20565
W O
W R L
O R L D
D A N
B A
B K
N K O P
O P E
E R
R A O N
T II O
A T N S
S V A
E V
E A L U A
L U A T
T II O
O N
N D E
D P A
E P A R T M
R T M E N T
E N T
Uganda
Post-Conflict
Reconstruction
Country Case Study Series
W O R L D B A N K O P E R A T I O N S E V A L U A T I O N D E P A R T M E N T
Uganda
Post-Conflict
Reconstruction
Country Case Study Series
Alcira Kreimer
Paul Collier
Collin S. Scott
Margaret Arnold
2000
The World Bank
www.worldbank.org/oed Washington, D.C.
Copyright © 2000
The International Bank for Reconstruction
and Development/THE WORLD BANK
1818 H Street, N.W.
Washington, D.C. 20433, U.S.A.
All rights reserved
Manufactured in the United States of America
The opinions expressed in this report do not necessarily represent the views of the World Bank or
its member governments. The World Bank does not guarantee the accuracy of the data included in
this publication and accepts no responsibility whatsoever for any consequence of their use. The
boundaries, colors, denominations, and other information shown on any map in this volume do not
imply on the part of the World Bank Group any judgment on the legal status of any territory or the
endorsement or acceptance of such boundaries.
The material in this publication is copyrighted. The World Bank encourages dissemination of its
work and will normally grant permission promptly. Permission to photocopy items for internal or
personal use, for the internal or personal use of specific clients, or for educational classroom use
is granted by the World Bank, provided that the appropriate fee is paid directly to the Copyright
Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A., telephone 978-750-8400,
fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For
permission to reprint individual articles or chapters, please fax your request with complete
information to the Republication Department, Copyright Clearance Center, fax 978-750-4470.
All other queries on rights and licenses should be addressed to the Office of the Publisher, World
Bank, at the address above, or faxed to 202-522-2422.
ISBN 0-8213-4682-2
Library of Congress Cataloging-in-Publication Data
Kreimer, Alcira, 1941–
Uganda: post-conflict reconstruction: country case evaluation/Alcira Kreimer...[et al.].
p. cm.—(Operations evaluation studies)
At head of title: World Bank Operations Evaluation Department, OED.
Includes bibliographical references.
ISBN 0-8213-4682-2
1. World Bank—Uganda. 2. Economic assistance—Uganda. 3. Uganda—Economic conditions—
1979– 4. Uganda—History-1979– I. Kreimer, Alcira. II. World Bank. Operations Evaluation Dept.
III. World Bank operations evaluation study.
HG3881.5.W57 U33 2000
338.91’096761—dc21 00-024903
Printed on recycled paper.
Table of Contents
Preface vii
Acknowledgments xi
Abbreviations and Acronyms xiii
1. Summary and Lessons Learned 1
Main Strengths and Comparative Advantages:
Stabilizing and Rebuilding the Economy 3
Partnerships with Other Donors, International Organizations,
and NGOs 5
Rebuilding Human, Social, and Cultural Capital 7
The Bank’s Institutional Arrangements 9
Implications for Monitoring and Evaluation 10
Summary of Main Findings of Individual Post-Conflict
Case Studies 11
2. Evolution of the Conflict 17
Independence and the Obote Government, 1963–71 17
The Amin Era, 1971–79 17
The Obote II Administration, 1980–85 18
The National Resistance Movement, 1986 to the Present 19
3. The Bank’s Role in Brief 21
Post-Conflict Reentry, 1979 and 1986 21
Lending for Recovery, 1986–97 24
4. The Bank’s Role in Aid Coordination 27
The Bank’s Relationship with Donors, NGOs,
and the International Community 27
Information and Public Relations 29
iii
Uganda: Post-Conflict Reconstruction
5. The Bank’s Role in Stabilizing, Reforming,
and Rebuilding the Economy 31
The Legacy of Civil War and Its Economic
Implications for Development Strategy 31
Effects on the Different Types of Capital 31
Effects on the Structure of Economic Activity 37
The Bank’s Recovery Lending Program 38
Three Bank Adjustment Lending Interventions 39
Suggested Areas for Improvement in Bank Performance 39
Low Taxation: The Better Option for Post-Conflict Economies 42
The Five Negative Economic Costs of High Taxation
in Post-Conflict Economies 44
The Bank’s Role in Rebuilding Infrastructure
and Economic Sectors 45
6. The Bank’s Role in Rebuilding Human and Social Capital 49
Human Development 49
Governance, Transparency, and Civil Service Reform 52
Participation and Civil Society 53
7. Special Issues 55
The Bank’s Role in Ongoing Conflict:
Northern Uganda Reconstruction Project 55
Security, Demobilization, and Reintegration 58
Bank Evaluations and Findings 62
OED Evaluation of Economic Recovery Credits 65
8. Conclusions: Post-Conflict Lessons
from the Uganda Experience 67
The Post-Conflict Context in Uganda and
Implications for Bank Activities 67
The Bank’s Overall Role and Comparative Advantage 68
The Bank’s Role in Aid Coordination 69
The Bank’s Role in Stabilizing, Reforming,
and Rebuilding the Economy 70
The Bank’s Role in Rebuilding Infrastructure
and Economic Sectors 71
The Bank’s Role in Rebuilding Human and
Social Capital 71
The Bank’s Role in Ongoing Conflict:
Northern Uganda Reconstruction Project 72
Special Issues 73
iv
Table of Contents
Annex: List of Persons Interviewed 75
Endnotes 77
Bibliography 79
Tables
3.1 Post-Conflict Lending Operations in Uganda 22
5.1 Private Investment in Uganda 34
5.2 The Composition of Ugandan GDP by War Vulnerability 37
v
Preface
Following its creation in 1946, the World Bank’s first loan to the
government of France was to help rebuild the country after the massive
destruction suffered in World War II. Since then, post-conflict recon-
struction has been a recurring theme in the Bank’s work. Given the
rapidly increasing number of post-conflict areas and the enormity and
complexity of rebuilding in each case, the Bank is taking a fresh look
at how it can best provide assistance to such troubled parts of the
globe. One way is take a look at past experiences in dealing with post-
conflict reconstruction in an attempt to draw some key lessons for
ongoing and future operations in post-conflict areas.
