World Bank Reprint Series: Number 180
Bela Balassa
Trade in Manufactured Goods:
Patterns of Change
Reprinted with permission from World Development, vol. 9, no. 3 (1981), pp, 263-75.
World Development, Vol. 9, No. 3, pp. 263-275, 1981. 0305-750X/81/030263-13/$02.00/0
Printed in Great Britain. � 1981 Pergamon Press Ltd.
Trade in Manufactured Goods: Pattems of Change
BELA BALASSA*
Joliins Hopkins Un iversity
Summary. - This paper looks at the changing pattern of trade in manufactured goods betwee,.
the developed industrial countries and the developing countries. It first reviews recent changes
in this trade and the policies applied. Then it analyses the comparative advantages of the indus-
trial and the developing countries and irdicates the commodity composition of this trade. The
implications of the results arc drawn with referencc to the benefits of trade in manufactured
goods for the participants.
This paper will examine the changing pattern modity classes 5-8 in the LIN Standlard Inter-
of trade in manufactured goods between the national Trade Classification less non-ferrous
developed industrial (for short industrial) and metals (68). Trade in manufactued goods will
the developing countries. Section ' of the paper be examined first in value terms, indicating
will review recent changes in this trade and tihe changes in the balance of trade, and subse-
policies applied. In Section 2, the comparative quently in volume terms. Chawnges in trade
advantages of the industrial and the developing volumnes will further be relatedl fo the rate of
countries in manufacturedt goods will be ana- economic growth, and the results will be
lysed and changes in the commodity compo- evaluated with reference to the traade policir,s
sition of this trade indicated. In conclusion, the followecd by individual countries and country
implications of the results will be drawn with groups.
reference to the benefits of trade in manufac-
tured goods for the participants.
(a) Chlaniges ini tlue balance of tradle
in inanufactiured goods
I. TRADE IN MANUFACTURED GOODS
BETWEEN INIDUSTRIAL AND The value of the manufactured exports of
DEiVELOPING COUNTRIES the industrial countries to the OPEiC countries
increased slightly more than fivefold hetween
l`his section reports on changes in trade in 1973 and 1978 while the reverse flow of these
manufactured goods between the industrial commodities remained at very low levels
countries andl the developing countries during (Tablel). As a result, the industrial countries'
the 1973 -1978 period. The former group is trade surplus in mnanufactured goods with OPEC
defined to include the United States, Canada, rose from $ 1 2.3 billioni in 1973 to $63.5 billion
the European (Common Market (EEC`)' the in 1978. This increase covered 69%$} of the rise
Furopean Free Trade Association (EFTA>2 in the oil import bill of the industrial countries
andi Japan; within the latter group, data for with Ol'FC, from $28.7 billion in 1973 to
OlPEC3 and for the non-OPItC developing $102.9 billion in 1978.5
countries4 are separately shown. The extent of coverage o.f the incrcase in oil
The two groupings exclutce tlhe countries imports by manufactured exports to OPlC,
of Southern E-urope (other tlhan Portugal); l1owever, varies to a considerable exrent amiiong
Australia, New Zealand and South Africa;
and tthe centrally-plannedl econoniies. Wliile *
the countrics of Southern E:urope rnay bec Thits paper was pircsented at the Synmposiumi on
c cIlndustrial Policies for the 80s', held in Mladrid on
appropriately included in the developing- 5 9 May 1980. 'The author is Professor of Political
country group, this had not been donc in the l conomy at Johns llopkins University and Consultant
present paper because the future trade pros- at the Worid Baank. Ile is indebte(l to Joung-Yong Lee
pects of these countries are intimately linkedl for research assistance.
with their accession to the Comnmon Market, The opinions expressed in the paper arc the author's
to take place during the 1 980s. own and should not be construed to represent the
Manufactured goods are defiriedl as conm- views of the World Bank.
263
264 WORLD DEVELOPMENI'
Table 1. Trade between the industrial countries and the developing countries in manufactured goods*
(billion dollars, f o.b.)
United States Canada EEC
EZxport Import Balance Export lmport Balance Export Import Balance
OPEC
1973 2.43 0.06 2.37 0.17 0.00 0.17 7.00 0.42 6.58
1974 4.27 0.09 4.18 0.29 0.00 0.29 i1.63 0.40 11.23
1975 8.00 0.09 7.91 0.56 0.00 0.56 20.00 0.46 19.54
1976 9.90 0.09 9.81 0.75 0.00 0.75 23.11 0.53 22.58
1977 10.50 0.09 10.41 0.83 0.00 0.83 29.42 0.66 28.76
1978 11.97 0.14 11,83 1.04 0.00 1.04 34.56 1.11 33.45
Non-OPEC
countries
1973 10.18 7.34 2.84 0.57 0.54 0.03 15.92 4.68 11.24
1974 16.10 9.74 6.36 0.88 0.75 0.13 23.04 5.83 U7.21
1975 17.75 8.89 8.86 1.04 0.75 C.29 25.81 6.85 18.96
1976 18.25 12.96 5.29 1.05 1.20 -0.15 25.21 8.92 16.29
1977 18.84 15.62 3.22 1.08 1.17 -0.09 29.74 10.88 18.86
1978 23.54 20.93 2.61 1.24 1.27 -0.03 37.28 14.10 23.18
Developing
countries
together
1973 12.61 7.40 5.21 0.74 0.54 0.20 22.92 5.10 17.82
1974 20.37 9.83 10.54 1.17 0.75 0.42 34.67 6.23 28.44
1975 25.75 8.98 16.77 1.60 0.75 0.85 45.81 7.31 38.50
1976 28.15 13.05 15.t0 1.80 1.20 0.60 48.32 9.45 38.87
1977 29.34 15.71 13.63 1.91 1.17 0.74 59.16 11.54 47.62
1978 35.51 21.07 14.44 2.88 1.27 1.01 71.84 15.21 56.63
EFTA Japan Industrial countries
Export Import Balance E.xport Import Balance Export Imnport Balance
OPEC
1973 0.69 0.04 0.65 2.56 0.00 2.56 12.85 0.52 12.33
1974 1.19 0.07 1.12 5.16 0.02 5.14 22.54 0.58 21.96
1975 1.96 0.05 1.91 8.12 0.00 8.12 38.64 0.60 38.04
1976 2.79 0.05 2.74 8.94 0.02 8.92 45.49 0.69 44.80
1977 3.21 0.07 3.14 11.50 0.03 11.47 55.46 0.85 54.61
1978 3.72 0.09 3.63 13.57 0.05 13.52 64.86 1.39 63.47
Non-OPEC
countries
1973 2.46 0.76 1.70 9.66 2.20 7.46 38.79 15.52 23.27
1974 3.53 1.02 2.51 14.90 2.52 12.38 58.45 19.86 38.59
1975 3.74 1.18 2.56 14.26 2.01 12.25 62.60 19.68 42.92
1976 3.77 1.38 2.39 15.57 2,88 12,69 63.85 27.34 36.51
1917 4.49 1.82 2.67 19.83 2.97 16.86 73.98 32.46 41.52
1978 5.89 1.93 3.96 25.64 4.27 21.37 93.59 42.50 51.09
Developing
count ries
together
1973 3.15 0.80 2.35 12.22 2.20 10.02 51.64 16.04 35.60
1974 4.72 1.09 3.63 20.06 2.54 17.52 80.99 20.44 60.55
1975 5.70 1.23 4.47 22.38 2.01 20,37 101.24 2C0.28 80.96
1976 6.56 1.43 5.13 24.51 2.90 21.61 109.34 28.03 81.31
1977 7.70 1.89 5.81 31.33 3.00 28.33 129.44 33.31 96.13
1978 9.61 2.02 7.59 39.21 4.32 34.89 158.45 43.89 114.56
Source: GATT,International Trade, 1976/77 and 197R/79.
