4.39 11.22
1983 103.10 150.00 0.88 17.28
1984 105.30 -177.10 2.13 18.07
1985 105.90 191.70. 0.57 8.24
1986 114.70 234.70 8.31 22.43
1987 118.08 296.94 2.95 - 26.52
-l 988 - - - 121.36 344.87 - 2.78 16.14
Note: Net barter terms of trade are defined as the ratio of export to import unit value
index. Income terims of trade are defined as the product of the net barter terms of trade
and the export quantum index.
575
576 DipakMazuimdar,
Table 12.A2 Major Economic Indicators, 1967-88
Real GNP Current Exports Bud et Unemployment
growlth account growth delficit rate
Year (% per annurn) (% GNP) (% (% GNP)
1967 .6.60 -4.12 28.00 - 6.2.
1968. 11.30 -7.49 . 42.20 - 5.1
1969 13.80 -7.76 36.90 - 4.8
1970 7.60 -7.35 34.00 1.60 4.5
1971 8.60 -9.38 27.90 2.30 4.5
1972 5.10 -3.56 52.10 4.60 4.5
1973 13.20 -2.28 98.60 1.60 4.0
1973 . 13.20 -2.28 98.60 1.60 4.0
1974 8.10 -13.05 38.30 4.00. 4.1
1975 6.40 -9.05 13.90 4.60 4.1
1976 13.10 -1.09 51.80 2.90 3.9
1977 9.80 0.03 30.20 2.60 3.8
* 1978 9.80 -2.17 26.50 2.50 . 3.2
1979 7.20 -6.43 18.40 1.40 3.8
1979 7.20 -6.43 18.40 1.40 3.8
1980 -3.70 -9.56 16.30 . 3.20 5.2
1981 5.90 -7.21 21.40 4.70 4.5
1982 7.20 -3.91 2.80 4.40 4.3
1982 7.20 -3.91 2.80 4.40 4.3
1983 12.60 -2.07 11.90 1.60 4.1
1984 9.30 -1.62 19.60 1.40 3.8
1985 7.00 -1.01 3.50 1.00 4.0
1986 12.90 4.39 14.60 1.80 3.8
1987 * 12.80 7.39 36.20 - 3.1
1988 12.20 7.84 28.40 2.5
Note; 1988 GNP growth rate is preliminary.
Source: Bank of Korea Principle Economic Indicators, Economic Statistics Yearbook
Thte Republic of Korea 577
Table 12.A3 Data Used for Unit Labor Costs in Manufacturing,
1967-86
Annual. Atnnual Maanufaciurlng Average
value added wages deflator CPI exchatige rate
Year (billion wonp) (billion woas) (1980 = 100) (1980 =100) (waon/uss)
1967 206.6 53.3 17.1 - 15.3 270.5
1968 300.1 76.6 18.3 17.0 276.7
1969 424.2 105.7 19.9 19.1 288.2
1970. 547.9 .137.1 22.3 22.2 310.6
1971 688.6 160.4 23.4 25.2 347.2
1972 899.3 211.5 26.7 28.1 392.9
1973 1,379.6 310.3 29.9 29.0 398.3
1974 1,867.2 451.3 37.7 36.1 404.5
1975 2,828.1 651.6 45.4 45.2 484.0
1976 4,075.1 1, 009;1l 52.8 52.1 484.0
1977 5,596.9 1,460.4 58.5 57.4 484.0
1978 8,193.0 2,221.8 66.2 65.7 484.0
1979 9,205.0 2,922.1 78.9 -77.7 484.0
1980 11,857.0 3,471.7 100.0 100.0 607.4
1981 15,412.0 4,133.5 113.7 121.3 681.0
1982 17,306.0 4,754.1 119.4 130.1 731.1
1983 20,912.0 5,499.6 122.8 134.5 775.8
1984 24,656.0 6,495.1 123.7 137.6 806.0
1985 26,737.0 7,244.5 126.7 141.0 870.0
.1986 32,882.0 8,607.3 131.3 144.2 881.5
Source: U.N. Yearbook of Industrial Statistirs, IMF fnternational Financial Statistics,
Bank of Korea.
578 DlipakMazumdar
Table 12A4 Annual Earnings per Worker in the Farming and.
Manufacturing Sectors, 1966-85
Real annual incomeper Real annual manufacturing
farm worker earnings.
Thfousands Percentage cliange Thousands Percentage change
Year .1980 won per annum 1980 won-. per annmtm
1966 400 - 467 -
1967 426 6.4 520 11.1
1968 474 1 1.2 593 14.0
1969 518 9.2 707 19.2
1970 537 3.7 787 11.3
1971 660 22.9 . 826 5.0
1972 688 . 4.2 858 3.8
1973 735 6.8 924 7.6
1966-73
average 555 11.9 710 . 13.9
1974 785 6.7 . 1,004 8.6
1975 822 4.7 1,018 1.4
1976 876 6.5 1,190 16.8
1977 964 10.1 .1,446 21.4
1978 1,004 4.1 1,696 17.3
1974-78
average 890 6.9 1,271 17.2
1979 1,072 6.7 1,845 8.7
1980 1,081 0.8 1,760 -4.6
1981 1,152 6.5 . 1,742 -0.9
1982 1,232 6.9 1,864 6.9
1983 1,292 4.8 2,023 8.5
1984 1,490 15.3 2,138 5.7
1985 1,579 5.9 2,294 . 7.2
1979-85
average 1,271 7.8 1,952 4.0
The Republic of Korea 579
Table 12.AS Annual- Household Income per Earner for Urban
Households, 1966-84
Salary earner households Wage earner households
Thousands Percentage change Thousands Percentage change
Year 1980 won per annum 1980 win per annum
1966 1,316 - 681 -
1967 1,569 19.2 981 44.1
1968 1,615 2.9 1,006 2.4
.1969 1,609 -0.4 1,038 3.2
1970 1,562 -2.8 975 -6.0
1971 1,652 5.7 1,041 6.7
1972 -1,685 1.9 1,073 3.0
1973 1,690 0.2 1,046 . -2.5
t966-73
average 1,587 . 4.0 980 7.6
1974 1,806 6.8 990 -5.3
1975 1,873 3.7 1,052 6.2
-1976 2Z428 29.6 1,146 8.9
1977 2,932 20.7 1,427 24.5
1978 3,284 12.0 1,851 29.6
1974-78
average 2,465 20.4 1,293 21.7
1979 3,507 6.7 2,058 11.1
1980 3,684 5.0 1,985 -3.5
1981 3,646 -1.0 2,006 - 1.0
1982 3,568 -2.1 1,787 -10.9
1983 3,760 5.3 1,938 8.4
1984 3,962 5.3 2,074 7.0
197944
average 3,688 -16.6 1,975 -16.6
Note on the Data
Data on daily farm wages and incomes were collected from the
Korean National Agricultural Cooperative's monthly report. These
sarne figures are repotted in the Statistical Yearbook published by the
Economic Planning Board. Census years were 1970 and 1975; all
580 DipakMazumdar
other years are based on sample surveys. The survey is carried out by
the Ministry of Agriculture and Fisheries and is based on a sample of
farm households engaged primarily in. farming and cultivating a plot
of land larger than 0.1 hectare. The survey is conducted monthly and
revised after censuses. Income includes agricultural receipts, side-
business receipts, nonbusiness receipts (wages, rent, and so. on) and
property (assets) receipts less farm and side-business expenses. Daily.
farm wages are also reported in these docuiments. Here, men's daily
wages (cash and in kind) are shown for all workers.
The price index used to deflate farm incomes was the prices paid
by farmers index, reported in the same documents. These prices are
.collected at 85 rural'markets covering 201 items.
Farm income was normalized to per 'worker farm income by
dividing total farm income (including income from nonfarm sources)
by the number of farm workers.
Data on manufacturing earnings were extracted from the Statistics
Yearbook published by. the,Economic Planning Board.' These statistics
are -collected by the Ministry of Labor in a monthly wage survey. The
survey covers all manufacturing establishments with ten or more
employees. The earnings.reported are the average monthly earnings
of all (men and women) regular employees. Regular workers are those
whose employment contract is for one month or more,: and who
worked for more than 45 days during the three' months prior to the
:reporting day. Monthly earnings include overtime pay, bonus pay,
and base pay.
The deflator used to estimate. real earnings for manufacturing and
urban household incomes was the all cities consumer price index. This
is reported in the Economic Planning Board's Statistics Yearbook.
Average price data are collected three times a 'month at nine principal
cities, including Seoul, on 394 commodities and services. Until 1965,
the city consumer price index survey was carried out only in Seoul.
Urban incomes data was also taken from the Statistics Yearbook.
This' data is based on the Family Income and Expenditure Survey
conducted by the Economic Planning Board each month. The survey
covers all households residing in one of Korea's 50 cities, excluding
farm households, 'fish'ing households, single. person households,
foreign households, and households whose income and expenditure
Tie Republic of Korea 581
are not easily identified. Income includes earnings, income from
subsidiary jobs, and other income.
582 Dipak Mazurndar
References
Amsden, Alice. 1987. -Project on Stabilization and Adjustment
Policies and Programs. Country Study No. 14. Helsinki,
Finland: World Institute for Development Economic Research.
. 1990. "South Korea's Record Wage Rates: Labor in,
Late Industrialization." Industrial Relations 29 (1).
Dank of Korea, Statistics Departmnent. Various years. Principle
Economic Indicators, Econtomic Statistics Yearbook. Seoul.
Bank of Korea. 1987. Report on Wage Survey. Seoul.
Collins, Susan, and Won Am Park. 1987. External Debt and
Macroeconomic Performance in South Korea. National
Bureau of Economic Research. New York. Processed.
Dornbusch, Rudiger, and Yung Chul Park. 1987. "Korean Growth
- Policy." Brookings Papters on Economic Activity 2: 389-453.
Grootaert, C. 1987. "The Labor Force Participation of Women in the
Republic of Korea: Evaluation of Policv Issues." Internal
Discussion Paper, Asia Regional Series. Washingtnn. D.C.:
World Bank.
IMF. Various years. International Financial Statistics Yearbook.
Washington, D.C.
Korea. Administration of Labor Affairs. 1980. Report on
Occupational Wage Survey, vol. I. Seoul.
Korea. Economic Planning Board. Various years. Report on Mining
and Manuifacturing Survey. Seoul.
. Various years. Korean Statistical Yearbook. Seoul.
- - . ______*Various years. Year Book of Labor Statistics. Seoul.
Krueger, A., and L. H. Summers. 1986. Reflections on the Inter-
Industry Wage Structure. Discussion Paper No. 1252.
Cambridge, Massachusetts: Harvard Institute of Economic
Research.
Lee, Jong Woo. 1983. "Economic Development and Wage
Inequality in Korea." Unpublished Ph.D. dissertation,
Harvard University.
The Republic. of Korea 583
Lindauer, David. 1984. Labor Market Behavior in the Republic of
Korea:. An Antalysis of Wages and Their Impact on thze
Economy. Staff Working Paper No. 641. Washington, D.C.:
World Bank.
Nishimuzu, M., and S. Robinson. 1984. "Trade Policies and
Productivity Change in Semi-Industrialized Countries."
Journal of Development Economics 16 (1-2): 177-206.
Park, Funkoo, and Torsacia Castaneda. 1987. "Structural
Adjustment and the Role of the Labor Market." Working
Paper No. 8705. Seoul: Korean Development Institute.
Park, Se-I1. 1980. "Wages in Korea: Determination of the Wage
Levels and the Wage Structure in a Dualistic' Labor Market."
Unpublished Ph.D. dissertation, Cornell University.