This review of Uganda is one of a series of field-based country case
studies assessing the Bank’s recent experiences with post-conflict recon-
struction. Other studies in this series include El Salvador and Bosnia and
Herzegovina.
Why choose Bosnia and Herzegovina, El Salvador, and Uganda for
field-based case studies of post-conflict reconstruction? Indeed, the selec-
tion of countries for case study was not self-evident, precisely because
each complex emergency is unique and because every situation presents
a special kind of learning opportunity. A number of considerations went
into these choices.
First, in each of these countries the Bank played a significant role in
attempting to assist with post-conflict reconstruction. Second, the coun-
tries are diverse in the causes of state failure and collapse, as well as the
factors that influenced the initiation or resumption of Bank operations.
Third, the case study countries are also diverse in terms of region and
phase of Bank assistance. For example, Bosnia and Herzegovina repre-
vii
Uganda: Post-Conflict Reconstruction
sents a relatively new recipient of post-conflict reconstruction assistance,
El Salvador is a more mature recipient of such assistance, and Uganda
presents a case in which post-conflict reconstruction assistance has been
essentially completed.
The field studies began with a review of project documentation and a
preliminary series of interviews with relevant Bank staff to identify the
main issues affecting recovery in each country. The in-country research
involved visits to project sites and interviews with a wide range of
respondents, including, for example, representatives from the central
bank; the prime minister’s office; the ministries of finance, education,
health, and reconstruction; donor and United Nations (U.N.) agencies;
international and local nongovernmental organizations (NGOs); parlia-
ment; veterans’ associations; and district administrations.
Overview of the Uganda Case Study
This case study assesses the Bank’s experience in assisting post-
conflict reconstruction in Uganda. Uganda’s civil war spanned a
period from 1972 to 1986, with a short-lived attempt at peace and
reconstruction during the second Obote administration from 1980
to 1985, following the overthrow of the Amin regime. With the
victory of the National Resistance Army (NRA) and the start of
President Yoweri Kaguta Museveni’s administration in 1986, inter-
national reconstruction assistance resumed. The Bank has been
intensively involved in reconstruction efforts, with more than 40
projects across a broad range of activities and sectors amounting
to around US$2 billion (including cofinancing).
Close collaboration between the Ugandan government and the
international community and strong government ownership have given
the country’s recovery the reputation of a “model reconstruction.”
While progress at the governmental and commercial center of Uganda
has been steady, much remains to be done in the country as a whole.
The poor outlying rural districts have shown little improvement in
basic social and economic conditions. In many areas, particularly in
the north, where conflict continues, pre-war conditions and levels of
livelihoods have still not been restored. Investments in social sectors,
such as education and health, have failed to improve overall levels of
service delivery. In addition, reconstruction measures have had to be
adjusted and adapted to respond to a recent government switch to
decentralization and to renewed threats to stability stemming from
conflict in the north and border tensions in the south. Reflecting these
viii
Preface
continuing challenges, the prevailing view inside the country is that
post-conflict recovery still has far to go.
An important consideration to keep in mind with respect to post-
conflict reconstruction is the differing timeframes, expectations, and
political necessities that influence the operations of both donors and aid
recipient governments. In this case, there are two separate but interact-
ing post-conflict time periods. The first is the actual time in which
economic and social recovery is expected to take place; in Uganda, it
is estimated to take two decades or more, as private and social capital
are gradually restored. By contrast, the donor timetable for policy re-
form and reconstruction assistance is more limited; it tends to be driven
by mandates and shorter-term domestic political considerations.
Although on balance successful, shortcomings in the Bank’s recon-
struction program can be attributed to this disparity of timeframe and
expectation. The Ugandan government, for its part, has operated on an
even more compressed timetable, one dictated by a third and more press-
ing consideration—political necessity. But, much to its credit, the
government’s support for policy reforms and the restoration of the Asian
community typify the kind of political commitment that has been vital
to the overall success of the reconstruction program.
ix
Acknowledgments
This report synthesizes the findings of an assessment of the World
Bank’s experience with post-conflict reconstruction. The objective of the
assessment is to distill lessons for ongoing and future operations from
the Bank’s experience in providing assistance for post-conflict recon-
struction. The outputs of the study include an overview and three coun-
try case studies, including this volume. The overview is entitled “The
World Bank’s Experience with Post-Conflict Reconstruction” (1998).
The reports were prepared by a team led by Alcira Kreimer and com-
prising John Eriksson, Robert Muscat, Margaret Arnold, and Colin Scott.
The following consultants contributed to the country case studies: Ann
Elwan, Bosnia and Herzegovina; Jose Marques and Mauricio Silva, El
Salvador; and Paul Collier and Zerubabel Ojoo, Uganda.
The report was produced as part of the OEDPK publication series by
a team under the direction of Elizabeth Campbell-Pagé. Barbara Balaj
provided editorial and writing support for the country cases. Caroline
McEuen (editor), Kathy Strauss and Aichin Lim Jones (graphics and
layout), Barbara Yale and Diana Qualls (editorial assistants), and Juicy
Qureishi-Huq (administrative assistant) comprise the publishing team.