* SITC commodity classes 5-8 less 68 (non-ferrous metals).
TRADE IN MANUFACTURED GOODS 265
Table 2. T/se rate of growth of manufactured trade between industrial and developing
countries and that of the industrial countries gross domestic produet (per cen tr
E:xports
Ii- lrports Apparent income
OPEC non-OPEC Ail LDC Ail 1,DC GDP elasticity
1963-1973*
Industrial countries - 8.2 16.5 4.6 3.6
1973- 1978
United States 23.7 6.3 10.6 11.1 2.5 4.4
Canada 33.7 8.8 16.5 6.8 3.4 2.0
EEC 23.3 6.2 12.5 12.1 2.1 5.8
EFTA 23.8 5.4 1(1.7 8.4 1.5 5.6
Japan 26.4 9.9 14.2 , 3.0 3.6 0.8
Industrial countries, total 24.2 7.2 12.5 10.2 2.5 4.1
of which 1974 45.3 24.5 29.7 4.2 0.4 10.5
1975 52.8 -4.5 11.5 - 1.5 -0.8 1.9
1976 16.4 0.7 6.7 27.3 5.2 5.3
11977 12.9 7.8 9.9 7.6 3.6 2.1
1978 1.1 9.6 6.0 15.5 4.0 3.9
Sources;
1963-1973: United Nations, Moeithly Bulletin of Statistics and Yearbook of National Accounts Statistics,
various issues.
1973-1978: Value of trade - GATT, International 'Trade, 1978/79.
Unit values - GATT, Network of Worid 7rade by Areas and Conimodity Classes, 1955 1976 (Geneva: 1978)
and International Trade, 1977/78 and 1978/79;and United Nations, Monthlly Bulletin of Statistics.
Gross domestic product -- OECD, National Accounts of OECD Countries, 1979; and Economic
Outlook (December 1979); United Nations, Yearbook of Na�ional Accounts Statistics, 1978.
* Estimated by use of regression analysis.
the industrial countries and country groups. Free Trade Association (2.4), the European
The relevant ratios for 1978 are, the United Common Market and the United States (2.3),
States, 38%; Cais-da, 56%; the European Com- and Canada (2.2). The differences are larger as
mon Market, 99S;; the European Free Trade regards the growth of manufactured imports
Association, 1 199r; and Japan, 65%. The ob- frormi the non-OPEC developing countries; the
served differences in these ratios are largely ratios of 1978 to 1973 imports are 3.0 in the
explained by intercountry differences in the European Cornmon Market, 2.9 `n the United
rate of growth of oil imports and in the initial States, 2.6 in the European Free Trade Associ-
coverage of oil imports by manufactured ex- ation, 2.4 in Canada, and 1 .9 in Japan.
ports to OPEC, while the rate of expansion of Changes in the industrial countries' trade
the exports of manufactured goods to OPEC balance in manufactured goods with the non-
has varied little among the industrial countries, OPEC developing countries have been further
Thus, the ratio of 1 978 to 1 973 oil imports affected by their initial position in this (rade,
ranged from 7.5 in the United States to 2.6 in as expressed by the ratio of manufactured
the lEuropean Common Market, and the 1973 exports to imports in 1973. This ratio was the
coverage of oil imports by manufactured ex- highest in Japan (4.4), followed by the Euro-
ports to OPUiC varied from 68% in EFTA to pean Comnion Market (3.4), the European
21% in Caniada. In turn, the ratio of 1978 to Free Trade Association (3.2)., with the United
1973 exports of manufacturedl goods to OPE1,C States ( 1.4) and Canada ( 1.1 ) far behind.
was between 5.0 and 5.5 in the major industrial Given its high initial trade position, slightly
countries and country groups; it was 6.1 in above-average export expansion, and substan-
C.anada. tially below-average import growth, Japan
Intercountry differences in the rate of experienced by far the largest increase in its
expansion of manufactured exports from the tracle surplus in mnanufactured goods with the
industrial countries to the non-OPEC develop- non-OPEC developing countries. This increase,
ing countries have also been rather small. Japan from $7.5 billion in 1973 to $21 .4 billion in
occupies first place, with a ratio of 1978 to 1978, accounted for fully one-half of the rise
1973 exports of 2.7, followed by the European in the combined manufactured trade suirplus of
266 WORLD DEVELOPMENT
the industrial countries with the non-OPEC for 1976 represented a 'rebound fromn the
developing countries from $23.3 billion to recession'.
$5 1.1 billion. At the sarne time, substantial differences are
The EEC and EFTA also increased their observed among industrial countries and
trade surplus in manufactured goods with the country groups in regard to both their GDI'
ncn-OPEC' developing countries; the increase growth rate and the rate of increase of their
was from $1 1.2 billion to $23.2 billion in the imnports of manufactured goods from the devei-
first case and fromn $ I1. billion to $4.0 billion oping countries. Japan leads in l.trms of the
in the second. In the same period, tihe US trade rate of GDP growth (3.6%) but trnlis the other
surplus with these couintries declined from $2.8 industrial countries by a considera; le margin as
billion to $2.6 b llion. With very similar export far as the growth of the volur,m- of mnanufac-
and irnport trends, differences in the 1973 tured imports from the developing counltries is
trade position of the EEC and EFTA, on the concerned (3.0%7'/year), with an apparent
one han,l, and the United States, on the other, income elasticity of ;mport demand of 0.8 for
explain the observed differences in the results. the 1' 93---1978 period. Canada was second to
Japan in terms,of GDP growth rates (3.4%/
year), and had the second-lowest growth rate
(b) Changes in the volutne of manufactured of manufactured imnports originating in the
trade and 4n the gross dontestic product developihig countries (6.8'%), with an app2,rent
ircome elasticity of 2.0.