Sactis, Jeffrey. 1985. "External Debt and Macroeconomic
Performance in Latin America and East Asia." Brookings
Papers on Economic Activity 2: 523-573.
U.N. Various years. Industrial Statistics Yearbook. New York.
.Wfong, Wontack, and Lawrence B. Krause, ed., 1981. Trade and
Growth of the Advanced Developing Countries in the Pacific
Basin. Seoul: Korean Development Institute.
World Bank. 1989. "Korea: Country Economic Memorandum."
Paper No. 7920-KO. Washington, D.C.
13
THAILAND
Chatongphob Sussangkarn
During the last 20 to 30 years, Thailand has been able to achieve a
satisfactory pace. of economic development despite several major
shocks in the world economy, and structural changes in world trade
and exchange rate systems. Through out this period, adjustment
problems associated with short-teun macroeconomic management and
external resource gaps have been relatively mild compared to the
experiences of many other counties. Although the chronic current
account deficit and external debt situation became an important focus
of policy concerns during thc early to rnid-1980s, the on-going boom
in manufactured exports, starting around 1986, has diluted this
concern substantially. The country is now being talked about as the
leader of the next wave of newly industrialized countries.
While overall macroeconomic growth has been good, Thailand
faces important structural adjustment problems concerning sectoral
and regional balance and income distribution. -During the last two
decades, the disparities between agriculture and nonagriculture,
between regions, and between income groups have been widening
noticeably. While industrialization is now proceeding very rapidly and
the share of agriculture in GDP is only around 16 percent, more than
60 percent of the labor force is still primarily engaged in agriculture.
Most of the dynamic and successful export industries are located in
and around the capital city, Bangkok, which is highly developed and
about 20 times larger than the next largest city (Chiang Mai). The key
question for the future is how to maintain the pace of economic
growth while ensuring more balanced development, with the benefits
from development spread more evenly among the population.
585
586 Chlalongphob Sussangkarn
In addition, with the current rapid growth of industries and services,
the compositioni of demand for labor is changing toward more highly
skilled labor. Already shortages of scientific and technical manpower
at the higher level are apparent. In the future, the labor market for all
types of labor with middle to high levels of education is also likely to
get tight. This will create another structural imbalance, one related to
labor of different skill types. This may have an adverse impact on the
competitiveness of Thailand in foreign trade, as well as implications
for the achievement of better distribution of income.
The nature and functioning of the labor markets are clearly related
to the problems of structural imbalance. Changes in the structure of
production lead to changes in the composition of demand for labor
by sector, by location, by skill types. Lags in adjustment in the labor
market, or rigidities, will lead to structural imbalance in the structure
of employment, which may reinforce other imbalances.
Macroeconomic Growth and External Balance
Table 13.1 shows the rates of growth of real GDP from 1960 to
1989. Between 1960 and 1980, the average rate of real GDP growth
was above 7 percent, which is very satisfactory. Breaking down the
Table 13.1 Growth of Real GDP and Real per Capita GNP, 1960-89
(percent)
Real
per capita
Year Real GDP Agriculture Industry Service GNP
1960-65 7.2 4.8 11.5 7.2 4.5
1965-70 8.6 6.0 10.4 9.5 5.8
1970-75 5.6 3.8 7.3 5.6 2.9.
1975-80 7.9 4.0 10.6 8.2 5.3
1980-85 5.6 4.9 5.0 6.3 3.5
1986 4.5 0.2 7.1 4.6 2.6
1987 8.4 -2.0 12.0 10.1 6.5
1988 11.0 8.6 12.8 10.7 9.1
1989 10.0
Note: 1970 figures are based on the New Series of National Accounts.
Source: NESDB (various years).
Thailand 587
period from 1960 to 1985 into five-year intervals, the table shows that
growth was lowest in the immediate aftermath of the two oil shocks:
1970-75 and 1980-85. However, the 5.6 perceat average growth
achieved during these periods is very high compared to other
countries' experiences. The impact of the first oil shock was
cushioned by a boom in commodity prices during the same period.
This helped to increase farm income, and improved the poverty.
situation considerably. In contrast,. after the* second oil shock, all
Thailand's main agricultural commodities suffered a declining price
trend. While Thailand still maintained a satisfactory rate of growth, the
impact on agncultural incomes was severe, leading to significant
increases in poverty incidence. However, after 1986, the economy
began a period of very rapid growth. Driven on by. fast growth of
manufactured exports (currently average 30 to 40 percent per annum)
and tourism, economic growth reached 11 percent in 1988 and was
abouit 10 percent in 1989.
As with most developing countries, Thailand has experienced trade
and current account deficits during the course of its development. In
1975, the ratio of the stock of debt to GDP was insignificant at 2.3
percent. Since 1975, however, this ratio has risen rapidly. The ratio
jumped to 16.1 percent of GDP in 1980, and to 39.0 percent of GDP
in 1985 (figure 13.1). After the second oil shock, as a result of the
recession in the world economy and declining agricultural commodity
prices, the ratio of debt to GDP increased particularly rapidly. In the
early to mid-1980s, this. issue was of major concern to the
government.
After the second oil shock, the Thai government, both at its own
initiative and with assistance from the IMF and the World Bank,
implemented various structural adjustment policies. to control the
external imbalance and associated . foreign debt problems (see
Sahasakul and others 1989). Some taxes were restructured, in
particular, the export tax on rice was progressively reduced and finally
eliminated. The government tried to control public expenditures, and
a major-policy in this connection that had important consequences on
employment was the ceiling on government employme-nt growth of 2
percent per annum, starting in 1983. This significantly affected the
588 Chalongp/aab Sussangkarn
Figure 13.1 Ratio of Stock of Debt to GDP, 1975-88
40
35 .
30
25.
-c 20 v
15
10
O I , I ___ I , 4 , ,- , 1 I I
1975 1977 1979 1981 1983 1985 1987
Year
Sources: National Bank of Thailand and NESDB.
employment prospects of the better educated workers,. who are
predominantly employed by the public sector.
However, the policies that were politically the most difficult to put
through were probably the devaluations of 1981 and 1984. In 1981,
the baht was devalued from about B20.5 to the dollar to B23.0 to the
dollar, and in 1984 a further devaluation took the rate to about B27.0
to the dollar. In both cases the government nearly fell. In contrast,
hardly any political problems arose after 1986 when the baht was tied
mostly to the U.S. dollar as the dollar depreciated substantially relative
to the yen and major L-iropean currencies, which meant substantial
effective depreciation of the baht with respect to the currencies of
Thailand's trading partners. These devaluations and depreciation of
the baht were substantial departures from previous exchange rate
policies. As figure 13.2 shows, the baht/dollar exchange rate had
moved very little between 1950 and 1980.
While the various adjustment policies did not keep the ratio of
foreign debt to GDP from rising, they nevertheless kept the problem
fThailand 589
Figure 13.2 Baht/Dollar Exchange Rate, 1985-88
28
271
26 26
525
=.24.
m23 -
22 -
21.
20 . -
1950 1955 1960 1965 1970 1975 1980 1983 1985 1988
Year
Source: Bank of Thailand.
from getting out of hand. This left Thailand in a good position to take
advantage of changes that occurred in the world, economic
environment starting around 1986. Partly because of the exchange
rate adjustments, partly because of the sharp decline in oil prices in
1986, and partly because of the transition of the. Asian newly
industrializing economies (Hong Kong, the Republic of Korea,
Taiwan, and Singapore) to more skilled and technologically intensive
exporters, the growth of the Thai economy accelerated significantly
after 1986. The growth was mainly driven on by sharp increases in
manufactured exports, which started to grow at a rate of about 30 to
40 percent per annum in 1986. These exports are very diverse, and
include processed food, textiles, shoes, gems and jewelry, artificial
flowers, integrated circuits, toys, and steep pipes. Basically, these
products require relatively unsophisticated labor, and are the kinds of
items through which the current newly industrializing countries had
achieved their success. At the same time, however, Thailand still
remains a major player in agricultural exports. Thailand is the largest
590. C/ialongphob Sussangkarn
exporter of rice and cassava in the world, and ranks among the top ten
exporters of rubber, sugarcane, maize, and fisheries.
The boom in exports led to a downturn in the ratio of debt to GDP,
which declined from about 39 percent in 1986 to about 32 percent in
1988 (figure 13.1). While the external resource gap is starting to
widen again in absolute terms, it remains manageable thanks partly to
the huge inflow of foreign direct investment. The key macroeconomic
issue has shifted more to the question of how to manage the extremely
rapid industrial and urban growth.
Structural Imbalance and Adjustment
Although Thailand has been able to achieve a very satisfactory
pace of economic growth during the past 20 to 30 years, and short-
run adjustment problems have proved to be relatively minor,
nevertheless, sectoral and locational imbalance and income
distribution problems are significant. Disparities between sectors of
production, between regions, and between income groups have been
widening.
The importance of agriculture in GDP has continually been on the
decline. In 1960, the share of agriculture in GDP was about 40
percent. This declined to 31.5 percent in 1975, and virtually halved to
16.7 percent between 1975 and 1986 (table 13.2). With the declining
importance of agriculture in GDP, the decline in the share of
employment in agriculture is not surprising. However, the decline in
the share of employment in agriculture has been much slower than the
share of agriculture in value added. While the share of agriculture in
GDP nearly halved between 1975and 1986, the share of employment
in agriculture only fell from 73 to 67 percent) This obviously meant
a substantial widening of the ratio of value added per head between
nonagriculture and agricultu�c. In 1975, the value added per head in
nonagriculture was 5.9 times higher than that in agriculture, and by
1986 it was 10.0 times higher.
1. This refers to the share of the work force whose main occupation is in
agriculture, and is based on data during the peak agricultural season.
T7hailand 591
Table 13.2 QUP and Employment by Sector, 1975, 1980, and 1986
Category 1975 1980 1986
GDP (rimillions of baht) 298,816 684,930 1,098,362
Agriculture 94,063 173,806 183,037
Nonagriculture 204,753 311,124 915,325
Share of GDP (percent)
Agriculture 31.48 25.38 16.66
Nonagriculture 68.52 74.62 83.34
Employment (millions) 18.182 22.681 26.672
Agriculture -13.270 16.092 17.803
Nouagriculturr 4.912 6.589 8.870
Share of employment (percent)
Agriculture 72.99 70.95 66.75
Nonagriculture 27.01 29.05 33.25
Per capita GDP (bahtlmonth) 1,369.6 2,516.6 3,431.7
Agriculture 590.7 900.1 856.8
Nonagriculture 3,474.0 6,464.7 8,599.9
Ratio of per capita GDP
Nonagriculture/agriculture 5.88 7.18 10.4
Sources: NESDB (varirus years); NSO (various years).
The gap between nonagriculture and agriculture widened
particularly rapidly between 1980 and 1986, when the prices of most
major crops were on a downward trend. The export price of rice
declined an average of 7.0 percent per annum, that of rubber by 5.1
percent per annum, and that of sugar by 9.1 percent per annum. The
result was that the per capita GDP. in agiculture declined in absolute
terms (table 13.2).
Because most agricultural households earn a significant part of
their income from nonagricultural activities-Sussangkamr and others
(1988, table 5.6) estimated that agricultural households earn about 46
percent of their income from nonagricultural activities-the disparity
in per capita income between agricultural and nonagricultural
households is not as large as the disparity in per -capita GDP.
Nevertheless, agricultural households earn about half as much as
592 Clialongplzob Sussangkarn
nonagricultural households, and the gap has been widening along with
the trend in per capita GDP. Table 13.3 shows that in 1975/76, the
ratio of per capita income of nonagricultural households to that of
agricultural households was 2.1. This increased to 2.7 in 1986. Also,
during the period from 1981 to 1986, when crop prices were on a
downward trend, nominal per capital income of. agricultural
households fell.