Director-General, Operations Evaluation: Robert Picciotto
Director, Operations Evaluation Department: Elizabeth McAllister
Manager, Sector and Thematic Evaluations: Gregory Ingram
Task Manager: Alcira Kreimer
xi
Abbreviations and Acronyms
AIDS Acquired immunodeficiency syndrome
CARE Cooperative for Assistance and Relief Everywhere
CAS Country Assistance Strategy
CEM Country Economic Memorandum
CG Consultative Group
DANIDA Danish International Development Agency
EDP Enterprise Development Project
EFMP Economic and Financial Management Project
ERC Economic Recovery Credit
ERL Emergency Recovery Loan
ERP Economic Recovery Program
ESW Economic and sector work
EU European Union
GDP Gross domestic product
GNP Gross national product
HIV Human immunodeficiency virus
IBRD International Bank for Reconstruction and Development
ICB International competitive bidding
ICR Implementation Completion Report
IDA International Development Association
IMF International Monetary Fund
NGO Nongovernmental organization
xiii
Uganda: Post-Conflict Reconstruction
NITEP Northern Integrated Teacher Education Program
NRA National Resistance Army
NRM National Resistance Movement
NURP Northern Uganda Reconstruction Project
OD Operational Directive
ODA Overseas Development Association
OED Operations Evaluation Department
Oxfam Oxford Committee for Famine Relief
PAPSCA Program for Alleviation of Poverty and Social
Costs of Adjustment
PEP Public Enterprise Project
SAC Structural Adjustment Credit
SAL Structural Adjustment Loan
SSRP Social Sector Rehabilitation Project
TA Technical Assistance
UEB Uganda Electricity Board
U.N. United Nations
UNDP United Nations Development Programme
UNICEF United Nations Children’s Fund
URA Uganda Revenue Authority
UPE Universal public education
USAID United States Agency for International Development
VAB Veterans’ Assistance Board
xiv
Summary and Lessons Learned
1
From Bretton Woods to the present day, the Bank has taken up the task
of post-conflict reconstruction. Some of the first loans the Bank made after
its founding helped to rebuild European countries after World War II.
During that era, the Bank concentrated on providing physical capital.
Today’s challenges are quite different. The end of the cold war and the
virtual explosion of civil conflicts in the 1990s have tested the ability of
the Bank and the entire international community to address the devasta-
tion of human and social capital on an unprecedented scale.
Post-Conflict Lending by Region
The most significant increase came in the first half of the 1990s, when
the majority of funds went to African countries (mainly Uganda). Of the
US$6.2 billion the Bank has lent to post-conflict countries since 1980, 56
percent, or US$3.5 billion, has gone to the Africa Region. The next
largest share has gone to the Latin Ameri-
can and Caribbean Region, which received Since 1980, the volume of Bank
just over 16 percent. The Europe and Cen- lending to post-conflict countries has
tral Asia and Middle East and North Africa increased more than 800 percent, to
Regions have received nearly equal shares, US$6.2 billion, and has touched every
region and economic sector.
about 11 percent each, and the East Asia
and Pacific Region received 4.5 percent. The
South Asia Region received the smallest portion, just over 1 percent,
consisting entirely of the Emergency Reconstruction Loan to Sri Lanka
in 1988, following the peace accords.
Over the years, lending commitments to post-conflict countries have
grown as a percentage of the Bank’s overall portfolio. Overall lending to
post-conflict countries constituted more than 16 percent of the lending
1
Uganda: Post-Conflict Reconstruction
commitments of the International Bank for Reconstruction and Develop-
ment (IBRD) and the International Development Association (IDA) for
fiscal year 1998. In several Regions, such as Europe and Central Asia and
the Middle East and North Africa, lending to post-conflict countries makes
up nearly half of the Regional portfolio. In Africa, post-conflict lending
makes up close to a third of commitments.
Post-Conflict Lending by Sector
Over the past 20 years, the Bank’s post-conflict reconstruction lend-
ing projects have covered every sector. Of the US$6.2 billion in lending
volume, the largest portion, 32.6 percent, supported multisector projects.
Multisector projects include central bank transfers to stabilize the mac-
roeconomic situation, technical assistance loans, and general emergency
reconstruction or recovery projects that have several economic and
social components designed to provide rapid assistance for the most press-
ing needs. Of the nearly US$2 billion in multisector lending, more than
half was in structural adjustment loans or credits for budget support.
The multisector also includes the demining implemented in Bosnia.
Nearly 10 percent of the US$6.2 billion financed agricultural projects,
and 9 percent supported transportation projects (mainly highway recon-
struction). The urban development sector received 8.5 percent of post-
conflict reconstruction assistance. Population, health, and nutrition
received 7.4 percent of the lending amount, and electric power and other
energy received 6.6 percent. Other sectors received smaller amounts of
support—education, 4.2 percent; water supply and sanitation, 4.2 per-
cent; and so on.
Need for Evaluation
In view of both the high risk and attendant potential returns entailed
in an expanded role for the Bank in post-conflict reconstruction, the
Bank’s Board of Executive Directors called on the Operations Evalua-
tion Department (OED) to assess recent and ongoing Bank operations in
such areas. The objective of this assessment is to distill lessons for ongo-
ing and future operations in post-conflict reconstruction. Specifically, it
aims to enhance the institution’s ability to respond more effectively and
efficiently to the needs of societies rebuilding after conflict.
Case Study Selection
Three cases were chosen for field study: Bosnia and Herzegovina, El
Salvador, and Uganda. The selection of countries for field case study
was not self-evident precisely because each complex emergency is unique
2
Summary and Lessons Learned
and because every situation presents a special kind of learning opportu-
nity. A number of considerations went into these choices.
First, in each of these countries the Bank played a significant role in
attempting to assist with post-conflict reconstruction. Second, the countries
are diverse in terms of the causes of state failure and collapse, as well as the
factors that influenced the initiation or resumption of Bank operations. Third,
the case study countries are also diverse in terms of region and phase of
Bank assistance. For example, Bosnia and Herzegovina represents a rela-
tively new recipient of post-conflict reconstruction assistance; El Salvador is
a more mature recipient of such assistance, and Uganda presents a case in
which post-conflict reconstruction assistance has been essentially completed.
The Bank’s Post-Conflict Response Mechanisms
Every post-conflict country is unique, and no single formula can re-
spond to all reconstruction needs. The Bank has used a broad array of
mechanisms to support the transition to peace and the resumption of eco-
nomic and social development. Its lending operations have consisted of
macroeconomic and sectoral adjustment reforms, direct investment, and
technical assistance in support of reconstruction. Nonlending services have
included damage and needs assessments, economic and sector work (ESW),
Country Assistance Strategies (CAS), and aid coordination.