Tab!e 2 provides data on changes in the vol- The European Common Market and the
umne of trade in manufactured goods between European Free Trade Association had the
the industrial and the developing couLntries in lowest ;GDP growth rates (2.1 and 1.5%) and
the 1973--1978 period. Information is also the highest apparent income elasticities of
provided on the rate of growth of GDP in the import demnand (5.8 and 5.6), with the volume
industrial countries and on their 'apparent' of manufactured imports from the de, -,ping
irncome elasticities of imnport demnand for manu- coiiitries rising at average annual rates of 1 2.1
factured goods originating in the developing and 8.4%, respectively. GDF growth rates in
countries, derived by dividing the rate of the United States equalledl the average for the
growth of the volurne of imports by that of industrial countries in the period 1973-,-1978
GDP.6 (2.5%), while the rate of growth of iniports
The table shows an apparent income elas- (Il.1') and the apparent income elasticity of
ticity of import demand of 4.1 for the indus- imnport demnand (4,4) were slightly above the
trial countries in the 1973--1978 period; their average.
combined manufactured imports from the Table 2 fturther shows that the volume of
developing countries rose at an average annual mnanufactured exports of the induistrial coun-
rate of 10.2%7r in volume terms as compared to tries to the developing countries increased at
an average rate of growth of GDP of 2.5<. I'he an average annual rate of 1 2.5%r between 1 973
volume of nianufactured imnports from the and 1978, the relevant results being 24.2% to
developing countries increased more rapidly in Ol'EC, and 7.2% to non-OPEC, countries.
the preceding decade (1 6.5%,',/year), when Relative rates of growth of exports and imports
considerably higher rates of GDP growth (4.6%S/ in value termns show identical results as the
year) were associated with a lower apparent termns of trade in manufacturc(l gooals clid not
income elasticity of import demand (3.6). chanige hctween the two groups of cOuntries.7
The results for the 1 973 - 1978 period were The tinue pattern of exports and imports,
affected to a considerable extent by the 1 974 - however, exhiiitedi substantial variations. The
1975 recession. In 1975, the industrial coun- volume of manufacture(l exports ['rom thte
tries' GDI' was slightly below the 1973 level industrial countrics to the OII.(' couLtries grew
and the volume of their iniports of' nanufac- hy nearly one-half in 1974 and hy more than
tured goods froni the developing countries was one-half in 1975, reflectinig the tripling of
only 3%� higher. In the following 3 ycars, 01PiC export earnings in 1974 and the adjust-
growth rates averaged 4.3'i' for GDP and 16.57'</ ment to the higher level of carnings in 1975.
year for the volume of manufactured imports, Bult, the rate of growth off nianufactured
i.e. an apparent income elasticity of 3.8. The cxports to the OP'k`' countries declined rapidly
relevant results are 3.8 and 1 1.57<, correspond- in subsequent years, with practically no chanige
ing to an apparent income elasticity of irm- shown in volume ternis in 1 978. The results are
port demand(l of 3.0, if we luiit our attention to e\plaincLI by the slow-(aowni in the rise of oil
the I. 't 2 years on the grounds that the results earnings in 1976 and 1977 and the ahsoltitc
TRADE IN MANUIYACTUREI) GOODS 267
decline of these earnings in 1978. countries to the oil crisis and the 1974-1975
The volume of manufactured exports to the world recession, The promotioni of exports, to
non-OPEC developing countries increased by be discussed later, was another form of policy
one-fourth in 1974, declined slightly in 1975, response. At the same time, the higli and rising
and remained practically unchanged in 1976. apparent income elasticity of import demanci
The relevant growth rates are 7.8 and 9.6% in does not give evidence of overall imnport substi-
1977 and 1978, respectively. For the 1973 - tution in manufactured goods during this
1978 period as a whole, these exports rose at period.12
ain average annual rate of 7.2%1r, corresponding The rapid growth of their exports of manu-
to an apparent inconie elasticity of import factured goods to the non-OiEC developing
demand of 1.8 as the GDP of the non-OPEC countries favourably affected economic activity
cleveloping countries grew 4.1%/year. The in the industrial countries. Such effects were of
apparent income elasticity is 3.0 for the OPEC particular im1portance in 1974, when the vol-
countries whose gross domestie product rose by uime of these exports rose by one-fourth while
8.2%!r1year between 1973 and 1978.8. the in(dustrial couritries were sli(ing into
The combined GDP of the developing coun- recession. In recent years, too, the increase in
tries ;ncr'ased at an average annual rate of 5.3% the export surplus of the industrial countries in
between 1973 and 1978 as compared to a rise mtanufactured trade with the non-OPEC devel-
of 1 2.5%1/year in the volume of their manufac- oping countries had a multiplier effect on their
tured imports frorn the industrial countries, national economies as it was not fully offset by
corresponding tcW an average apparent income higher prinrary imports.
elasticity of import deemand of 2.4. The rate of Notwithstancling the rise in their export sur-
growth of GDP was higher (6.2' 5) and that of plus, the indiustrial couintries' imports of iiantu-
manufactured imports from the industrial factured goodls from the developing couintries
countries lower (8.2%n) in the 1963-1973 grew rapidly, with the apparcint incoome elas-
period, when the apparent incomiie elasticity of ticity of iniport tdeimian(d rising from 3.6 ilu
itmport demanid was 1.3.9 liowever, (uring the 1963 1973 to 4.1 �i 1973 1978. These
carlier period, the income ancl the import shares results do not provide evidence of increased
of the OPEC countries were much smialler. protectionismn on the part of tie indtustrial
countries. While the effects of the 'new protec-
tionism' were noticeable in 1977 when the
(c) Thie policies applied volume of snanufactured inmports from the
developiing countrii s increase(l by only the
Following the quadrupling of oil prices, the rise was 1(v-; in 1 978 when improvements were
OPEC countries attempted to increase their made on the protection front.'3 Available
absorption of' manufactured imports at a rapid information poinits to further rapicl increases
rate. While in 1974 these countries had an in 1 979 .14
export surplus of $84 billion, the surplus fell The observed, changes coUld not, however,
to $53 billion in 1975 and to $43 billion by be explained on the basis of lemand consider-
1978.10 Apart from the rising imports of the ations alone. While access to industrial country
capital-surplus oil exporters, foreign borrowing markets bas provicled opportunities for export
on the part of the other OPEIC countries con- expansion, the exports of manufacturedl goods
triblute(l to this result. from the developing countries had respon(dcd
lForeign borrowing also contribtutecl to the to the policies followedl by these coun tries. A
higli rate of growth of the nianufactuLred im- number of' developing countries adopted an
ports of the non-OPEC developing countries export-oriented strategy diuring tIe 1960s andl
from the incdustrial countries. 'Flie net overall have continucd with this str.itcgl after 1973.