Naturally the disparity between agriculture and nonagriculture was
reflected in the disparity between urban and rural areas. However, in
the case of Thailand, extreme differences exist in the economic
conditions in Bangkok compared to the rest. of the country. Whereas
the Bangkok region (including the five surrounding provinces)
contain 15.6 percent of the total population in 1985, it accounted for
45.5 percent of total GDP. Of the GDP from industry, 63.7 percent
originates from the Bangkok region. For the most dynamic exporting
industries such as textiles and garments, the proportions originating
from the Bangkok region rise to over 90 percent. Per capita GDP in
Bangkok is more than sevenfold that of the northeast (the poorest
region), and about 2.8 times higher than that of the central region (the
second richest region). Taken by itself, the Bangkok region is already
a "newly industrializing country."
The sectoral and regional disparities are reflected in the distribution
of household income. Table 13.4 shows. the shares of incomes
received by various quintiles of households, with the top and bottom
two deciles also separated out. Between 1975 and 1986, the income
Table 13.3 Mean per Capita Income, Agricultural and Nonagricultural
Households, Selected Years, 1975/76-86
Baht per month
Nonagricultural:
Year Agriculutral Nonagricultural agricultural
1975176 247 513 2.1
1981 503 1,154 2.3
1986 481 1,312 2.7
Source: NSO (1975/76, 1981, 198).
Thailand 593
Table 13.4 Income Share by Quintile of Population, Selected Years,
1975/76-86
(percentage of total)
Quintile 19 75/76 1981 1986
First 49.3 51.5 55.6
Top 10 percent 33.4 35.4 39.1
Second 10 percent 15.9 16.1 16.5
Second 21.0 20.6 19.9
Third 14.0 13.4 12.1
Fourth .9.7 9.1 7.9
Fifth .6.0 5.4 4.5
Second bottom 10 percent 3.6 3.3 . 2.7
Battom 10 percent 2.4 2.1 1.8
Source: NSO (1975/76, 1981, 1986).
share of the richest 20 percent of households increased from 49.3
percent to 55.6 percent while that of the poorest 20 percent of
households declined from 6.0 percent to 4.5 percent. The shares of all
quintile below the top quintile continually declined between 1975/76
and 1986. Of the top two deciles, the second decile. made only a slight
gain. Only the top decile increased its share of income substantially.
Between 1981 and 1986, apart from the worsening in the
distribution of income, absolute poverty also increased. Because most
of the population depended for their livelihood on agriculture, the
decline in crop prices led to a large increase in the share of the
population living below the poverty line, from 23 percent in 1981 to
29.5 percent in 1986 (table 13.5). As the table shows, this was a
reversal of the excellent progress made in poverty alleviation since the
late 1960s.
After 1986, agricultural prices picked up substantially, which
helped to reduce poverty. The percentage of the population living
below the poverty line declined to 25.2 percent in 1988. This wvas,
however, still higher than the percentage below the poverty line in
594 Chalongphob Sussangkarn
Table 13.5 The Incidence of Poverty, Selected Years, 1968/69-1988
Percentage of
population below
Year the poverty line
1968/69 39.0
1975/76 30.0
1981 23.0
1986 29.5
1988 25.2
Note: Figures up to 1986 are calculated from various socioeconomic surveys
conducted by the NSO. The 1988 figure was based on a Thailand Development
Research Institute simulation using the THAM2 model. Sanitary districts are
classified as rural.
Sources. Hutaserani and Jitsuchon (1988); Meesook (1979).
1981. The income distribution situation since 1986 still shows a slight
worsening trend. Although crop prices increased, the boom in
manufactured exports also led to rapid increases in the incomes of
those in urban areas, particularly those in the Bangkok area. To some
extent, the exchange-rate policy and the agricultural taxation policy
counteracted the worsening trend of poverty and income distribution
between 1981 and 1986.
The devaluations in 1981 and 1984 increased farm. incomes more
than would have occurred without these devaluations. Although
Thailand does have some market power in the major export crops
such as paddy, econometric estih-ates generally find the foreign
demand curve to be fairly elastic, and that devaluation would increase
domestic farm incomes. Thailand has also had a long history of
export taxation on rice.2 This included export tax, a rice premium,
and export quotas on rice exports.. The estimated burden on rice
farmers is some 30 percent of the fo!reign price of rice in 1980
(Siamwalla and Setboonsarng 1987). As rice prices fell on the world
market, the rate of taxation gradually declined, and in 1986 all taxes
on rice exports (explicit and implicit) were removed. This helped to
2. There are also smaller taxes on rubber still present today.
ThralJand 595
keep the domestic farm gate price from falling in line with world
prices. Nevertheless, these policies could not reverse the trend of
falling farm incomes and worsening inequality.
Labor Markets and Structural Adjustments
The problems of imbalance, poverty, and income distribu-tion
described above are clearly related to adjustments occurring in labor
markets in response to structural changes .on the production side. This
section discusses the structure of the labor market in Thailand and the
relationship between the labor market and the structural imbalance
and disparity issues.
General Employment Situation
As already mentioned and indicated in table 13.2, most of those
employed in Thailand work in agriculture. Currently, over 65 percent
of all employed individuals have their main occupation in agriculture.
Thus, in terms of the employment structure, Thailand is basically an
agrarian society. Aggregate indicators do not reveal any serious
employment problems. Open unemployment is generally very low,
around 1 percent of the work force, although the rate appears to be on
the rise (figure 13.3), presumably partly due to the recession in the
first part of the 1980s. The low open unemployment rate is not too
surprising. Most people in Thailand are either self-employed or
unpaid family workers, mainly in agriculture. While the share of self-
employed and unpaid family workers in total employment has been
gradually falling (table 13.6), in 1988 they still accounted for over 70
percent of all employed people. Also, as in many other developing
countries, the informal sector offers many employment opportunities.
Here barriers to entry are low, so most of those who really want to
work carn find something to do.
Underemployment has also been generally low in Thailand. Table
13.7 shows that in 1977, about 4 percent of the employed work force
worked less than 20 hours per week. This percentage declined to 1.3
percent in 1984. Despite regional variations, no region had a really
serious underemployment problem. The underemployed are almost
equally divided between men and women, and over 80 percent of the
underemployed are self-employed or unpaid family workers. When
596 Chlalongp/iob Sussangkarn
Figure 13.3 Open Unemployment Ratus, Selected years, 1977-86
1.4
1.2
1.0
i0.8
0.6
0.4
0.2
0
1977 1980 1983 1986
Year
Table 13.6 Employment by Work Status, Selected Years, 1978-88
Status 1978 1983 1988
N'umber of people
Public employees 1,020,528 1,780,834 1,839,900
Private employees 3,291,002 4,470,462 6,176,700
Employers 282,424 253,586 353,500
Self-employed 6,544,122 7,456,259 8,550,600
Unpaid 10,669,761 11,222,386 12,543,300
Total 21,807,837 25,183,527 29,464,000
Percentage share
Public employees 4.68 7.07 6.24
Private employees 15.09 17.75 20.96
Employers .1.30 1.01 1.20
Self-employed 30.01 29.61 29.02
Unpaid 48.93 44.56 . 42.57
Total 100.00 100.00 100.00
Source: NSO (various years).
Tlhailand 597
Table 13.7 Percentage of Employed Working Less than 20 Hours per
Week, July-September, 1977 and 1984
Area 1977 1984
North
Municipall nreas 1.69 0.65
Non-municipal areas 1.63 1.55
Northeast
Municipal areas 2.05 1.57
Non-municipal areas 6.2 0.65
South
Municipal areas 3.8 0.94
Non-municipal areas 7.49 3.49
Central
Municipal areas 1.7 0.96
Non-municipal areas 1.95 1.22
Bangkok 1.07 0.76
Whole country .3.97 1.27
Source: NSO (1977, 1984).
asked whether they desired more work, about 64 percent of the
underemployed stated that they did not want more work (NSO Labor
Force Survey 1984).
Labor Market Problems and Structural Adjustmnent
Three basic types of adjustment problems affect the labor market,
namely:
The seasonal nature of agricultural production results in a
short-term problem: in the dry season, there is a lack of
agricultural activities in nonirrigated parts of the country, and a
very high proportion of the labor force becomes seasonally
unemployed.
Lags in labor movement in line with changes in the production
structure result in sectoral and locational disparities.
598 Chlalongplob Sussangkarn
Mismatching of supply and demand by educational types is a
problem mostly relevant to the middle to upper educational
levels.
SEASONAL UNEMPLOYMENT. As most of those employed are dependent on
agriculture and work in agriculture has a seasonal nature, seasonal
unemployment affects many people. Table 13.8 shows that three to
five million people are seasonally unemployed each year. The
northeast, the poorest region, is the worst affected, with over 30
Table 13.8 Seasonal Unemployment Numbers and Rates, by Region,
1979-85
Year/category Northj Northeast South Central Total
1977
Seasonal unemployment 1,065,740 2,306,910 53,660 537,310 3,963,620
Seasonal rate (No) 24.51 3032 2.34 13.78 21.84
1978
Seasonal unemployment 863,930 2,673,870 38,850 445,410 4,022,060
Seasonal rate (Vo) 19.55 35.20 1.54 11.09 21.68
1979
Seasonal unemployment 985,570 2,823,780 128,080 431,890 4,369,320
Seasonal rate (%) 21.60 36.18 5.38 10.86 23.34
1981
Seasonal unemployment 1,497,200 3,274,970 48,440 749,980 5,570,5910
Seasonal rate (%Si) 28.98 39.48 3.01 16.48 28.39
1982
Seasonal unemployment 1,482,030 3,442,910 71,470 460,620 5,457,030
Seasonal rate (%) 27.35 40.13 2.57 9.75 25.38
1983
Scasonal unemployment 992,420 2,775,220 75,320 573,810 4,416,770
Seasonal rate (%) 20.05 35.90 2.89 13.00 22.42
1984
Seasonal unemployment 675,410 2,770,270 77,880 244,020 3,767,580
Seasonal rate ({6) 12.79 33.26 2.87 5.37 18.05
1985
Seasonal unemployment 946,680 2,771,820 121,390 348,620 4,188,510
Seasonal rate (So) 17.41 31.47 4.29 7.24 19.14
Source: NSO (various years),
Tlailand 599
percent of the work force seasonally unemployed. About 40 percent
of the seasonally unemployed are self-employed farmers who may
not have the opportunity to find off-farm work during the dry season,
particularly if this involves seasonal migration. Of the rest, most are
unpaid family workers, and consist mainly of the relatively young
(14-24 years old) and women.
A study by Bertrand and Squire (1980) suggests-that most of the
seasonally unemployed are voluntarily unemployed, with the younger
workers and females drawn into the labor force only to help in the
peak season. However, more recent examinations by Phongpaichit and
Baker (1984) and Sussangkarn (1987) contridict this view. The latter
analyzed a data set on seasonal migration and showed that most of the
seasonally unemployed would like to find work in the dry season,
particularly the younger workers and women. Many, however; cannot
find jobs or do not know how to look for seasonal migration jobs.
Seasonal unemployment is an important problem for the rural
population. It is intrinsically tied to the seasonal nature of agricultural
activities. The situation will depend on changes in the cropping pattern
in agriculture, as different crops require different amounts of labor at
various times of the year. It will also depend on the possibilities for
extending irrigation into currently rain-fed areas.