Key Issues Analyzed
In the case studies of Bosnia and Herzegovina, El Salvador, and
Uganda, six key issues were analyzed, including the Bank’s: main
strengths or comparative advantages; partnerships with other donors,
international organizations, and nongovernmental organizations (NGOs);
role in reconstruction strategy and damage and needs assessments; role
in rebuilding the economy and institutions of governance; internal man-
agement of resources and processes; and implications for monitoring
and evaluation.
Main Strengths and Comparative Advantages: Stabilizing
and Rebuilding the Economy
Supporting Macroeconomic Stabilization
Supporting the achievement of macroeconomic stabilization is an
area of comparative advantage for the Bank and should be one of its
highest priorities in post-conflict situations. Moving to macroeco-
nomic stability as soon as possible in a post-conflict setting is crucial
to economic recovery.
3
Uganda: Post-Conflict Reconstruction
The Bank’s performance in supporting fiscal and other structural eco-
nomic reforms, such as privatization and tax
The case studies confirmed that on the policy, has been mixed. Thus, the pursuit of
grounds of relevance, efficacy, and
economic reforms should continue incremen-
efficiency, stabilizing the economy is
the area of strongest Bank perfor- tally, taking into account the country’s his-
mance in post-conflict reconstruction. torical and current policy and institutional
enabling environment.
Rebuilding Physical Infrastructure
The Bank’s assistance for rebuilding physical infrastructure, includ-
ing the policy and institutional aspects, has been another strong area of
performance. The areas typically most important to facilitating recov-
ery are roads, transportation, power, telecommunications, basic hous-
ing, and water and sanitation.
Bosnia and Uganda received comprehensive reconstruction support
from the Bank. In Bosnia and Herzegovina, the Bank helped start the
reconstruction effort in 1996, without waiting for financial normaliza-
tion and membership in the Bank. Sixteen emergency projects for the
country addressed all major infrastructure and social sectors. They also
incorporated a focus on employment generation and support for institu-
tional development.
The Bank’s involvement in Uganda’s reconstruction was particularly
comprehensive. Since the National Resistance Movement (NRM) gov-
ernment took power in January 1986, the Bank has coordinated closely
with international donors to support the country’s reconstruction effort.
In the first five year after the conflict (1987–92), the Bank supported
approximately 25 lending operations amounting to more than US$1 bil-
lion. It also provided financing for an Economic Recovery Credit (ERC)
and a series of sector investment and reform programs (agriculture, edu-
cation, health, railways, telecommunications) aimed at rehabilitating
key economic and social infrastructure affected by the war.
In El Salvador, several high-quality Bank lending operations and non-
lending services made significant contributions to post-conflict recon-
struction and economic recovery.
Supporting Basic Production Sectors
The Bank’s approach to assisting basic production sectors, such as
agriculture and industry, has been for the most part indirect, concentrat-
ing on policy and institutional enabling environments and on infrastruc-
ture. Operations intended to strengthen public sector entities in agriculture
in El Salvador encountered protracted problems in attempting to work
4
Summary and Lessons Learned
with weak, inefficient, or rigid bureaucracies. This led to the canceling
or restructuring of operations and to a search for alternative approaches
to such functions as research, extension, and service and input delivery,
relying more on the private sector and NGOs. The performance of
operations that have assisted intermediary organizations in support of
microenterprise and other participatory local development efforts has
been mixed. The record has been relatively satisfactory in Bosnia and
Herzegovina, and less satisfactory in Uganda.
Demining
Demining is unique to post-conflict situations and essential for reha-
bilitation and recovery. Although there are strong economic and
humanitarian reasons for demining, the main lesson learned from this
assessment is that Bank involvement should focus primarily on indirect,
nonclearance activities such as coordination, information and mine aware-
ness, training, and institution building.
Partnerships with Other Donors, International
Organizations, and NGOs
Consultative Group Meeting Benefits
The Bank’s use of consultative groups has been particularly effective
for mobilizing aid resources, including facilitating the clearing of ar-
rears (as in Bosnia), seeking a coordinated approach to macroeconomic
issues, and providing information on needs and assistance flows. Some
of the benefits perceived by aid recipients of a Bank role in aid coordina-
tion included the quality of its analytical and advisory services, its unique
relationship with the International Monetary
Fund (IMF), its credibility in the eyes of the Early leadership or participation in
rest of the donor community, and its poten- damage and needs assessments in
tial to mobilize funds from other sources. post-conflict countries has facilitated
the Bank’s subsequent involvement in
It helped enormously to have the Bank’s developing reconstruction strategies
leadership role clearly defined and mandated and in aid coordination.
by key members of the international com-
munity, as well as strongly backed by top Bank management, as in the
case of Bosnia and Herzegovina after the Dayton Accords. Even in the
absence of such a designation, it is important to have a clear request
from the government early on, as in El Salvador. It was also valuable
for the government and the main donors to achieve a consensus on a
policy framework for sustainable recovery and reconstruction, as in Bos-
nia and Herzegovina and Uganda. In El Salvador, holding a CG meet-
5
Uganda: Post-Conflict Reconstruction
ing before the peace accords were signed contributed to the effectiveness
of coordination.
Coordination in Undertaking Damage
and Needs Assessments
One area where a coordinated, joint approach among donors was
particularly desirable was in undertaking damage and needs assessments,
as illustrated by the Bosnia and Herzegovina experience. A joint effort
in this area is not only more efficient but can also help set the stage for
effective cooperation and coordination between donors and governments
when implementing post-conflict reconstruction initiatives. Assistance
for preparing a national reconstruction plan should also be coordinated
to avoid the imposition of unnecessary burdens on the government.