trade deficit of these countries rose froim $15 Available evidenice wggLi:sls thati countrics
billion in 1973 to $37 biillioni by 1)78,'' witb following an export-oriented s(rraltcgv wcre
mtuch of the inerea.se being financed by borrow- better- able to surniiount the adlverse ellects of
ing abroad. Foreign borrowing permittecd the the quadrupling of oil prices andI the 1974
non d-P(' clevelopinig countries to increase 1975 world recession than countries witth an
their import surplus in manufacturecl goocls vis- import-suli.sfil ltion orientation.15
�-vis the industrial countries from $23 billion Althougt. supply factors bave had a crucial
in 1973 to $51 billion in 1978, notwithstanding role in determining the overall rate of expan-
[heir higher oil bill. sion of niantufci ultrcd expor�s hy lie develop-
Foreign borrowing representcd a policy re- ing countries, the allocation of these exports
sponse on the part of the non-OI> t dIcveloping aniong the industrial countries lias beern affec-
Table 3. Imports of manufactured goods by industrial countries
1973 1978 Incremental ratios (1973-1978)
M/GDP AILDC/GDP MLDC/M MiGDP MLDC/GDP MLDC/M AAI/t'GDP CMLDC/IGDP eMLDCIZAM
United States 3.28 0.57 17.37 4.60 1.00 21.74 6.71 1.69 25.17
Excluding Canada 2.49 0.57 22.90 3.65 1.00 27.42 5.50 1.69 30.70
Canada 14.83 0.43 2.90 15.29 0.61 3.99 16.05 0.87 5.42
Excluding United States 3.43 0.43 12.53 3.51 0.61 17.37 3.64 0.87 23.89
EEC 11.32 0.48 4.24 13.44 0.77 5.73 15.93 1.11 6.97
Excluding intra-EEC trade 3.86 0.48 12.45 4.89 0.77 15.74 6.09 1.11 18.23
Excludingintra-European trade 2.65 0.48 18.08 3.31 0.77 23.26 4.08 1.11 27.20
EFTA 18.82 0.48 2.55 18.34 0.62 3.38 13.15 0.56 4.26
Excluding intra-EFTA trade 15.38 0.48 3.12 15.38 0.62 4.03 11.36 0.56 4.93
Excluding intra-European trade 3.32 0.48 14.44 3.48 0.62 17.84 2.67 0.56 20.98 -
Japan 2.38 0.53 22.57 1.80 0.44 24.43 1.38 0.37 26.80 z
IndustrialCountries 7.19 0.52 7.23 8.39 rP.78 9.30 9.61 1.07 11.13
Excluding US-Canada and 2.60 0.52 20.03 3.17 0.78 24.58 3.78 1.07 28.27
intra-European trade
Source: GA'-, International Trade, 1978/79; OECD, Economic Outlook (December 1979) and NVationalAccounts of OECD Countries; and United Nations, Year-
book of National Accounrts Statistics, 1978.
Notes: M = total imports of manufactured goods.
MkLDC = manufactured imports from developing countries.
GDP = gross domestic product.
TRADE IN MANUI:ACTURr)D GOODS 269
ted by tlhe trade policies followed by these indilstrial countries. For this purpose, it is
countries. It has often been said that the United further necessary to examine the commodity
States has traditionally been more open to composition of imports, since somne products
imports from developing countries than the encounter non-tariff barriers in the industrial
other industrial nations. Data for 1973 offer countries while others are subject only to tariffs.
only weak evidlence in support of this propo- TIhis qluestion will be takeen up in Section2.
sition. Thus, excluding trade between the
United States and Canada that takes place 2 COMPARATIVI ADVANTAGI AND
largely in the framework of multinational '. COMMADATY ADVANI OF
corporations and one-third of which is subject T DE COMMOI)ITY INOMPOSITION 0F
to free trade treatment under the US -_ Canada TRADF IN MANUFA(TUREI) (;OOI)S
automotive agreement, as well as trade within In an earlier paper,19 the authlor hfas shown
the European free trade area in manufacturedl that the pattern of worl exports of nanufac-
goods enconipassing the FEC and FFTA, the tured goods can bc explained in terms of
ratios of imports from developing countries to intercommodity differences in capital-labour
the total imports of manufactLures and to the . .
gross domestic produict were not substantially ratosme ntsrh i icfto ne in capite
higher in thie United States than in the EEC
and Japan (Table 3).16 advantage derived in that paper willJ he utiltzed
e e o . in the fol1owing to analyse the pattern of
e heonytether with the dtr rion export inport ratios in mnanuifactured trade
economy, togetber with the deterioration of ewe h ntsra niti o-1E
its competitive position, contributed to rap�idbetwenpthe industrias and7 Stienonuent
increases in lUS imports of manufactured goods
between 1973 and 1978, with the developing celanges in the comnodity compiliosition of this
bctwen 173 nd 178,withthetrade wvill further be examlined. with consider-
countries increasing their share in the total. ation given to the factors thal inflwencedo tile
Thus, the incremental ratio of nianufactuired
observed resullts. linally, the role of the newly-
imports originating in the developing countries idstrialrizin countriys ti the anel-
to the gross domestic product was substantially
higher during this period in the United States factured goods fromn the developing countries
(1 .69) than in the European Common Market will be noted.