LABOR MOVEMENT AND CHANGING PRODUCTION STRUCTURE. Another important
adjustment problem is the imbalance between the sectoral and
locational distribution of production and employment. This has led to
growing disparities between agriculture and nonagriculbire and
between different regions. The case of Thailand seems somewhat
unusual when compared to other countries. Comparing the ratio of
the share of employment in agriculture to the share of agriculture in
GDP for a number of countries, Thailand stands out as having one of
the highest ratios (table 13.9). This ratio indicates the difference
between the value added per head in agriculture and nonagriculture,
and is a rough indicator of the disparity between agriculture and
nonagriculture.
Part of the explanation for the very large ratio of the share of
employment in agriculture to the share of agriculture in GDP may be
600 Chalongphob Sussangkarn
Table 13.9 Agricultural Indicators for Selected Asian Countries
(1) (2)
Percentage of Shlare of
the laborforce agriculture
in agriculture in GDP Ratio
Country (1980) (1982) (1)1(2)
Bangladesh 7 5 47 0.63
China' 74 37 0.50
India 70 33 0.47
Indonesia 57 26 0.46
Korea, Rep. of 36 16 .0.44
Malaysia 23 23 1.00
Myanmar 53 48 0.90
Pakistan 55 31 0.56
Philippines 52 22 0.42
Sri-Lanka 53 27 0.51
Thailand 71 22 0.28
Source: World Bank @various years).
because many of those who have their main occupations in agriculture
also work in nonagricultural occupations throughout the year, but
particularly in the off-season. While this may be true to some extent, it
is not the main explanation of the very high ratio in Thailand
compared to other countries. First, those engaged in agriculture in
other countries are also likely to be engaged in nonagricultural
occupations. Second, the nonagricultural opportunities are not
abundant everywhere, as evidenced by severe seasonal unemployment
problems in many parts of the country. Further, the comparison of
incomes of agricultural and nonagricultural households in table 13.3
includes household incomes from all sources, and this shows
agricultural households falling behind nonagricultural households in
line with widening value added per head between agriculture and
nonagriculture.
Two main reasons explain the difference between the ratio of
employment in agriculture and the share of agriculture in GDP in
Th ailand 601
Thailand compared to other Asian countries. The first is the past
ready availability of forest areas that could be converted to arable
land. These were the main destination for migrants from the rural
areas irn response to population pressure up ur'til about 1980. Instead
Of migrating to urban areas, rural migrants would go to the forest
areas (often illgally), settle down to cultivate the land, and in effect
take ownership. The main migration pattern in the 1960s and 1970s
was rural/rural. As a result, until.the late 1970s, the rate of expansion
of cultivated land in Thailand was 3 to 4 percent per annum (figure
13.4), and was greater than the rate of population growth in the rural
areas.
A second reason is the very high proportion of Thai farm
households who are owner cultivators.. In 1981, the NSO
Socioeconomic Survey gave this proportion as 83.3 percent of all
Figure 13.4 Total Cultivated Areas of Major Crops, 1972-84
100
90
~8O
-70. 7
60 -
50 , , , , , , , I . . . .
1972 1974 1976 1978 1980 1982 1984
Year
* 1 rai =1,600 square meters
Note: Rubber and second rice crop not included.
Source: Ministry of Agriculture and Cooperation.
602 Chialongpitob Sussangkarn
farm households. This is likely to be a factor working against large-
scale migration into urban areas. The mar-ket for the sale of land in
rural areas is thin, and thus owner cultivators who wish to sell their
land and migrate to the urban areas may only get a low price for their
land. This would increase the opportunity cost of migration.
Migration from self-cultivating households would therefore be limited
to a few family members such as sons or daughters.
A factor that would reinforce the above reason is that many of the
so-called owner cultivators do not, have title to their land. The
migrants who went into the forest areas to open up new land were
taking possession of the land illegally, even though the authorities did
not really try to enforce the law. The result is that about 30 percent of
private land is not legally documented. This would make it even more
costly for a farmer to abandon the land and migrate out of the rural
areas (see Chalamwong and Feder 1988).
Until the end of the 1970s, the farmers' migration to open up new
agricultural land was logical. Plenty of land was theoretically still
available in the forest areas, and crop prices were high and rising.
However, once crop prices started to decline in the early 1980s, it was
difficult for the farmers to move out of agriculture because of the thin
market for land or the lack of legal title to the land.
The high opportunity cost the farmers faced were they to- migrate
into the urban areas led to long lags in the adjustment of the
employment structure to the produiction structure. However, some.
adjustments have been occurring, arni the pace of adjustment is likely
to accelerate, given the rapid indur,zrialization that is currently taking
place in Thailand. The benefits of migrating to the urban areas, where
the demand for semi-skilled workers is growing rapidly, are getting
larger. Data on the rate of growth of the urban population show
clearly that rurallurban migration has been increasing since 1980.
Between 1960 and 1980, the growth of the urban population was
remarkably steady at about 3.5 percent per annum. Between 1980 and
1985, however, the rate jumped to 6.6 percent per annum (figure
13.5). This was the time when crop prices were falling and when the
availability of new land for agricultural expansion had become scarce.
Thus, there are clearly labor market responses to changes on the
production side. However, the responses could not keep pace with
Thailand 603
Figure 13.5 Growth per Annum of Urban Population, 1960-85
6.6
6
5
c4 3.7 3.
-gL,3. ~ ~ 3. 3. 3.5-1
2
0
1960-65 1965-70 1970-75 1975-80 1980-85
Period
Source: World Bank (various years).
changes that were occurring in the production structure. This led to
problems of imbalance and income disparities as discussed earlier.
The current industrial boom will likely lead to even more rapid
urban growth. While recent migration data are incomplete (pending
results of the 1990 census), migration rates into the Bangkok area
have probably accelerated. If Thailand's pace of growth remains
similar to past patterns in the newly industrializing Asian countries,
population movements out of agriculture and into urban areas are
likely to be the key demographic transition over the medium to long
term.
THE MISMATCH OF EDUCATED MANPOWER. While the overall open
unemployment rate in Thailand is very low, the rates of open
unemployment among educated people are quite high, particularly
those with a vocational education. This was an important problem
around the mid-1980s, and was related to the cutback in government
employment growth to 2 percent per annum. Cuirrently, howVever, due
to the economic boom, the problem of educated open unemployment
is much less severe, and instead the problem of shortages of
engineering and scientific manpower has gained importance. These
604 CAalongphob Sussangkarn
problems of a mismatch of the demand for and supply of educated
manpower relates to structural changes on the demand side, the
functioning of the labor market, and the education system.
Table 13.10 shows that the open unemployment rate of those with
a primary education or below was about 0.5 percent in 1984 and
1986. At the vocational level, however, the rate was over 10 percent in
both years, while at the other levels, the rates were some 3 to 4 percent.
Until the econonmic boom that started after 1986, the number of
educated open unemployed had been increasing rapidly. The two
groups with the highest growth of open unemployment were those
with a university education (averaging 14.3 percent per annurm
between 1977 and 1986), and a vocational education (averagirg. 22.'7
percent per annum), which was the group with the most serious open
unemployment problem. It had the highest open unemployment rate
and also the fastest growth in the number of open unemployed. In
1986, the number of unemployed with a vocational education was
about equal to the sum of the unemployed with a secondary
education, university education, and teacher training combined.
Four basic reasons explain why the educated open unemployment
problem was getting worse during the early 1980s. First, the supply of
the better educated work force had been rising rapidly: since 1980,
workers with a university and vocational education have been
increasing at about 15 percent per annum. This was much faster than
the overall growth in the labor force.
Table 13.10 Open Unemployment Rates by Education, 1984, 1986,
1988
(percent)
Level of education 1984 1986 1988
Primary or below 0-42 0.56 0.38
Secondary 3.19 3.55 2.47
Vocational 10.22 10.86 5.31
University 4.19 4.58 2.87
Teacher training 3.19 3.96 1.62
Total 0.97 1.26 0.83
Soun e: NSO (various years).
T7hai/and 605
Second, a fundamental change occurred in the main source of
demand for better educated workers. Up until 1983-84, the main
absorber of better educated workers was the public sector. Figure 13.6
shows the importance of the government for the employment of the
better educated. For those with less than a primary education, the
proportion in government employment was only about 0.5 percent.
For completed primary education, the proportion was still low at 2.1
percent. As figure 13.6 shows, the importance of the public sector
increased rapidly for higher levels of education.
From the mid 1970s to 1984, government employment growth
amounted to about 10 percent per annum. This was faster than the
growth in the size of private employees, or the numbers of self-
employed and unpaid family workers. The high level of growth of
government employment created much needed jobs for the rapidly
Figure 13.6 Share of Government Employment by Education, 1984
Total 6.4
University 56.3
Teacher 84
Vocational 41.4
Secondary 22.1
Elementary 2.1
Less than primary 0.5
0 20 40 60 80 100
Percent
Note: The high number of teachers is because the government is the country's main
supplier of education.
Source: NSO (1984).
606 Chtalongphob Sussangkarn
increasing pool of workers with relatively high levels of education.
While this did not slow down the growth in educated open
unemployment completely, it helped to contain the problem. However,
as part of the structural adjustment policies adopted in the early 1980s,
the growth of government employment was cut drastically to only 2
percent per annum. As discussed earlier, in the early to mid-1980s the
saving-investment gap and the rapidly rising foreign debt burden were
important issues for Thailand. As civil service salaries were taking up
more and more of the budget (48 percent of all government revenues
by 1984), the government imposed an upper limit on civil service
growth to 2 percent per annum, which is still the case for most parts of
the civil service today. The consequence was that educated workers
could no longer rely on the public sector as the main employer.
The third reason for rising open educated unemployment is that the
education system cannot respond quickly to changing needs in the
labor market. The skill mix the civil service requires is very different to
that for private sector employment needs. Even with the current
economic boom, there is an oversupply, of graduates in the humanities
and social and political sciences, while at the same time there is a severe
shortage of graduates in the more technical and scientific disciplines.
The education system is mostly public and highly bureaucratic, and
reducing the size of any department, despite clear needs to reallocate
resources among disciplines te keep up with changing labor market
needs, is very difficult.
The fourth reason concerns the structure of the labor market.
Previous analyses of the labor market point to the existence of labor
market segmentation, with wages not fully responsive to demand and
supply (Sussangkarn 1987). Relative wage data by education level
(figure 13.7) do not reveal any clear falling trend in the relative wage
*of the vocational and university groups, even though their open
unemployment rates had been rising the most rapidly. Econometric
estimations. of a segmented labor market model revealed high wage
differentials between the formal sector (the public sector and the large
private firms) and the informal sector, and the better educated find that
being unemployed and waiting for an opening in the formal sector is
more worthwhile than going to work in the informal sector.
Thailand 607
Figure 13.7 Relative Wage by Educational Level, 197884
(index, primary education =00)
3S51
250
100
50 --
50
O00. . . . _ _.. , 4 ,
1978 19; 9 19BO 1981 19S2 1983 19B4
Yea.r
- - University - -Teacher -------- Vocatimal
* Secondary - -Primary
Source: NSO (various years).
The econometric study found that returns to education above the
primary level were zero or negligible in the informal sector, while
returns to education in the formal sector were large. Thus, the wage
differential between the formal and informal sectors increased with the
level of education. For a male, nonmigrant, private employee in
Bangkok aged 35, the estimates predicted a small formal/informal wage
differential at the primary educational level. This differential rose to
over 400 percent for those with a university education.