The resource mobilization function of aid coordination may lead to the
cofinancing of operations. This is particularly desirable for general pro-
gram assistance for balance of payments and budget support, for which
conditions and tranches must be synchronized. However, the Bank had
difficulty in mobilizing adequate cofinancing for budget support in Bosnia
and Herzegovina and El Salvador. Some donors were ambivalent toward
the governments, because of concern either about the depth of their com-
mitment to peace accords or about their accountability and legitimacy.
In-Country and Sectoral Coordination
Another challenging aspect of coordination for the Bank has been at the
in-country and sectoral levels. Misunderstandings and differences can arise
with other donors that have substantial resources, those that have had a
long-standing coordination role, or those that have both. The case of Bos-
nia and Herzegovina demonstrates the importance of partnerships in the
reconstruction process. The sector task forces, for example, were credited
with success in information-sharing within sectors and with some success
in avoiding contradiction and overlap among donors.
Coordination with NGOs
The Bank’s coordination with NGOs, both local and international,
and its relationship with civil society in general are also important. A
strong resident mission with an NGO liaison officer in Bosnia and
Herzegovina, for example, facilitated these relationships. The Bank is
seen as an effective bridge between the government and NGOs there.
The Bank’s rather rough but improving relationship with Salvadoran
NGOs could arguably have been more harmonious sooner had there
been a resident mission.
6
Summary and Lessons Learned
Despite the presence of a resident mission in Uganda, the Bank has
been criticized by NGOs and some U.N. sources for its coordination and
partnership role there. Some felt that the
Bank’s leadership had focused the develop- Partnerships are key to effective aid
ment community on macroeconomic reforms coordination. It is critical to establish
at the expense of poverty alleviation pro- at the outset of the recovery process
grams. In addition, the Bank was criticized what each donor will do.
for employing a “standard economic model
approach” to coordination rather than adopting a special post-conflict
approach. NGOs also felt that the rhetoric of partnership and participa-
tion “was moving faster than reality.”
Rebuilding Human, Social, and Cultural Capital
Human Capital
Investment in human capital, including education and health spend-
ing, is an important component of post-conflict work. Violent conflict
can decimate the human resources of a country as people are killed,
maimed, or displaced in large numbers. Human capital services are
typically the first to be disrupted by conflict. Education, health, and
community services stop, bringing the realization of human potential to
a halt. Schools, hospitals, clinics, and community centers are destroyed,
as is the government’s capacity to administer services. Conflict also cre-
ates newly vulnerable groups, such as the unemployed, ex-combatants,
households headed by women, children, and the disabled, who are
legitimate beneficiaries of reconstruction aid for socioeconomic as much
as for humanitarian reasons.
Human capital or social sector work was not generally a priority in
the Bank’s post-conflict projects or in its ESW, both of which concen-
trated initially on macroeconomic, sector, and infrastructure work. El
Salvador and Bosnia and Herzegovina were
exceptions, with the latter portfolio contain- The restoration and development of
ing an early and balanced inclusion of so- human capital in the post-conflict
cial sector work. In Bosnia, four of the sixteen phase is essential to establishing a
projects contained social sector components. base for economic rebuilding.
Uganda operations produced unsatisfactory
results, notwithstanding some effort to prioritize health, education, and
economic support for the rural poor and vulnerable groups.
The El Salvador Basic Education Modernization Project was built on a
Bank-supported Community Managed Schools Program (EDUCO) pilot
project and employed a community-managed approach, initiated during
7
Uganda: Post-Conflict Reconstruction
the conflict by the communities themselves. The participatory nature of the
El Salvador program contributed to consensus building and the sustainabil-
ity of the peace process. Strong government and Bank commitment have
also been important for performance and the achievement of objectives.
Social Capital
Inherent in violent civil conflict is the destruction of social capital,
particularly institutions of governance and civil society and such basic
attitudes and behaviors as trust and participation. While the severity of
this problem in post-conflict countries is increasingly recognized, nei-
ther the Bank nor any other international donor has an obvious com-
parative advantage in this area. Analyses undertaken by the Bank and
others have identified some key components of good governance and
vigorous civil society that need to be addressed. These components
include transparency, accountability, the rule of law, and the profes-
sions. Several recent Bank-supported pilots
Reconstruction efforts must consider (as in Bosnia and Herzegovina), and World
the possible productive difficulties Development Institute efforts (as in Uganda)
women face in post-conflict show promise in these areas.
situations (such as lack of labor at
critical times in the agricultural
season), since in many post-conflict The Role of Women
countries a third or more of the The role of women in rebuilding social
working-age men have been killed capital should also be examined and
and women are the productive basis exploited. Often considered a “vulnerable
for restarting the economy. group” in post-conflict settings, women have
potential as strong community leaders who
can facilitate the rebuilding of social capital and may be overlooked.
It is only in Bosnia and Herzegovina that the Bank has made a spe-
cific operational effort to address the particular needs of women.
Demobilizing Ex-Combatants
The related activities of the demobilization, reinsertion, and reintegra-
tion of ex-combatants into the civilian economy and society can be impor-
tant to economic recovery and sustained peace. Bank-supported efforts in
this area expanded in the 1990s, and the Bank has played a leadership
role through the analysis of its experience in Africa. The experience of the
Bank and other donors has generated several lessons, as well as some
unresolved issues. For example, when the parties to the peace accords are
not fully committed, one side or the other can manipulate the process. An
approach narrowly targeted on ex-combatants, as opposed to an area or
countrywide approach, can create resentment among local populations.
8
Summary and Lessons Learned
Demobilization can also sometimes be premature, particularly where
a continuing threat to national security exists, as may have been the
case in Uganda. The Uganda experience also demonstrates that the avail-
ability of land is an important determinant of success in the reinsertion
and reintegration of ex-combatants.
Land is not a guarantee of success, however, as illustrated by El Salva-
dor, where a substantial number of ex-combatants who have been pro-
vided land under the accord-mandated Land Transfer Program abandoned
their land for a variety of reasons, including poor land quality, lack of
supporting services and credit, and lack of aptitude and interest.