(1.11) although the differences are smaller as
far as the incremental share of the developing (a) Tue structure of comparative achatage
countries in their manufactured imports (30.7 In or(ler to examine the comparative advan-
anci 27.2%, respectively) is concerned. tages of tile industrial and the developing
The contras� with Japan is much more countries i,i their mutual trade in manufactured
pronounced. With an apparent income elas- goods, the capital labouir ratios reported in the
ticity of import clemand of only 0.8, the ratio earlier paper have been averaged in the 1 1
of Japan's manufactured imports from the commodity category breakdown employed in
developing countries to its GDP fell from 0.53 GATT statistics, using US prodluction data as
in 1975 to 0.44 in 1978, corresponding to an weights 20 Tlie estimated ratios pertain to
incrernental ratio of 0.37 between 1973 and physical as well as to hiuman capital. The
1 978. In the same period, the incremental former has been obtained as tlhe ratio of the
share of imports from developing countries (physical) captial stock to employment; the
in the total iimports of manufactured goods latter has been estimated as the discountedl
was 28.3% in Japan. value of the �liffcrcncc between the average andl
Japan used a variety of formal and informai the unskilled� wage, taken to represent tIhe
measures to liiit the imports of manufacturedl return on invesinment in huImn capital. 'flie
goods, in particular from developing cotuntries. relevant data are shown in T able 421
These measures were liberalized in 1978.17 As A/lthough the 11 commnodity category
a result, the volume of manufacture(l imports sciieme involves a considerable degree of aggre-
from the developing countries increasedl by gation, substantial differences are observed in
more than one-fourth between 1977 and capital- labour ratios among the individual
1978,18 foliowing a decline in the 1973-1977 categories. The ratio of physical capital per
perio(d. worker (expressed in 1000 UIS dlollars per
A consideration of the imports of all manu- worker) is the highest for iron and steel (27.7)
factured goods originating in the developin�g and for chemicals (21.4); it is the lowest for
countries does not sullfie, however, to appraise clothing (2.4) an(l for other consuLmler gootds
the effects of the trade policies followed by the (6.7). The ratio of hunman capital per worker
Table 4. Comparative ad!ontage ratios and trade in manufactured goods between industrial and developing countries
Capital per worker Exrourt surplus* Export-import Ratio of 1978 to 1973 tradet
ratio* - Ratio of NlCsit
Physical human total 1973 1978 Export lmport Export in imports from
($1000) (S billion) 1973 1978 non-CPEC LDC OPEC non-OPEC LDCs
Iron and steel 27.71 28.15 55.85 3.00 5.73 6.45 5.82 1.95 2.16 3.45 76 �
Chemicals 21.37 25.51 46.88 5.43 12.02 6.90 7.91 2.26 2.53 3.38 47 r
Other serni-manufactures 19.59 24.21 43.80 --0.11 0.58 1.00 1.10 2.57 2.34 5.17 58 e
Engineering products, subtotal 9.61 29.38 38.99 18.52 45.16 6.01 4.61 2.59 3.38 5.74 83 e
Machinery for specialized industries 9.44 28.34 37.79 6.81 16.79 *3.98 29.46 2.50 4.54 5.01 90 <
Office and tekcornmunication equipment 7.91 35.22 43.15 1.13 2.31 i.88 1.62 2.50 2.91 5.09 75 r
Road motor vehicles 12.89 25.40 38.39 3.39 8.76 38.67 21.37 2.64 4.78 5.31 91 �
Other machinery and transport equipment 9.66 30.27 39.93 6.62 16.80 7.07 5.20 2.70 3.67 6.81 77
Household appliances 8.29 39.09 47.38 0.47 0.31 1.42 1.08 2.56 3.36 5.43 95 z
Textiles 10.00 16.62 26.62 0.32 0.05 1.16 1.01 1.58 1.81 2.50 55
Clothing 2.37 11.00 13.37 -3.03 -8.75 0.11 0.09 2.18 2.81 5.88 81
Other consuTrr goods 6.73 25.99 32.72 -0.86 -3.70 0.80 0.49 2.55 3.22 7.24 90
Maanufactured goods, total 11.89 26.11 38.00 23.29 51.14 2.50 2.20 2.41 2.74 5.04 75
Source: Capital per worker: Bela Balassa, 'A "stages approach" to comparative advantage', op. cit.
Exports and imports: GATT, International Trade 1978/79 and GATT tapes.
* Industrial countries' exports to, and imports from, the non-OPEC developing countries.
t Ratio of the industrial countries' exports and imports, respectively, in trade with developing countries.
I Newly industrializing countries, defined to include Argentina, Brazil, Chile, Mexico, Uruguay, Israel, Hong Kong, Korea, Singapore and Taiwan.
TRADE IN MANUFACTURED GOODS 271
is also the lowest for clothing ( 11.0), followed non-OPEC developing countries in the case of
by textiles (16.6); it is the highest for house- machinery for specialized industries (54.0) and
hold appliances (39.1) and for office and tele- road mnotor vehicles(38.7). In turn, theexport-
communication equipment (35.2). Combining import ratio was 7.1 for the other machinery
physical and human capital, iron and steel and transport equipmeint category, which also
(55.9), household appliances (47.4) and cherni- includes some relatively simple products, such
cals (46.9) are at the upper, clothing (13.4) as bicycles.
textiles (26.6) and other consumer goods (32.7) Finaliy, 1973 export- import ratios were
at the lower, end of the range (Table 4). relatively low for office and telecommunication
If engineering products are considered as a equipment (1 .9) and for household appliances
single group, 1 973 export--import ratios in ( 1.4). In both instances, physica! capital -labour
trade between the industrial countries and the ratios are substantially luwer thaen average,
non-OPEC developing countries largely corre- although relatively high human capital intensity
spond to the pattern of comparative advantage raises their overall capital-labour ratio. At the
as represented by capital-labour ratios, the same time, within the first category, develop-
only exception being semi-manufactures.22 As ing countries exported chiefly electronic com-
shown in Table 4, iron and steel and chemicals ponents that are highly labour-intensive while a
with the highest overail capital-labour ratios large share of exports in the second category
also had the highest export-irnport ratios (6.5 were radios that are produced by relatively
and 6.9, respectively) in 1973. Apart from semi- simple techniques.
manufactures, engineering prodlucts are placed We have further calculated weighted averages
next in terms of overall capital -output ratios of capital- labour ratios for the exports and
(39.0) as well as export-irnport ratios (6.0). the imports of the industrial countries in their
At the other extreme, the exports of the trade in manuifactured goods with the non-
industrial countries to the non-OPEC develop- OPEC developing countries, the weights being
ing counitries hardly reached one-tenth of their the value of exports and imports in the 11 com-
imports in the case of clothing that exhibits by modity group breakdown. In 1973, the relevant
far the lowest capital-- labour ratio. The next ratios for exports (expressedl in 1000 lUS dollars
lowest export --import ratio (0.8) is shown for per worker) were 13.8 for physical capital and
the other consumer goods category, including 27.5 for human capital, totalling 41.3. In turn,
shoes, travel goods, toys, sports goods and a average capital--labour ratios for the manufac-
variety of miscellaneous products, which ture( imports of the irndustrial countries from
exhibit the second-lowest capital intensity. the non-OPt.( (lcveloping countries were 9.8,
A seemingly aberrant result is observed in 25.1 and 34.8, respectively.