The proportion of those working in the formal sector increases with
educational level. In 1988, of those with a primary education and
below, only 6.6 percent worked in the formal sector. The ratio quickly
rose to 31.6 percent and 35.5 percent for those with a lower and upper
secondary education, respectively. For those with a vocational
education it was 63.2 percont, and the ratio reached 88.4 percent for
those with a university education. Furthermore, most of the educated
608 Chalongphob Sussangkarn
unemployed came from relatively well-off families who could finance
their periods of unemployment.
With the economy experiencing double digit growth, the situation in
the labor market is changing rapidly. As table 13.10 shows, by 1988
the open unemployment problem for the better educated had improved
tremendously. The problem now is a severe shortage of university
educated scientific and technical manpower, especially engineers. The
wages of better educated workers have increased rapidly. The larger
private firms are bidding workers away from the public sector and the
small- and medium-size private firms. The situation is unlikely to
improve for some time given the continuous increase in foreign direct
investment from Japan, Hong Kong, Taiwan, and elsewhere.
At the vocational level, the technical manpower situation is still one
with general excess supply, although shortages are apparent in specific
industrial fields. Part of the, reason is that many of the booming
industries rely more heavily on semi-skilled workers. Another reason is
the quality of output at the vocational level. Coordination between the
vocational schools, which are mostly public, and the private companies
is insufficient. The machinery vocational students learn on is mostly
outdated, and the schools do not have enough money to keep pace with
the rapid t;chnological advances.
Adjustment Issues for the Future.
Thailand's policy focus should be to maintain the pace of economic
growth while enrsuring more equitable sharing of the benefits of growth
among the population. To achieve this goal the government needs to
tackle several human resource and labor market problems in line with
expected changes in the production structure.
The government must anticipate and plan for the expected rapid
growth in rural/urban migration. Bangkok is already very congested
and the necessary infrastructure lags far behind actual needs. Currently,
the eastern seaboard region between Chonburi and Rayr:ng, some 100-
200 kilometers from Bangkok, is developing rapidly. This will help to
divert some of the industrial expansion and population movement away
from the Bangkok area. However, the eastern seaboard's urban centers
need careful management and planning. The social infrastructure is
already lagging behind the population movement.
Thtiland 609
Whereas Thailand has a ready siupply of relatively cheap labor
available, most of the workers have just a primary education. In
Thailand, more than half of those who finish the six years of
compulsory education drop out of the formal education system.
Currently, 75 percent of the work force have only a primary
education. The gross enrollment ratio at the secondary level is only 30
percent. This is very low compared te the newly industrializing Asian
countries, and also lower than many countries in the region, who are
likely to be Thailand's main competitors in the future (figure 13.8).
For the future, Thailand needs to upgrade the technological and
skill base of its industries and services, including agriculture, which is
diversifying into higher value added products that require greater
levels of skill and knowledge. This. will require a more highly
educated work force. Current low enrollment at the secondary level
Figure 13.8 Gross Enrollment Ratios, Selected Asian Countries, 1984
Hong Kong
Indonesia
Malaysia
Philippines
Singapore
Republic of Korea
Taiwan
Thailand
0 20 40 60 80 100 120
Percent
* Tertiary 0 Secondary EU Primary
Source: World Bank (1987).
610 C/talong phol Sussangkarn
means that the market for workers with more than a primary education
will begin to get tight over the next few years. When this happens, the
basic wages of the modern manufacturing sector will begin to rise
rapidly, and Thailand's competitiveness will be eroded despite its large
pool of low wage workers. Such a development may affect the.
sustainability of the current growth trend, and also lead to undesirable
income distribution consequences, because the majority of workers
with just a primary education will continue to fall further and further
behind those with a better education. Policies. are needed to increase
secondary enrollment and to develop effective training programs to
upgrade the skills of those who have left the formal education system.
At the higher education level, the education system needs to be
made more flexibile so that it can respond better to the labor market's
skill requirements. Fields of study that produce an excess supply of
workers need to be trimmed, while those where more graduates are
needed should be allocated more resources to expand. Private
universities and colleges need to be given more flexibility, and more
cooperation is needed between the public and private sectors to develop
study programs that will meet the labor market's needs. In the future,
public education, the dominant source of education, needs to cater to
requirements for employment in. the private sector rather than to
employment in the public sector as in the past.
References
Bertrand, T. J., and L. Squire. 1980. "The Relevance of the Dual
Economy Model: A Case Study of Thailand." Oxford
Economic Papers.
Chalamwong, Yongyuth, and Gershon Feder. 1988. "The Impact of
Landownership Security: Theory and Evidence from
Thailand." World Bank Economic Review 2(2):187-204.
Hutaserani, S., and S. Jitsuchon. 1988. "Thailand's Income
Distribution and Poverty Profile and their Current Situation."
Paper presented at the 1988 Thailand Development Research
Institute Year-End Conference on Income Distribution and
Long-Term Development, December.
Thiailand 611
Meesook, Oey Astra. 1979. Incomne, Consumnption atnd Poverty in
Thailantd, 1962/63 to 1975/76. Staff Working Paper 364.
Washington, D.C.: World Bank.
NESDB (National Economic and Social Development Board).
Various years. National Income of Tlailand. Baingkok.
NSO (National Statistical Office). Various years, 1977, 1984. Labor
Force Survey. Bangkok.
1 1975/76, 1981, 1986. Socioeconomic Survey.
Bangkok.
Phongpaichit, P., and C. J. Baker. 1984. "Bertrand's Choice and
Seasonal Unemployment Reccinsidered." Bangkok, Thalland:
Chulalongkorn University, Faculty of Economics. Processed.
Sahasaklul, C., N. Thongpakde, and K. Kraisoraphong. 1989. Lessons
from the World Bank's Experience of Structural Adjustment.
Loans (SALs): A Case Study of Thailand. Bangkok, Thailand:
Thailand Development Research Institute.
Siamwalla, Ammar, and Suthad Setboonsarng. 1987. "Agricultural
Pricing Policies in Thailand: 1960-1985." Bangkok,
Thailand: Thailand Development Research Institute,
Agriculture and Rural Development Program.
Sussangkarn, Chalongphob. 1987. "The Thai Labour Market: A
Study of Seasonality and Segmentation." Paper presented at
the International Conference of Thai Studies, Australian
National University, Canberra.
Sussangkarn, Chalongphob, Pranee Tinakorn, and Tienchai
Chongpeerapien. 1988. "The Tax Structtire in Thailand and
its Distributional Implications." Paper presented at the
Thailand Development Research Institute Year-End
Conference, Cha-am, December.
World Bank. Various years. World Development Report. New York:
Oxford University Press.
INDEX
(Page numbers in italics indicate material in figures or tables)
Added worker effect, 17, 18 in, 69-71; economic indicators for,
Adj ,1; fiscal proxies to mimic 65; economic policy (late 1970s),
Adjustment, 1;fsa rxe ommc 68; employment in manufacturi'ng,
devaluation in, 266; gradual ap- 78; employment by sector, 79; ex-
proach to, 53; labor market flexibil- change rate policy and wages in, 84;
ity and success of, 143; political failure of short-term stabilization
sustainability -of programs for, 207; famlure of immigration to
public sector employment and, 46; program ini, 68; immigration to ur-
public spending cuts in, 266; resnlts tand povert in, 9-; in flation
of (for middle income countries), tion 67, 68, 70,84, 93; inward-on-
143; sources of problems with, 483; ent got sre of, 74; labor
successful and less successful, 44- ented growfl strategy of, 74; labor
45; timing of, 12; varieties of expe-
rience with, 8-9, 12, 14; wage flexi- patio' rates in, 85-86, 87 8& labor
bility and, 22; see also individual market institutions in, 80-81, labor
country entrdes a market perforrnance in adjustment
cO.ntry - fl ,*WS in, 82, 84; labor market segmenta-
Adjustment policies: quantification and, tion and increasing inlbrmality in,
12; timing of, 13; see also individ- 79-80; labor scarcity in, 84; lessons
ual country entries for policy makers from, 94; macroe-
Agricultural indicators for Asian coun- conomic policies in, 62-64; military
tries, 600 government's objective in, 67; min-
Agriculture: -in Argentina, 63, 78; in imum wages in, 81; overvalued ex-
Bolivia, 101, 112, 120; in Costa change rate policy of, 63; policy fa-
'Rica, 218, 225, 227, 231-32; -in ronng nontradables and, 77; politi-
Cote d'lvoire, 268-69, 28182, 288- cal and economic developments
89, 291 n.10; debt crises and shift (1970s), 64; 67-6p ; population
out of, 33; in Egypt, 332, 335, 345, growth in, 76; publc; sector em-
349; in Ghana, 359, 397, 401; in ployment in, 89-90; real output
Kenya, 406, 408, 415, 435, 436 variability of, 68-69; real wagcs and
foreign terms of trade, 66; real wages
Argentina: adjustnient frustrated in and per capita GDP, 66; real wage
(1987-88), 71-72; aggregate em- -trends in, 74-75; regional and sec-
ployment in, 84-87, 89; agricultural toral labor allocation-in, 76-78; ru-
employment in, 78; balance of pay- ral population decline in, 76; sectors
ment crisis in, 63; crisis of 1980s in the economy of, 63; suggested la-
613
614 Labor Markets in an Era of Adjustment
bor market policies for, 93; trade and,'109; promotion of small scale
policies of, 62-64; trade reforms in, enterprise and, 107; stabilization af-
67; unemployment in, 88, 89-90; ter 1985 and, 105-11; stabilization
urban population growth in, 76; ur- and the labor market and, 119-20,
ban unemployment in, 85; wage de- 121, 122-25; tariffs in, 106; unem-
termination in, 81; wage policy re- ployment reduction program and,
sults in, 78; women in labor force 108; wages structure and bonus
of, 90-91; see also Buenos Aires changes and, 108
Banque Centrale des Etats de 1'Afrique de Bolivian labor market, 10 7-8, 111;
lOuest (BCEAQ), 261, 265 agriculture and, 112, 120; domestic
Berry, A., 240, 241, 250 service and, 118; indicators for,
113; labor force indicators by city
Bolivia, 102; agriculture and agrarian re- and, 128; migrant labor in, 118-19;
form in, 101; best export potential ramifications of clanges in, 135-36;
for, 101; composition of the unem- i d
ployed in, 17; economic collapse of 117-18;l sifeencts and, 114-15 116a,
(1984(85), 104; Emergency Social
Fund and employment in, 109,119- .120; self-employed workers in, 123;
20; GNP in, 109;- inflation in, 100, sources of data on, 137-38; urban la-
109; informal sector in, 117, 122- bar.market idicators by gender and,
24; internal conflicts in, 103; 130,431; women in, 115
macroeconomic indicators for, 100; Bolivian labor market adjustment impli-
macroeconomic problems of, 104; cations: for income distribution,
migration within, 112; mine nation- 125-26, 127; for long-run economic
alization in, 101; political history ' vgrowth,m1, 1233, 134; for regional
of, 103; problems from 1952 indus- development, 127, 128, 129; for
trialization strategy in, 101; real - women, 129-30,131,132
wages in, 120, 121, 122, 125; r;-' Bonuses, 45; in Bolivia, 108; in Korea,