The Bank has acquired some expertise in public sector downsizing
programs that can be applied to demobilization efforts. Training schemes
have had a mixed record in both areas.
Cultural Capital
What is the justification for assisting in the protection and conserva-
tion of cultural heritage in situations of complex emergencies? Applying
scarce resources to conserving cultural heritage in a post-conflict situa-
tion may seem frivolous at first glance. How-
ever, cultural heritage has the power to In attempting to reestablish civil
inspire hope and remind people of their cre- society in the face of ethnic rivalries,
ativity. Its destruction is a decisive way to protecting cultural heritage is one of
assert primacy and control and can become the tasks requiring attention in any
a symbol of the brutality and insanity of war. assistance strategy.
So sensitive are cultural heritage sites that
they can become flashpoints in ethnic and civil conflicts, as in the bomb-
ing of Bosnia and Herzegovina’s Mostar Bridge.
Although the Bank has carried out a number of projects with cul-
tural heritage conservation components, it has little experience doing
so in post-conflict situations.
The Bank’s Institutional Arrangements
The case studies reveal the great importance of management deci-
sions regarding the staffing and structuring of post-conflict country
teams and the programming, design, and implementation of opera-
tions. The country director must have a mandate to give the country
substantial if not full-time attention. Resident representatives need suf-
ficient authority to make a wide range of programming and implemen-
tation decisions in the field.
Early post-conflict situations require timely and flexible program-
ming, design, and implementation. From the initial base of a well-
9
Uganda: Post-Conflict Reconstruction
prepared damage and needs assessment, the Bank should refine its strat-
egy over the few years that follow through a series of workshops and
policy notes. A full CAS can be prepared
Despite the high up-front costs, when time and resources permit.
well-staffed resident missions are Expeditious preparation, piloting and
a precondition for successful Bank bridging funds, and loan instruments
intervention in post-conflict situations.
should be resourced at sufficient levels to
enable the Bank to be effective earlier in
post-conflict situations. Mechanisms for rapid procurement and dis-
bursement should be devised for post-conflict situations, consistent
with sound practice.
Implications for Monitoring and Evaluation
The effective implementation of post-conflict operations requires inten-
sive monitoring, and the Bank must be prepared to allocate sufficient
administrative budget resources for this task. Monitoring efforts should
also draw on external expertise about a post-conflict country and make
better use of ESW, which should include longitudinal household and com-
munity studies. Considering the need for completion information in often
highly volatile post-conflict settings, the Bank should sharply reduce the
interval between project closing and completion reporting, with appropri-
ate streamlining of the process.
10
Summary and Lessons Learned
Summary of Main Findings of Individual Post-Conflict
Case Studies
Bosnia and Herzegovina
Early, Comprehensive, and Inclusive Approach
The Bank’s response to post-conflict needs in Bosnia and Herze-
govina was early and comprehensive. Its role in reconstruction and
economic recovery is widely perceived to have been successful by
the country’s authorities, other donors, NGOs, and beneficiaries.
The OED team agreed with this assessment. The Bank’s nonlending
activities are appreciated, including providing a framework for
reconstruction and guidance to donors. There is widespread recog-
nition that Bank-supported projects were implemented quickly.
Elements of Success
Other elements of success include a fairly wide dispersion of ben-
efits, both geographically and to a broad range of beneficiaries;
involvement of stakeholders; an early and balanced inclusion of the
social sector projects in order to rebuild human capital; and a con-
tribution to local implementation capacity. The country’s authori-
ties particularly appreciated the sense of ownership afforded them
by Bank projects.
Ability to Move Quickly
The peace implementation agenda dictated a compressed time-
table for launching the reconstruction program. The Bank was able
to adhere to this timetable for a number of reasons: the availability
of planning resources, support from the Dutch Executive Director,
strong and visible support from the Bank’s President, its role in coor-
dinating the planning work, coordination with other actors, and
superb staff quality and dedication.
Highly Qualified and Dedicated Staff
In addition to speed, the special attention given to the program
within the Bank contributed to the overall quality of the program as
well. A very skilled country director was appointed solely for Bos-
nia and Herzegovina. Her commitment to the process, attention to
speed, and quality of work were important in meeting the com-
pressed timetable for project processing. The resident representative
11
Uganda: Post-Conflict Reconstruction
also played an important role in the process. His ability to deal with
differing perspectives and work with all involved parties in the reso-
lution of issues were very much appreciated by the government,
donors, and other agencies.
Successful Project Implementation and Dispersion
of Benefits
The OED team found that the benefits of Bank-assisted projects
to date are being felt throughout the federation and are increas-
ingly reaching Republika Srpska. The range of benefits has been
delivered quickly to a diverse set of beneficiaries while involving
stakeholders and building up local implementation capacity. Sev-
eral factors contributed to success in the implementation phase,
including early establishment of a resident mission, streamlined
project process and procurement procedures, sector diversification,
pilot projects and project preparation, local ownership, and wide-
spread participation.
Economic Management in Peace Agreements
The provisions for fiscal strategy built into the Dayton Accords
are of particular importance for the reconstruction work of donors,
especially for the Bank. The Bosnia and Herzegovina experience
demonstrates the importance of incorporating into a peace accord,
to the extent possible, economic management provisions that are
more likely to enhance than to obstruct the economic recovery pro-
cess. Despite Dayton’s built-in obstacles to effective economic gov-
ernance and the continuing political tensions, the first year and a
half saw substantial economic recovery. In this context, the Bank
has also played a central role in the mobilization and application of
external resources critical to this recovery.
12
Summary and Lessons Learned
El Salvador
Overview
The Bank’s assistance to El Salvador focused mainly on macro-
economic reform but also included projects addressing the health
and education sectors to improve the country’s poor social indica-
tors. Bank assistance was critical to laying a sound macroeconomic
foundation for an impressive economic recovery during the first half
of the 1990s, as well as mobilizing international economic support.