the case of other semi-manufacturcs that had rhe results indicate the comparative advan-
an export -import ratio of 1.0 in 1973, not- tages of the industrial countries in capital-
withstandiing their relatively high capital intensive commodities vis-�-pis the non-OPEC
labour ratio (43.8). Hlowever, several of the developing countries. Th, conclusion applies
.ornrnodities includled in this category are also to the industrial countries and country
natural resource products, which tact provides groups, taken individually, with percentage
an advantage to developing countries that dlifferenices in average capital labour ratios for
possess the resources in question. Also, the their exports andl imports in trade in manufac-
category is rather heterogeneous as it includes tured goods with thle non-C)l't (.' dleveloping
capital-intensive products, such as pulp and countries rangirig between 1 8': in the Linited
paper, as well as labour-intensive pro(lucts, States and 36%' in Canada. At thle same time,
such as leather ancl rubber nianufactures, when for the industrial coiunitries, takien t)gclith(�, as
weighting by UJS productiion imparts an upwarrd well as lfor the individuadl couitrics and country
bias to the estimai,lted average capital labour groups, the extent of cotinparativc ;idva;ntage
ratio for tile group as a wholc . vis-�-vis the non-OPE(l develop�nig countries
Apart from thieir above-average capital inten- appears to be greater in regard to physical than
sity, the sophistication of the prod tict ioni for hluiman capital.
process and the need for the availability of
precision-engineered parts, com ponients and (b) Changes l�n (lie cONtl ,nod�it 'v -omttposition
accesssories limited the export possibilit es of of fracle, 1973 1(>78
the (P veloping countries in a variety of engin-
eering products. These factors in large part (i) 1ndustria/-countay cx ports
explain Lie very highi export -iiimport ratios in In the exports of the industrial counitries to
the trade of the industrial countries with the OPEC' in the 1973 19(78 period, ab-'ve-,average
272 WORLD DEVFLOPMENT
increases are shown for consumer goods; the $16.8 billion in speci;alized rnachineiy and from
ratio of 1978 to 1973 exports was 5.3 for $6.6 billion to $16.8 billion in motor vehicles.
motor vehicles, 5.4 for household appliances, -I'he next largest increases, with the ratio of
5.9 for clothing, and 7.2 for other consunier 1978 to 1973 imports being 3.7 and 3.4, took
goods as against an overall average of 5.0.23 In place in the other-machinery-and-tranisport-
turn, the lowest ratios are observed in the case equiprnent ancl the hoLusehold appliances car-
of iron and steel (3.5), chernicals (3.4) and egories, Shipbuilding an(d the exportation of
textiles (2.5). Finally, machinery and e(uip- parts and components of' machinery and trans-
ment exhibited average ratios (5.0-5.A1), except port equipment iinportantly contributed to the
for the high ratio shown for the 'other-machin- growth of iriiports originating in the non-OlPEC
ery-and-transport-equipment' category (6.8) developing countries in the first case, ancd radics
where aircraft and other military equipment ancl T.V. sets in the second. But while the trade
are of importance. surplus of the industrial countries declined in
Given the limited doomestic production of the second category, a substantial increase is
manufactured goods in rnos. of the OPlC(' shown for the first where aircraft and other
countries, these results tend to reflect patterns military produicts are of importance.
of domestic use. It would then appear that, on Ail in all, the industrial countries' imnports of
the whole, increases in oil earnings were used engineering products fromn the non-OPEC devel-
inore to increase consuinptiorn and military oping countries shows above-average increases,
expenditure than to raise investment levels. with the ratio of 1978 to 1973 irlports being
Iron and steel, chemicals and textiles experi- 3.4. As a result of these changes, tby 1978
enced smnaller chan average increases in indus- engineering products came to account for
trial-country exports of manufactured goods to 29,4%70 of tlie manufactured imnports of the
the non-OPEC developing countries also. In industrial countries froni the non-OPIK(' devel-
these three cominodity categories, thie ratios of oping countries as compared to )3.9%yo in
1978 to 1973 exports were 2.0, 2.3 ancd 1.6, 1 973.: In absolute terms, the most important
respectively, as compared to an overall average categories are other machinery and transport
of 2.4. The relevant ratio is 2.6 for the con- equipment ($4.0 billion ini 1978), office andT
sumer goods categories, the only exception telecommunication equipment ($3.8 billion),
being clothing (2.2). Finally, the ratio of 1978 and �ouschold appliances ($3.7 billioli).
to 1973 imports of machinery and equipment The ratio of 1978 to 1 73 imports from the
was 2.5, except that the import of military non-OPECl developing countries is the lowest
equipment raised this ratio to 2.7 for the other- for textiles (1.8) that have long been subject
machinery-and-transport-equipment category. to restrictions in the industrial countries. The
These results conflict with the popular inldustrial couintries have also benefited froin
image, which has found its way into economie technological chang that has led to the appli-
models, that the non-OPEC developing coumn- cation of more capital-interisive techniques.
tries would limit their consumer good imnports Nevertheless, with textile imports from the
in favour of the imports of' machinery and non-OPEC developing countries rising more
equipment. In fact, withi clothing imports being rapidly than exports to these countries, the
small in absolute terms, the imports of con- trade surplus of tie industrial countries i�l
surmer goods appear to have incr. ased some- textiles declined from $0,3 billion in 1'973 to
what more than the imports of capital goods, practically zero in 1 978.
and increases in the imports of military equip- In the same period, the trade deficit of the
ment were even larger. indiLstrial countries with the non-OPE;C devel-
oping countries in clothing rose from $3,0
(ii) 1ndustrial-counltry itiports b�llion to $8.8 billion. As noted carlier, the
In the imports of the industrial countries comilparative advantage of the non-01li(' devel-
from the non-OI'I-i.(' developing countries, the oping countries is the strongest in clotliing;
largest increases are observed in the specialized nor have there been technological changes
machinery ancI the road motor vehicles cat- unfavourable to them in this industry. Thus,
egories; the ratios of 1978 to 1973 exports are no�twithstaniding the limitations imposed in the
4.5 and 4.8, respectively, as compared to an framework of the Multifibre Arrangement,
overall average of 2.7. But this result was the ratio of 1978 to 1973 imports was 2.8 for
achieved from a very small base; in fact, be- clothing, exceeding the overall average.