sults of fiscal deficit of, 116; sav- 552, 560
iwgs in, 101; taxes and tax base in, Brazil: during the adjustment period,
104-5; trade balance in, 110; sug- 147-52; balance of payments of,
gested research on, 134; unusual 149; debt crisis in, 146; employ-
problems of, 101 . ment and wages in the urban formal
Bolivian adjusiment program, 99-101; sector of, 145; exports of, 144, 14;9;
Emergency Social Fund changes and, failure'to adjust macroeconomic bal-
. 108; foreign exchaige market re- ances in, 143, 146; growth in
forms in, 105; import prohibitions (1970s), 144; heavy borrowing by,
_and licensing requirements abolished 144-45; import decline in, 149; in-
in, 106; industrialization strategy fant mortality rate in, 166; inflation
in, 101; labor market changes and, in, 150, 151, 152, 163; macroeco-
107-8; market liberalization and, nomic results in second half of
107; moves to cut govemment ex- 1980s in, 148, .151-52; manufac-
penditures and raise taxes and, 106; tured export development in, 144;
obstacles to success of, 110-11; new mortality rates in, 166; the poor.and
econoalc policy (NPE) and, 105-8; adjustment in, 164-66; poverty in,
. policies to reactivate the economy 144, 145, 162-63, 164-66; prob-
Index 615
lems with data in, 164, 165, 167; strategy and, 157; minimum wage
public sector and adjustment in, 146; decrease and, 161; occupations of
regional differences in distribution heads of poor households and, 154;
of formal sector workers, 153-54; poor macroeconomic results of labor
social mobility in, 145; unemploy- bargaining system and, 158; rise of
ment in, .159; worker eligibility for labor unions and, 155-59; segmenta-
state benefits in, 152, 153; workling tion of, 152-55; strikes and, 157;
class expectations in, 146 wage control and indexation results,;
Brazilian adjustment (1980s): balance of 149; wage guidelines flaunting and,
payments during, 149; consumption 156
during, 151-52; control of imports Brazilian macroeconomic policy and its
during, 148 n.2; Cruzado Plan and, outcomes during adjustment, 147
150-51, 162-63; debt moratorium Brazil's adjustment problem, 144-46
and, 151; failure of, 143,'146;- im- - -
and,s 151; failuresof,r14, 146; im Budget deficits: in C6te.d'lvoire, 260; in
ports and exports during, 149; in- Ghana, 361-62; in Kenya, 409, 412;
come. distribution and, 162, 164; in- in Malaysia, 465-66, 467, 468,
flation and, 150, 151, 152, 163; in- 4- ways offiacn, 467
* , . ~~~~~~~472-73; ways of financing, 467
terest on government's internal debt
and, 164; internal balance and, 167; Buenos Aires, income distribution in,
investmnqut and, 151, 152; labor 95
movement's voice in policies and,
163; macroeconomic indicators dur-
ing, 147; minimum wage decrease
and, 161; policies leading to infla- Chile, 169-70; deflation in, 23, 25; em-
tion during, 149; political stalemate pirical definitions of variables in
(1985-89) and, 150; the poor and, study of, 210-11; implications of
164-66; private sector savings ab- case study of, 208-9; model for ad-
sorbed by public sector deficits, justment policies and labor market
164; savings. and investment response in, 197-207; rigidities in
changes.during, 148; shock stabi- labor market of, 169; unemployrment
lization program and, 150-51; suc- in, 14, 204
cess in trade and external balance Chilean adjustment program's labor
and, 166-67 market effects, 197-207; on average
Brazilian labor market in the adjustment real wages, 181; distribution results
process: composition of labor force and, 195-97; on earnings functions,
by sector and, 152; earnings in for- 194-95; on employment, 181, 182-
mal and informal sectors and, 154; 83, 184-86, 195; private rate of re-
earnings of heads of households and, turn to schooling and, 195; privati-
155; flexibility in, 143, 159, 167; zation and, 197; rate of return to ex-
government control over labor perience and, 195; on real wage per-
unions and, 155-56; government formance, 187-90; on relative
credibility and, 167; growth of labor wages, 192-94; social indicators
unions and, 157; impact of adjust- and, 196; social rate of return to
ment measures on informal sector schooling and, 19495; subsidies to
.and, 159; labor market outcomes the poor and, 196-97; wage indexa-
and, 160; labor unions' targeting tion and, 189-92
616 Labor Markets in an Era of Adjustment
Chilean economic setting- changes in primary exports of, 218; problems
labor market institutions (1979-82), from attempt to stimulate the eco-
177; economic reforms in 1970s, omy of, 223; public expenditures on
172-78; external sector, 179; finan- education and social services, 249-
cial crisis of 1982 and policy re- 50; public sector employment in,
spouses, 178-81; labor law changes 229, 231; public sector growth of,
after 1982, 177-78; macroeconomic 219-20; recovery from 1980-82
indicators in, 173; market deregula- downturn, 251; rural employment in,
tion and, 176-78; privatization and, 231; saving and investment as per-
175-76; puiblic sector reforms in, centage of GDP, 228; share of labor
175-76; socialist experiment and, force in agriculture in, 218, 227; so-
171-72; stabilization program in, cial indicators in, 218; stabilization
178; structural adjustment in, 180- measures in (1982), 224; Standard
81, 181-97; trade reforms and, 174- Fruit and, 231, 233; terms of trade
75; wage indexation, 177, 178 and, 220, 223, 225; underemploy-
Competition in product markets, 4 ment in, 229; the unemployed in,
17; unemployment in, 224, 227,
Competitiveness, 21, 22 229, 230, 231, 249, 252; U. S. aid
Costa Rica: agricultural production and to, 226; value added by trade-related
products, 218, 225, 231-32; Central sector in, 238; wages in, 233, 234,
American Common Market and, 219, 240, 242, 243-245, 248, 252, 253;
223, 235; comprehensive structural women and the crisis in, 251
adjustment program in, 225-26; em- Costa Rican labor market: before the
ployment in, 229, 230, 231, 237, aids, 227-2a, 230, 231-32, indica-
239; foreign debt of, 223, 226-27; tors of, 230
free trade zones in, 226; gradual im- ,.-
freetrae znesin, 26;graualim- Costa Rican labor market institutions in
plemention of structural adjustment the adjustment pracess, 232; real
programs in, 226; gross savings and wae ad,u2t2-34 sctoral
investment as percentage of GDP in, ploment shifts and, 234-35, 237,
228; import-substituting industrial- . 9 240n seitsa wages and, 240,
ization polircy of, 219; income dis- 239, 240; sectoral wages and, 240-
tribution, poverty, and unemploy- 42, 243-45,246-47, 248-49
inent in, 230, 242, 248-49; indica- Costs, policies to counter increase in
tors of internal and external finance nonwage, 22
in, 222; inflation in, 224; invest-
mnent financed by foreign savings C6te d'lvoire, 259; agricultural labor
and, 220, 227; labor mobility in, force in, 268-69; central bank
237, 253; labor unions in, 233, p3CEAO) for West Africa and, 261,
254; macroeconomic experience be- 265; crops in, 281-82, 291 n.10;
fore the crisis in, 217-20, 223-27; currency of, 261; decisions on send-
macroeconomic indicators in, 221; ing children to school, 284-86;
merchandise exports of, 235, 236; deficit elimination in, 260; educa-
minimum wage in, 231, 232, 233, tion of formal sector employees,
241; oil price hike (1974) and, 220; 277; effects of exogenous events on,
one hundred day stabilization plan 264-66, 267, 268; employment
of 224-25; political stability of, shift from services to manufacturing,
218; poverty in, 227, 248-49, 252; 306; exchange rate and foreign debt
Index 617
of, 267; exports of, 263; export Devaluation: as alternative to wage flex-
subsidies and, 260, 266-68, 271-72; ibility, 21; fiscal proxies to mimic,
females working in public, and pri- 266; formal-informal wage gap. and,
vate sectors, 277; fertility analysis 207; in Korea, 54546, 571; in
(descriptive statistics) for, 311, Thailand, 588, 594
312; fertility and the role of women
in, 301-3; French franc and, 261; uragedworkereffect, 18
impact of education on urban family Distributive conflicts, 25-29
size in, 302; income distribution in, Dutch disease model, 321, 322, 326,
298; inflation in; 262; labor supply 328, 471
model for, 308-9; living standards
in, 262; Living Standards
Measurement Survey (LSMS) on em- Earnings functions, 38-39, 40, 41
ployment, 269, 272-79; migration Economic performance, indicator of, 9
and economic incentives in, 303-5; Economic scenarios, exhilarationist and
modem manufacturing firrns data in, stagnationist, 20-21
269, 272; poverty in, 298-301; Education: in Chile, 194-95; in Cote
reentry into full-time education in, d'Ivoire, 277, 282, 283, 284-86,
307; rural sector in, 280-82; school 287, 294, 302; in Egypt, 340, 341,
attendance in, 285, 286; school sys- 342; in Ghana, 363, 368, 369,375,
tern in, 284; as a sverely indebted 397; in Kenya, 413, 441-42, 443,
middle-income country, 262; stabi- 448, 451; in Malaysia, 505, 506,
lization in, 265-66; statistical 507-8, in Thailand, 606, 607, 609-
model of transitions for, 310; struc- . 10; see also School enrollment;
tural adjustment in, 265-68, 300, Women's education
306; unemployment and structural
adjustment in, 300; urban labor mar- Educational enrollments in Asian coun-
ket of, 268-79; wages and gender in, tries, 609
301, 302; Western Africa Monetary Egypt: adjustment measures to reduce
Union (UMOA). and, 259, 261, 262, fiscal imbalance in, 351; agricultural
263, 265; see also Ivoirian entries exportables lag in, 349; agriculture
Country studies, 6-8 .as a labor reservoir in, 335, 345;
bias against employing women in,
Crop price increases, 41 . 342; construction sector in, 332,
335, 345-49; contractual arrange-
ments for labor in, 346-47; demand
Data: complications of interpreting, 18; for crop labor in, 332; economic
limitations of, 7-8; misleading ag- liberalization (Infitah), in, 321,
gregate, 279; problems with 322; education's impact "n employ-
B3razilian, 164, 165, 167 ment in, 340, 341, 342; employ-
ment in, 329, 332; employment af-
Debt export ratios, 544 . ter the oil windfall in, 339-45; ex-
Deflation in Chile, 23, 25 change rates in, 321, 324; exports.
of, 324; female and child labor in,
Demand: real wages and, 23; wage level
. . . - ~~~~~~334, 349; gradualism in adjustment
as determinant of aggregate, 20 in, 320; guaranteed employment for
do Tray, D., 298, 299 graduates in, 325, 333, 343, 351;
618 Labor Markets in ani Era ofAdjusiment
imports of, 323; inflation in, 318; in Kenya, 406, 408, 410, 414; in
investment in, 32S, 326-28, 329, Korea, 544, 559; in Malaysia, 464,
349-50; investment deflator in, 353; 465, 469, 470; shift in output to-
labor mobility in, 334, 337, 345; ward, 5; of Thailand, 589-90, 591
labor market in, 333-34, 352; labor Export subsidies: in C6te d'lvoire, 260,
force structure in, 331; labor short- 266-68, 271-72; in Korea, 559
ages in 323; liquidity of banking
sector in, 327; macroeconomic fea-
tures of economy of, 318, 319, 320- Family size, education and (in Cote
21; migration from, 322, 332, 333, dIvoire), 302
345; oil and, 318, 321, 322, 324, Female workers: development and par-
326, 349, 350; output shift away ticipation in work force by, 563; see
from exportables, 329, 332; policy also Women
response to external shocks and, Fiscal contraction, long-term growth
320; poverty in, 334; price controls and 52-53
and, 326; private investment in,
330; productivity per worker -in, Food, tradable and nontradable, 49
350; public employment in, 324, Foreign exchange and exchange rates: in
332, 340; public expenditures in, Argentina, 63, 84; in Bolivia, 105;
322, 324; public investment in, in CUte d'Ivoire, 261, 267; in Egypt,
329; public sector industry and, 328; 321, 324; in Ghana, 391-92, 358-
real wages by sector in, 335, 336, 59; in Kenya, 408-9, 414; in Korea,
338; revenues from oil and Suez 571; in Malaysia, 470-73, 475,
Canal and, 324; the state as preferred 529; three concepts of exchange rate
employer of graduates in, 341-42; and, 471
tariff protection and, 328; wage set-
ting mcchanism in, 337-38;. unem-- .