While the Bank lagged behind several other donors in the volume of
its post-conflict assistance, its leadership of a series of four CG meet-
ings between 1991 and 1995 made a vital contribution to post-
conflict reconstruction and economic recovery. Several high-quality
lending operations and nonlending services also made significant
contributions. However, no effort, with or without donor support,
has yet shown any real promise for reversing perhaps the most seri-
ous socioeconomic legacy of the conflict: violent crime, or
“microinsecurity,” of epidemic proportions.
Reengagement before the End of the Conflict
By establishing effective working relationships with the govern-
ment a couple of years before the Peace Accords, the Bank was able
to provide timely macroeconomic assistance and coordination sup-
port to El Salvador’s reconstruction program. The timing of the Bank’s
re-entry in El Salvador was propitious, although not entirely inten-
tional. It is hard to say whether, in the absence of the earthquake
disaster, the pre-Peace Accords relationships between the govern-
ment, the private think-tank Salvadoran Foundation for Economic
and Social Development, and the Bank would have developed as
expeditiously and productively as they did.
Lending Operations
Early operations, including the two SALs and the SSRP, were
well-timed. It is also clear that the Bank pursued comparative
advantages in policy and institutional reform in its programming.
The Success of Post-Conflict Education Lending
In the social sectors, the Bank seized a window of opportunity in
supporting EDUCO, the community self-managed education initia-
13
Uganda: Post-Conflict Reconstruction
tive of the government. This expanding initiative promises to cor-
rect and reverse some basic deficiencies in the pre-conflict educa-
tional system of El Salvador. The EDUCO experience provides a
good example of post-conflict support being more than just recon-
struction but also entailing a redirection of development. However,
the Bank’s experience providing support to the health sector in El
Salvador during the early post-war years has been mixed and has
not yet led to a full follow-on project.
Good Aid Coordination
The donor coordination role played by the Bank just before the
Peace Accords and during the first few years of post-conflict recon-
struction was indispensable. The holding of a CG meeting before the
Peace Accords were signed and the commitment of experienced and
capable Bank leadership to the process contributed significantly to
the effectiveness of coordination. It was also important that the gov-
ernment gave priority to donor coordination and endorsed the Bank’s
supportive role.
Effective Bank Country Team
A high degree of capability, experience, and continuity combined to
make for an effective Bank Country Team in the crucial years just
before and after the El Salvador Peace Accords. This team made a
critical contribution to the Bank’s performance in a tense environment.
Views on establishing a resident mission are mixed, with some believ-
ing it unnecessary, while others favor such a presence to enhance the
Bank’s image and relations with civil society and other donors.
14
Summary and Lessons Learned
Uganda
Comprehensive Involvement and Main Successes
Bank involvement in the reconstruction efforts in Uganda have
been particularly comprehensive. In the first five years after the con-
flict (1987–92), the Bank supported approximately 25 lending
operations amounting to more than US$1 billion and closely coordi-
nated with international donors. The Bank’s role was key in strength-
ening the Ministry of Finance and the Central Bank, removing the
monopoly of the Coffee Board, assisting in sugar rehabilitation, and
rebuilding roads.
Program Shortcomings
Despite good performance in reforming and rebuilding the
economy, there have been several respects in which Bank involve-
ment could have been improved: insufficient attention to consensus
building; excessive use of conditionalities; and, most important, a
seriously dysfunctional emphasis on raising taxation. The Bank did
not always fulfill its potential comparative advantage, for example,
in the power sector, a key element in recovery. Nor did it fully
convert its coordination role into creating an overall strategy for
reconstruction or a sector-by-sector plan.
Social Sector Shortfalls
The Bank’s performance was relatively poor in the social sectors,
particularly in strengthening health and education institutions. In-
vestment in the social sectors was often premature, implemented
through weak ministries, and inadequate forethought to the nature
of the services the Bank was trying to support. The post-conflict
period required major health sector reforms, which fell short of needs.
Education investment was equally disappointing, with only an esti-
mated 37 percent of funds reaching schools. There was too much
dependence on the existing bureaucracy, and not enough use was
made of NGOs.
Bank Processes and Institutional Arrangements
Regarding Bank processes and institutional arrangements, project
design did not fully reflect the need in Uganda’s unsettled institu-
tional environment for a flexible, process-oriented design. This was
particularly evident in the social sectors, where education and health
15
Uganda: Post-Conflict Reconstruction
ministries were too weak to accommodate spending, and superven-
ing events such as decentralization and renewed conflict changed
priorities. Changes in key resident mission staff, task managers, and
government staff were inevitably disruptive at times. Bank staff were
not generally familiar with working in conflict countries or with the
international relief and rehabilitation system. Staff resources and
time were concentrated in Kampala, often with little understanding
or contact with international and local NGOs.
Differing Timetables for Post-Conflict Recovery
A final shortcoming of Bank projects was that, where they were
not sequential, many were too short to address the projected length
of recovery. There are at least two different timetables in post-war
recovery, well-manifested in Uganda. The first is a real-time dura-
tion of the recovery process, which typically requires at least two
decades of sustained effort, with the risk of war a recurrent phe-
nomenon. The second timetable is set by donor considerations,
which may be guided by different objectives and manifests itself
through programs that do not necessarily take into consideration
the very long recovery process. The government has to observe
both timetables, weighing the political expediencies of short-term
measures to provide security and boost confidence with the needs
for longer-term recovery.
16
Post-colonial events in Uganda can be considered according to four time
periods corresponding to the dominant political regimes: the newly indepen-
dent government of Obote, 1963– 71; the Amin era, 1971– 79: Obote II, 1980–
85; and the National Resistance Movement, 1986 to the present.
2
Evolution of the Conflict
Independence and the Obote Government, 1963–71
At independence, Uganda was more fortunate than most of its African
neighbors in that the country had promising economic potential and
comparatively good health and education services. The industrial sector
produced a range of commodities satisfying much of domestic needs.
The export sector was buoyant, especially with respect to cash crops
such as coffee, cotton, and tea. Some minerals, including copper, were
also exported. The social sector, especially education, had a good repu-
tation throughout Africa. The provision of health services was good,
and health facilities were well managed and stocked with medicines.