tween 1973 and 1978, the trade surplus of the The industrial countries also experienced a
industrial countries with the non-OPEC clevel- risc in their trade cleficit in other cons����icr
oping countries increased from $6,8 billion to goods with the non-OPEC developing countries
TRADE IN MiANUFACTURED GIOODS 273
from $0.9 billion in 1973 to $3.7 billion in The bulk of the rnanufactured exports of the
1978, with the ratio of 1978 to 1973 exports non-OPEC developing countries to the indus-
being 3.2. The commodities included in this trial countries originates in the newly-industrial-
category have relatively high unskilled-labour izing developing countries. This group has beecn
intensity and average skill intensity. With few defined to include developing countries where
exceptions, these commodities have not been per capita incomes exceeded $1 100 in 1978 andl
subject to restrictions in the industrial coun- rnanufacturing production sccounted for at
tries. Ieast 20% of' :he gross domestic product. The
Finally, increases in the imnports of the countries in questioni are Argentina, Brazii,
industrial countries fiom the noil-OPEC devel- Chile, Mexico, Uruguay, lsrael, [long Kong,
oping countries were below-average in the case Korea, Singapore and Taiwan.
of intermediate produets, where the relevant As shown in Table 3, in 1978 ihe share of
ratios are 2.2 for iron and steel, 2.5 for chemi- the newly-industrializing countries in the manu-
cals, and 2.3 for semi-manufactures. These factured imports of the industrial countries
products have the highest physical capital from the non-OPEC developing countries was
labour ratio, although, as we have seen, the 75%. This compares with the 46% share of'
serni-manufactures category also includes natu- these countries in the combined GDP of the
ral-resource and labour-intensive products. non-OPEC developing countries. If comnpari-
sons are made with all (OPEC and non-OPFC)
(c) Somne implications of the results developing countries, the relevant shares are 73
and 33%. The four F ar kastern countries in the
It appears that the OPEC as well as the non- group, Hong Kong, Korea, Singapore and Tai-
OPEC developing countries have utilizecl wan, accounted for 5% of the gross doniestic
increases in foreign exchange availabilities to product and 52% of the manufactured ex-
expand thc imports of consumer goods and ports of the developing countries to developed
military equipment more rapidly than the country markets.25
imports of investment goods, Apart from con- Compared to the other non-OPEC deveioping
flicting with popular preconceptions, these counries, the newly-industrializing countries
results raise questions about the allocation of tend to export relatively skill-intensive com-
foreign exchange between current expenditures modities. They have a high export share in
and investment in these countries. engineering products (8.3%;) that have higher
It further appears that the developing coun- skill-intensity than any other commodity
tries have made little headway in exporting category Within the engineering group, housc-
comnmodities that are highly intensive in physi- hold appliances with a 95</ export share of the
cal capital, require sophisticated technology, newly-industrializing countries are the most
or necessitate the availability of precision- skill-intensive. The export share of the newly-
engineered parts, components and accessories. industrializing countries is also high in other
Several of the developing countries, however, c�risumer goods (90%/c) that have average skill-
have made progress in exporting skilled-labour- intensity, and it is the lowest in textiles (55%)
intensive ccinmodities, such as ships and T.V. that have low skill-intensity. They also have low
sets, as well as parts, components and access- shares in chemicals (47%) and in other sermi-
ories of engineering products to the industrial manufactures (58%7c) that are intensive in nhysi-
countries Also, growth has been rapid in the cal capital ancd several of wvhich embodo natural
exports of other consuiner goods thtat have resources. And while the newIly-in'i.istrializing
average Jkill intensity ancl are relatively inten- countries have an above-average sh.re in cloth-
sive ini unskilled labour. ing exports (81 '�i) that are intensive in unskilled
A few exceptions apart, proclucts in these labour, more sophisticated items have a higher
comrnmoclity categories do not encounter non- than average share in thieir exports of these
tariff barriers in the industrial countries. It cornmodlities,
should be addedl, however, that the exports of
textiles and clothing, subject to limitations first
under the (`otton Textiles and, subsequently, 3. CONCLUSIONS
the Multifibre Arrangement, have also experi-
enced a high growth rate. The exports of these This paper has examined changes in the pat-
commodities increased at an average annual tern of trade in manufactured goods in the
rate of 7.7% in volume terms as compared to 1973- 1978 period. T'he growth of mnanufac-
the average rate of expansion of manufactured tured exports of the industrial countries Io the
exports of 10,2%/o. developing countries acceleratecd duririg this
274 WORLD DEVELOPMENT
period, with an average annual rate of increase down of economic growth in the industrial
of 12.5% as compared to 8.2% between 1063 countries on their imports from the developing
and 1973. As the gross domestic product of countries, but do nlot provide evidence of the
the developing countries rose 6.2%/year in the effects of increased protectionism that would
first period and 5.3%/year in the second, their have been expected to lead to a fall in the
apparent income elasticity of import dernand income elasticity of import demand. Protec-
increased from 1.3 in 1963-1973 to 2.4 in tion, however, appears to have limited the
1973- 1978. growth of imports into Japan while the imports
The rise in the apparent income elasticity of of textiles and clothing into all the industrial
import demand in the developing countries cotuntries have been unfavourably affected by
reflects both the growth of export earnings of the operation of the Multifibre Arrangement.
the OPEC countries and the increase of foreign Nevertheless, these exports rose at an average
borrowing by the non-OPEC developing coun- annual rate of 7.7% in the 1973-1978 period
tries. In fact, the apparent income elasticity of and came to account for 59.8% cf the imnports
import demand was 3.0 for the OPEC cot ntries (again excluding US--Canada trade and trade
and 1.8 for the non-OPEC developing countries within the fEuropean free trade area), and 4.9'%'i
in the 1973-1978 period. of the domestic consumption, of textiles and
Increased exports of manufactured goods to clothing in the industrial countries in 1978.
the developing countries favourably afftcted The relevant shares for manufactured imports,
economic activity in the industrial countries, in taken together, wvere 24.6 and 1[5%'.27
particular in 1974 when the volume of these The results further provide evidence of the
exports rose by one-fourth while the industrial changing pattern of comparative advantage,
countries were sliding into a recession. At the with the newly-industrializing countries �n-
same time, the developing countries have as- creasingly exporting skill-intensive commodities
sumed increased importance as markets for the and the Icss-develope(l countries starting t
industrial countries. accounting for 47.3% of export unskilled-labour intensive commoclities.
their exports (excluding US Canada trade and In turn, the exports of the industrial cotuntries
trade vithin the European free tracle area) and continue te be doiminated by prodcucts that are
for 5.1% of their production of manufactured intensive in physical capital, require sophisti-
goods in 1 978.26 cated technology, or necessitate the availability
The industrial countries experienced average of precision-eegineered parts, components and
annual increases of 4.6%1r in their GDP �nd accessories.