Ghana: adjustment program's primary
ployment and unemployed in, 17, goal in, 391; agriculture in, 359,
318,xa334, 339, 34-36, 350; wages n 397, 401; budget deficit in, 361-62;
indexation in, 336, 350; wages in, budget surplus resulting from
* 347 Economic Recovery Program (ERP),
Emergency employment programs, 17 366; cocoa and, 359, 378, 391, 397;
Employment: by sector, 31-32; sectoral commerce as main declining sector
shifts in, 18, 29-30, 33-34, 37-39, of, 393; consequences of crisis of
41, 55; shift from formal to informal 1983 in, 362-64; consumable short-
sector in, 18; see also entries for ages in, 361; data for labor market
specific countries analysis in, 357-58; debt burden of,
Exchange rates; three concepts of, 471; 362; decline in farmers' earnings in,
sxnee rals Frei echnge and ex- 359; distribution aspects of adjust-
see also FoTeign exchange and ex-
change rates ment program in, 398-401; determi-
nants of earnings and labor supply
Exchange rate/wage tradeoff, 27-28 in, 375, 378-81; drought and famine
Exports: in Brazil, 144, 149; in Costa in, 362, 363; earnings pattern
Rica, 218, 235, 236; in Cote changes in, 396; economic crisis
d'lvoire, 263, 266; in Egypt, 324, (1983) in, 358, 362-64; education
329, 332, 349; in Ghana, 358, 391; in, 363, 368, 369, 375, 397; ex-
Index 619
ports of, 358, 391; fastest growing lors in, 367, 397; employment sec-
sectors of, 391; food production per [or by education and sex in, 369;
capita in, 363; forcign exchange hourly earnings in, 374, 376-77;
earnings for, 358, 359; foreign ex- houschold labor income in, 371-73;
change shortage in, 359, 361; for- income effect of second job in, 385,
eign exchange system of, 291; gov- 390; intersector flow of workers and,
ernment sector. growth in, 392; 397; labor participation rates by age
health standards in, 363; incentives and sex, 368; labor supply and, 381;
for workers to relocate in, 395; in- labor supply elasticities and, 390;
flation in, 359, 361,.362, 365, 366; main and secondary jobs in, 381,
infrastructure development in, 358; 382-83, 384, 385; migration and,
investment under ERP in, 366; labor 374, 397, 400, 401-2; monthly la-
force in agriculture in, 359; labor bor income determinants, 373; per-
market's role in adjustment process sonal labor incomes and education,
in, 390-98; macroeconomic indica- 375; real wage trends in, 399; self-
tors for, 359, 360; macroeconomy employed women in, 370; self-em-
since ERP in, 365-66; manufacturing ployed workers in, 370. 401; sex-re-
in, 361; migration out of the country lated wage differentials, 373-74; to-
and, 361, 364, 374, 397; migration tal income vs. labor income in, 372;
within the country, 397, 399, 400, underemployment and, 371; unem-
401-2; per capita. GDP, 362, 363; ployment and, 370-71; women's ed-
poverty in, 363, 401; reduction in ucation and, 368; working age and,
corruption in, 395, 398, 402; re- 367
turns to education in, 397; real wage Gindling, T. H., 240, 241, 250
gains in, 393; regional diffeTentials
during adjustment in, 395, 399; sec- Glewwe, P, 298 299
tor growth rates in, 361; sector spe- Government (public sector) employment
cific earnings functions in, 385, and size: in Argentina, 89-90; in
389-90; service prices declines in, Costa Rica, 219-20, 229-31; in
393; trade surplus resulting from Egypt, 324, 332, 340, 341-42; in
ERP in, 366, 391; two sector model Ghana, 369, 392, 397; in Kenya,
of economy of, 398401; unioniza- 412, 430, 454; in Thailand, 605-6
. tion in, 379, 399; wages in public Government wages during adjustment,
sector in, 392; wage trends in, 393, - 37
394' 395 Gross domestic product (GDP) growth
Ghana's Economic Recovery Program rates, 9, 10; time pattern of, 12
(ERP), 357, 362, 364-65; central el- Growth, labor market adjustment and
ement of, 398; impact on poverty long-run 52-54
of, 401; stabilization aspects of, '
390; workers adversely affected by,
400, 402 Human capital investment, 53-54;
Ghana's labor market: in the adjustmnent stabilizationli and structural
process, 390-98; child labor and,
367; education and, 368, 369, 375;
employment and cducation and, 368; Income distribution, 5-6, 46-49; struc-
employment in public and other sec- tural change and, 56
620 Labor Markets in an Era of Adjustment
Inflation, 45; in Argentina, 67, 6B, 70, in, 277-79; wage employment vs.
84, 93; in Brazil, 149, 150, 151, self-employment in, 274; see also
152, 163; in Chile, 24-25; in Costa Cote d'lvoire
Rica, 224; in Egypt, 318; in Ghana,
359, 361, 409; in Kor3a, 547, 550K Kenya: adjustment problems of, 406-12;
406, 08, 49; inKorea 547,550- adjustment targets, instruments, and
results in, 412-16; agricultural
informal sector: in Bolivia, 117, 122- prices and terms of trade in, 435,
24; in Brazil, 154, 159; in Cote 436; agricultural reform in, 415;
d'lvoire, 276-77; definitions of, agriculture in the economy of, 408;
488-89; in Kenya, 430-35; in budget deficit of, 412; Budget
Malaysia, 488; shift of employment Rationalization Programme, 413;
to, 18, 56-57 capital formation in, 454; coffee and
Investment, stabilization, and structural tea and, 406; economic growth pat-
adjustment policies, 6 tern of, 408; education and eamings
Ivoirian macroeconomics, 261-65; cur- in, 448, 451; educational expendi-
rent account and, 264; economic in Lures of, 413; employment in educa-
dicators of, 263; exports and, 263, tional system of, 429; employment
d266; investment, government ex- of university and training program
penditures, and trade and, 264; real graduates in, 412, 419; exchange
effective exchange rate depreciation rate policy of, 418-9, 414; exports
and, 266; Western African Monetary 408, 409, 410; financial policy of,
*Union (UMOA) and, 261, see also. 40,0,41;fncilpiyo,
Union (UMvOA)e and, 261; seealso 413; financial sector reoron in, 410,
O*le dlvaire 413, 415; first oil crisis and, 406;
Ivoirian rural labor market, 280-81; foreign exchange reserves of, 414
agricultural enterprises and house- government dexficit of, 409, 412;
hold producLion aovernent,dficit f, 40,g412
hold production and, 281-82; agri- growth of GDP (1963-73) and, 406;
cultural prices and labor supply in, indirect taxes and, 410; industrial
288-93, 293-98; schooling and- structure of 411; industry and trade
work in household enterprises and, reforms in, 415; inflation in, 406,
282, 283, 284-86, 287; work and 408 409; informal sector in, 430-
school status and, 287; see also Cote 409 informal
35; infrastructure investmentin
d'lvoire 440-41; interest rates and, 410, 413;
international economic events and,
Ivoirian .urban labor market: formal 405; investment spending decline
firms' incentives to become informal in, 440; labor unions in, 421;
and, 276; incentives in manufactur- largest source of government rev-
ing sector and, 270-71; informaliza- enue for, 410; male/female wage dif-
tion, self-employment, and family ferentials in, 449; macroeconomic
enterprises in, 276-77; in the mod- indicators for, 407; manufacturing
era sector, 270-75; movement of la- sector share of GDP, 412; migration
bor between industries and, 274-75; and cultural and tribal differences in,
real wages in, 279; returns to educa- 439; migration to informal sector
tion in, 278; structure of, 268; wage in, 446; poverty in, 445; price con-
determination and real vtage growth trol removal in, 415; public sector
Index 621
growth in, 412, 454; public sector Korea: capital-intensive technology in,
job advantages in, 430; real earn- 564; currency devaluation in, 545-
ings in, 405; real GDP of, 405, 408; 46, 571; cycles in economy of, 535-
school fees in, 441442, 443; second 42; data in study of, 579-81; earn-
oil crisis and, 408, 454; stabiliza- ings per worker in, 578; economic
tion without proper adjustment in, indi:ators for, 576; efficiency incen-
412; structural adjustment policies tives in, 552-53; emigration of
-of, 415; structural changes in, 410; workers and, 553; exchange rate pol-
tourist industry in, 410; trade bal- icy in, 571; export subsidies in,
ances of, 405, 409, 414; trade pot- 559; firn size, bonuses, and wages
icy of, 414; unemployment in, 419, in, 560, 573; firm specific skills in,
420; wage bill for public sector in, 552-53;. food policy in, 546; gov-
429; wage policy in, 415; women in ernment interv'ention in wage set-
informal sector of, 432 ting in, 45, 551,553; household in-
Kenya's labor market: agriculture as come per urban worker in, 579; in-
source of jobs in, 416; employment flation in, 547, 550-51; investment
in urban and rural sectors and, 417; rate in, 551-55; labor aristocracy in,
rural participation rates in, 418, 572; labor union in, 551, 560, 574;
419; population growth, 416 -17; large firn sector shrinkage during
trade unions in, 421; urban female adjustmnent in, 561-62; manufactur-
participation rate in, 417, 418, 419; . ing labor oDst data, 577; net barter
urban informal sector growth in, and income terms of trade in, 575;
416; wage determination in, 421-22 overtime pay and bonuses in, 552,
560; profit-shari'ng in, 552, 553,
Kenya's labor market adjustment, .422; 5 p
capital formaLion and, 440-41; earn- 572; subsidies for export/large scale
ings functions analysis and, 448-49, scecor in, 559; total factor.produc-
tiVity growth IU,- 551; unemploy-
450, 451, 452, 453; employment by mevttin,r14,h17, 553; unit plaor
industry and, 426-27; excess labor determinants in, 547-49; unit labor
absorbed into self-employed sector,
434 fmal wge roilet 44 costs and external competitiveness
434;th ofremale minimum wage pr4, in, 544, 545; value added per worker,
560; wage determination in, 551-55
428; income distribution and, 445- 5 e i mnftu
-571; wage explosion in manufactur-
47, 448; informal sector eanmings
and employment, 430-35, 443; la- treins in, 549-51, 572, 574; wages
bor~ ~ ~ ~ ~ ~~~~ted moblit 549d,1 572 574e wageges
bar mobility and, 439; male wage in large scale manufacturing sectors
profiles and, 444; public sector of, 572
wages and employment and, 429-32,
433, 434; rural sector and, 434-35; Korean labor markets and wage differen-
sectoral wages and employment and, tials, 555-56; farm/nonfarn differ-
423, 424-25, 426-27, 428, 429; ru- ential, 556-59; female participation
ral-urban migration and, 436-40; ur- and, 563, 564, 564-71, 5731
ban employment and, 434; wage male/female differences and, 563,
rates by occupation and, 435; 565-71; wage differentials by sector,
women in the labor market and, 441- 572-73; wage difference by size of
43, 444, 445, 454 firms and, 559-63, 564
622 LaborAMarkets in anEra of Adjustnent
Korean manufacturing establishments: Labor market role during adjustment:
earnings differential by sizC of, 562; distributive conflicts and, 25-29; la-
technology, size, and productivity bor market institutions and, 41-46;