1 6.5%7o in their imports of manufactured goods It would appear, then, that the expansion of
from the developing countries between 1963 manufactured tracle between industrial and
and 1973, corresponding to an apparent income developing countries has made it possible for
elasticity of import demand of 3.6. The income these countries to specialize according to their
elasticity increased to 4.1 in the 1973-- 1978 comparative advantage. I'`his pattern of spechil-
period, when the decline in the rate of growth ization, in tuirn, has favourable effects in re-
of GDP (2.5% in 1973 -1978) exceeded that in source allocation and economic growth in the
the rate of growth of imports ( 1 0.2%). countries participating in international trade.
The results show the effects of the slow-
NOT ES
1. Belgium, Denmark, irance, ('.ersnany, Sreland, Zealand), excluding centrally-planned economies and
Italy, Luxembourg, the Netherlands and the United the OPE'C countries.
Kingdom .
2; Austria, I;iniand, Icelanid, Norway, Portugal, 5. GATT, International Trade, 1978/79 (Geneva:
2.d autria, Switzland, lcln,Nra,Prua, 1979), Tables B3- F -- the data for 1978 are prelimii-
nary.
3. Algeria, I,cuador, Gabon, Indonesia, Iran, Iraq,
Kuwait, Libya, Nigeria, Qatar, Saudi Arabia and the 6. l'he expression 'apparent' income elasticity is
United Arab Emirates. used here to refer to the f`act that this measure neglects
the effects of changes in relative prices on the volume
4. The countries of Latin America and the Carib- of imports. Nor does the measure distinguish between
bean, Africa (other than South Africa), Asia (other the effects on imports of changes in population and
than Japan), Oceania (other than Australia and New in per capita incomes.
TRADE IN MANUFACTURED GOODS 275
7. GATT, International Tride, 1978/79, op. cit., 19. 'A "stages approach" to comparative advantage',
and Table 1. in Irma Adelman (ed.), Economic Growthl and Re-
sources, Vol. 4, National and Intermational Issues
8. United Nations, Yearbook of National Accounts (London: Macmillan, 1979), pp. 121 -156; published
Statistics, 1978, and World Bank Data Base. in an abbreviated form as 'The changing pattern of
comparative advantage in manufactured goods',
9. Estimated by regression analysis from data Review of Economics and Statistics (May 1979), pp.
published in United Nations,Monthly Bulletin oJfStat- 259.- 266.
istics and Yearbook of National Accounts Statistics.
20. lhe capital-labour ratios theniselves have been
10. GATT, International Trade, 1978/79, Table G. derived from the data of the US Industrial Census.
Their use in the present context assumes equal substi-
11. ibid. tion elasticities between capital and labour across
industries.
1 2. On the relevant methodology and its application
to three developing countries, see Bela Balassa, 'Poiicy 21. In addition to these 'stock' coefficients, in the
responses to external shocks in selected Latin Ameri- �arlier paper 'flow' coefficients are also reported,
can countries', presented at the NBER/FIPE/BEBR
can ounries, peseted t te NER/FPE/EBR represcnting profits per worker in the case of physicai
Conference on Trade Prospects among the Americas: capital and the average wage in the case of human
Latin American Diversification and the New Protec- capital. The stock coefficients are preferredo, however
tionism, held in Sao Paulo Brazil on 24-26 March in view of tluctuations in profit rates over time and
1980, and to be published in the Quarterly Review of the inclusion of the unskilled wage in the average wage
Economics and Business and in a Portuguese trans- figure.
lation in Estudios Economicos.
22. The OPEC countries have been excluded from
1 3. Bela Balassa, 'The Tokyo Round and the develop- these comparisons as tliey export practically no manu-
ing countries', Journal of World Trade Law (Marcth/ factured goods.
April 1980), pp. 93-- 118.
14. GATT, 'International trade in 1979 and present 23- For lack of price indices on a commodity group
prospects: first assessment by the GATT Secretariat' basis, export ratios have bern xpressed in terms of
(Geneva: 1, FIebruary 1980). current prices.
15. See Bela Balassa, The Newly-Industrializing Coun- 24. Table 3 and the sources cited therein.
tries After the Oil Crisis (Washington, D.C.: World
Bank, forthcoming). 25. Data for the gross domestic product pertain to the
1 6. The differences are larger if comparison is made year 1975 and have been expressed in 1975 prices and
wi.The Canadaand,as fare lasrgher ifscoaresof develop exclange rates. lmport data have been derived from
with Canada and, as far as the share of developing the trade statistics of the industrial countries and
countries in total manufactured imports is concerned, overest�mate the shares of tlong Kong and Singapore
with EFTA. In the case of Canada, protection against in the total bth including their re-exports.
LDC products affected the outcome. In turn, the
results for the EFTA countries reflect the importance
of the United States as a supplier. 26. Bela Balassa, 'Prospects for trade in manufactured
goods between industrial and developing countries,
17. 'The Tokyo Round and the developing countries', 1978- 1990' (Washington, D.C.: World Bank, June
op. cit. 1980), mimeo.
18. For sources, see Table 2. 27. ibid.
THE WORLD BANK
Headquarters:
1818 H Street, N.W.
Washington, D.C. 20433, U.S.A.
European Office:
66, avenue d'I�na
75116 Paris, France
Tokyo Office:
Kokusai Building,
1-1 Marunouchi 3-chome
Chiyoda-ku, Tokyo 100, Japan
The full range of World Bank publications, both free and for sale, is
described in the World Bank Catalog of Publications, and of the continuing
research program of the World Bank, in World Bank Research Program: Ab-
stracts of Current Studies. The most recent edition of each is available with-
out charge from:
PUBLICATIONS UNIT
THE WORLD BANK
1818 H STREET, N.W.
WASHINGTON, D.C. 20433
U.S.A.