differentials, 561 real wages and, 14, 16, 17-25; sec-
Korean stabilization and adjustment toral employment shifts and relative
policies: banking and credit policies wages and, 29-30, 33-34, 7-39, 41;
and practices and, 543; debt/export unemployment and, 14, 15, 16, 17-
ratio and, 544; export promotions as 25
major strategy in industrialization Labor militancy, 41
of, 542; factors affecting external Labor shedding, structural adjustment
competitiveness, 544; foreign bor- and, 34
rowing and, 543-44; instruments of and 34
Labor shifts, sectoral, 18, 29-30, 33-
targeted industrial development, 34 37-39 41 55 56
543; recovery from external shocks 5 56
and, 544; tariff policies and, 543 Labor unions, 3-4, 44; in Brazil, 155-
59, 163; in Costa Rica, 233, 254; in
developing countries, 41-42; in
Labor market: in Argentina, 79-80, 80- Ghana, 379, 399; in Kenya, 421; in
81, 82, 84, 93; in Bolivia, 107-8, Korea, 45, 551, 553, 560, 574;
119-20, 121, 122-25, 130, 131; in linked to political parties, 43; in
Brazil, 143, 152-55, 155-59, 160, Malaysia, 481-82; weak 45
161, 167; in Chile, 169, 177, 197-
207; contract between buyers and Lorch, K., 270-72, 276
sellers in, 2; in Costa Rica, 227-28, Lucas, R., 481, 482
230, 231-32, 234-35, 237, 239, Luxury unemployment hypothesis, 17
240; in Cote d'lvoire, 268-79; 280-
81, 283, 288-93, 293-98; defined,
2; in Egypt, 323, 333-34, 346-47, Macro policy instruments, 1
352, iII Ghana, 381, 390-98; in Malaysia: balance of payments in, 470;
Korea, 555-56, 561-71, 571-73; in budget deficit of, 465-66, 467, 468,
Malaysia, 481-82, 483-84, 487, 472-73; business cycle phases in,
489; rigidity in, 43; role in struc- 464-70; causes of unemployment in,
tural adjustment of the, 4-5; sector 514-16; construction sector in, 480,
transitions of, 39; that is working 484,-491, 492; consumer price index
well, 2; in Thailand, 595-98, 599- in, 463; debt/GDP ratio in, 468;
608 earnings in the female labor market,
504, cairnings of males and of fami-
Labor market adjustment consequences, 50 4; earning ando i-
,.lies in, 520-24; earning trends- in,
56; on income distribution, 46-49; 488-504; economy of, 459-63; the-
on long-term growth, 52-54; for educated unemployed in, 17; educa-
women, 49-52 tion of unemployed workers in, 501;
.Labor market flexibility, success of ad- electronics and electrical machinery
justmrent and, 143 industries in, 460, 469, 519; em-
Labor market institutions, 41-46; labor ployues and self-employed in vail-
make riidt and 435 ous categories in, 500-1; employ-
rnarket rigidity and, 43, 55
ment in agriculture in, 492; em-
Labor market reforms, 44 ployment growth rate in, 487; em-
i1ndex 623
ployment trends in, 488-504; excess domestic exchange rate (RDER) in,
of spending over income in, 472; 471, 475; real effective exchange
exchange rates in, 470-73, 529; ex- rate (REER) in, 471, 472; real ex-
pansion in education in, 486, 505, change rate in, 529; real wages in,
506, 507-8; exports of, 464, 465, 475, 476, 478-83; recession and re-
469, 470; external terms of trade covery in, 484, 529; rice and, 462-
and, 467; financing the deficit in, 63: rubber and, 464, 475; the self-
468; fiscal deficit of, 469; fiscal employed in, 490, 530-31; shift to a
policy of, 467; food in the economy service economy and, 494-96; short-
of,462-63; formal and informal sec- run problems of adjustment in, 528-
* tor definitions and, 488-89; GDP 29; short-run problems of the
growth rate and terms of trade and, macroeconomy of, 470-85 structural
463, 464, 469; government wage adjustment in, 485-86; terms of trade
bill in, 461; growth and cycles in and, 469, 471-73; the unemployed
the economy of, 463-70; inflation in, 17, 510-12; wage setting in,
in, 467; informal sector in, 488; in- 482-83; wages in manufacturing and
stability of export earnings of, 461; construction in, 480-81; work force
*interest costs in, 529; internal mi- distribution by industry in, 490-92
gration of labor in, 525-28; labor.
fbrce growth rat": in, 489; labor mar-
kets and competitiveness in, 483- Malaysian adjustment and labor markets
84; labor supply and demand in, in the long run, 530-32; age and
487; labor unions in, 481-82; land. earnings and, 497, 498-99, 500; ag-
development in, 486; manufactured gregate supply of and demand for la-
goods in, 469; mode of employment bor, 487; earnings dif'erences be-
in, 489-90; monetary restraint in, tween formal and informal sectors,
467; movements in consumption 497, 498-99, 500-2, 503, 504; earn-
and product wages, 475; new class of ings differences by industry and
industrial workers in, 520; new eco- mode of employment, 492-97; earn-
nomic policy (NEP) in, 485-86, new ings of employees and of self-em-
jobs in private services in, 492; ployed and, 501-2, 503, 504; earn-
nominal effective exchange rate ings -and skill levels, 495; educa-
(NEER) and, 471; paddy sector in, tional expansion and change in
486, 493-94, 530; palm oil and, occupational structure, 505, 506,
464; petroleum and, 464; population 507-8; initial conditions and
growth rate in, 487; poverty in, objectives of structural adjustment
485-87; product wages in, 475-76, and, 485-86; new employment in
477; prepayment of external debt agriculture, 492; regional effects of
by, 470; private investment in, 469; labor market adjustment and, 525-
private savings in, 466, 467-68, 28; trends in distribution of the work
469, 471; public sector in, 461-62. force by industry, 490-92; trends in
482; public sector restraint by, 470; employment and earnings in formal
racial disparities in income in, 485- and informal sectors and, 488-504;
86; racial imbalance in educational unemployment trends and causes,
attainment in, 505; racial quotas in 508-12, 513, 514-16; women in the
employment in, 486; rate of return labor market, and adjustment, 516-
to education in, 505, 507-8; real 17, 518, 519-24, 532
624 Labor Markets in an Era of Adjustment
Models, Dutch disease type, 321, 322, Stabilization, 1; control of fiscal and
326, 328,- 471 monetary policy and, 106; role of
Morales, A., J. A., 100, 103, 104, 106, labor market in, 3; unemployment
107 during, 19-20
Stabilization and adjustment policies:
investment and, 6; labor markets and
Nominal effective exchange rate
(tIEER)>471 ~~~~macroeconomic success of, 1; see
also eitries for individual couantries
Standards of living, stabilization and ad-
Ormacbea, E., 117,118 justment pDlicies and, 1
Structural adjustment, 1; restructuring
Pareto optimal outcome, 2 incentives for trade and, 106; role of
Pollack, M., 240, 241 labor markets in, 4-5; tariff-subsidy
schemes and, 266-67; women and,
Poverty, 48; in Brazil, 144, 145, 162- 50-51; see also Adjustment and en-
63, 164-66; in Costa Rica, 227, ti f i c
24849, 252; in C6te d'lvoire, 298- triesfor idividual countries
301; in Egypt, 334; in Ghana, 363, Structural transformation during debt
364, 401; in Kenya, 445; in crises, 30
Malaysia, 485-57; in Thailand, 593
Tanzi effect. 151 n.4
Real domestic exchange rate (RDER), Taylor, L., 20
471 Thailand: adjustment problems that af-
Real effective exchange rate (REER), 9, fect labor in, 597-608; agricultural
11, 12, 14, 471 exports of, 589-90, 591; agricultural
Real wages, 14, 17-25; aggregate de- household earnings in, 591-92;
mand and, 4; demand and, 23; in- agriculture in, 589, 590; Bangkok's
dexes of, 16, 35, 36; macroeco- economy in, 592; basis for eco-
nomic consequences of decline in, 4; nomic growth of, 587, 589; demand
too high for maintaining employ- for skilled labor in, 586; devalua-
meat and output, 4; see also Wages tions in, 588, 594; education sys-
Republic of Korea, see Korea tem's future needs-in, 609-10; educa-
Research suggestions,d57 tion system's response to labor mar-
ket in, 606; employment opportuni-
Rice: 462-63, 594; as a nontradable, 26 ties in the informal sector in, 595;
employment by work status in, 596;
enrollment in secondary education
in, 609; foreign debt of, 587, 588;
Saachs, J., 100, 104, 106 future adjustment issues and, 608-10;
School enrollment, recession and, 55 government employment in, 605-6;
Severance pay, 43 household income distribution in,
592-93; industrial and urban growth
Shocks, external, I in, 590; infrastructure problems in,
Skill differentials, 38 608; macroeconomic growth and ex-
Skilled labor, growth of, 53 ternal balance of, 586-90; manufac-
Index 625
tured exports of, 589; migration in, Unemployment benefits, 4
601, 602, 603; new agricultural land
in, 602; owner cultivators in, 601X2; Unions, see Labor unions
policies to control foreign debt Uthoff, A., 240, 241
problems, 587, 588; poverty in,
593; retums to education in, 607;
rice export taxes in, 594; rural mi-
grants in, 601; shortage of scientifiec Value added, wages in, 22 484
and technical manpower in, 586, Verry, D., 481, 482
603, 606, 608; structural imbalance
and adjustment in, 590-95; trade and
current account deficits of, 587; un-
doculmented private land in, 602; un- Wage differentials: formal/informal, 37-
employment and underemployment 38; role of, S
in 595, 596, 603, 604, 606, 608; Wage flexibility, 22; devaluation as al-
urban growth in, 603 temative to, 21
Thailand's labor market and structural ad- Wage fluctuations, 19
justment, 597-98; general employ- Wage indexaion: in Latin America, 43;
ment situation and, 595-97; labor unemployment and output loss andi ,
movement and the changing produc- 4
tion structure in, 599-603; mismatch
of demand for and supply of educated Wages: changes between sectors in rela-
manpower, 603-8; seasonal unem- tive, 34; exchange rates and, 27-28;
ployment and, 598-99 in value added, 22, 484; see-also
Relwages,
Tradables, shift of output toward, 5 Real
Western \Africa Monetary Union
UMOA. See Western Africa Monetary (UMOA\259 261, 263
Union (UMOA) Women: in \ Argentina, 90-91; in
Underemployment, 17; in Costa Rica, Bolivia, 115, 129-30, 131,132; in
229; in Thailand, 595 Costa Rica,\ 251; in CBte d'lvoire,
277, 301, 306.?, 387; in Egypt, 334,
Unemployment, 15; during adjustments, 342, 349; in\electronics industry,
14, 17; in Argentina, 85, 88, 89-90; 50; in Ghana, et68, 370, 373-i4; in
Bolivia, 108; in Brazil, 159; causes Kenya, 417, 4fl, 419, 44143, 444,
and explanations of, 4, 19-20, 54; 445 454; in Kbrea, 563, 564-71,
in Chile, 14, 204; in Costa Rica, 573; labor market adjustment and,
224, 227, 229, 230, 231, 249, 252; 49-52; in Malaysia, 516-17, 518,
cross-country comparisons of, 14; 1
in Egypt, 17, 318, 334, 339, 340- and, 512
43, 351; in Kenya, 419, 420; in and, 5-
Malaysia, 17, 508-12, 513, 514-16; Women's education: labor force paitici-
in Thailand, 595, 596, 598-99, 603, pation and, 50, 51, 52; see also
604, 606, 608 inedividual country entries