The civil service, made up of a motivated and trained work force, was
regarded as functioning well. But, trouble was brewing on the political
front. Growing political conflict and Milton Obote’s creation of a one-
party state in the mid-1960s led to increasing tensions, culminating in
the overthrow of Obote in 1971 through a military coup led by Idi Amin.
The Amin Era, 1971–79
The Amin era was characterized by notorious political mismanage-
ment leading to severe economic decline, political instability, and increas-
ing civil violence. An autocratic government took power, banning political
activities, severely intimidating civil society, and eliminating prominent
individuals perceived to be a threat to the regime. Ugandans of Asian
origin, who over the years had dominated and controlled the industrial
and commercial sectors, were expelled and their property confiscated as a
form of political patronage. This loss of entrepreneurs and skilled man-
power resulted in the breakdown of commercial and industrial sector
17
Uganda: Post-Conflict Reconstruction
management. Other skilled Ugandans fled the country and were replaced
by illiterate soldiers and civilians who then occupied managerial posi-
tions. Standards in the civil service eroded. Shortages of goods became
widespread. Much of the earnings from the coffee boom of the mid-1970s
were used for military hardware and other imported merchandise. The
military overthrow of Amin by a combined force of the Tanzanian Army
and Ugandan exiles marked the peak of destruction. More social and
economic infrastructure was destroyed, and the educated and managerial
classes embarked on another wave of flight from the country.
Following the demise of Amin, there was much political uncertainty.
Competition emerged among the many exiles who participated in the
overthrow. As such, they failed to achieve the cohesiveness and political
unity needed to support the reconstruction of a shattered economy and
society. This difficult situation was exacerbated by controversies sur-
rounding the 1980 elections. The Obote II government was formed, and
conflicts soon started in Luwero and neighboring districts—the focus of
opposition groups—and to some extent in West Nile by Amin soldiers.
Lives and property were lost on a greater scale than in the 1970s, and
reconstruction efforts came to naught.
The Obote II Administration, 1980–85
Leaders in the Obote II administration were drawn from a range of
military and civilian backgrounds and political leanings, which made it
difficult to resolve conflicts. Many political leaders lacked the capacity,
experience, and skill to meet the challenges of reconstructing an economy
mismanaged for so many years, let alone a civil society that had under-
gone years of intimidation and the breakdown of ethics.
In such circumstances, the contribution of donors in conflict-resolution
and reconstruction efforts was limited. Foreign exchange rationing contin-
ued, and inflation was high. The gross domestic product (GDP) continued
to fall, and the industrial sector never revived. The appointment of indi-
viduals through political patronage continued. The civil service was poorly
motivated and failed to deliver public goods and services. Above all, the
military was undisciplined in areas of conflict and security. The military
overthrow of the Obote II government in July 1985 and the subsequent
military defeat of the Okello Junta in January 1986 marked the end of the
widespread civil strife that had engulfed Uganda since 1971.
18
Evolution of the Conflict
The National Resistance Movement, 1986 to the Present
The National Resistance Movement (NRM) assumed power in Janu-
ary 1986, inheriting a shattered economy with a GDP that had shrunk
by more than 20 percent from its peak value in 1970. Inflation was
raging at an annual rate in excess of 240 percent. In addition, coffee
earnings contributed more than 70 percent of export earnings.
The NRM embarked on a stabilization and structural adjustment pro-
gram in 1987, with the assistance of the Bank and other donors. Priori-
ties included restoring macroeconomic stability and taming inflation.
These goals proved elusive until 1992–93, when the government adopted
a strict cash budget rule aimed at preventing the printing of money to
finance deficits. Subsequent results have received international recogni-
tion. Inflation has fallen to below 10 percent. Economic growth rates
have averaged 6.4 percent a year for eight years, which translates into
an annual increase of 3.4 percent in average incomes since the structural
adjustment program began. The government has implemented a wide
range of policy and institutional reforms that have increased incentives
to work, save, and invest.
Many observers have noted the persistence and patience of the Bank in
this reconstruction effort. The NRM government was initially wary of the
Bank following controversy in some African countries that experienced
difficulties with structural adjustment programs. Since 1992, the govern-
ment, including the President, has demonstrated a stronger commitment
to economic reform, and liberalization of the economy has accelerated.
19
Bank involvement in the reconstruction efforts of Uganda has been particu-
larly comprehensive. In the first five years after the conflict (1987– 92), the
Bank, in close coordination with international donors, supported approxi-
mately 25 lending operations amounting to more than US$1 billion. Bank
support focused on assisting the government in achieving economic stabil-
ity and in promoting long-term economic growth. The Bank has also been
instrumental in rebuilding social and physical infrastructure. In this context,
the Bank has also encouraged the Ugandan Diaspora to return in order to take up the task of
rebuilding the nation. Still other Bank initiatives in Uganda have focused on urgent social con-
cerns, such as primary education and targeting assistance to vulnerable groups. From 1987 to 1997,
the Bank’s entire lending portfolio focused primarily on reconstruction and rehabilitation.
3
The Bank’s Role in Brief
From 1980 to 1985 and from 1987 to 1996, the Bank had an active
portfolio in Uganda (see table 3.1). The goal of its initiatives was to
assist the government with both the immediate needs of reconstruction
and demobilization of the military and as the longer-term goals of eco-
nomic stability and growth.
Post-Conflict Reentry, 1979 and 1986
The first attempted reentry by the Bank was during the post-Amin era
after 1979. Unfortunately, it was cut short by the onset of further con-
flict. According to most observers, there was little the Bank could have
done differently to avert the renewal of conflict or to play a more effec-
tive role. The post-1986 reentry, in contrast, raised a number of issues
about the choice and timing of Bank interventions. Despite the hiatus
resulting from renewed conflict in the early 1980s, earlier initiatives
were “carried over” to some extent.
A Bank reconnaissance mission visited Uganda in July 1979 to identify
country assistance needs. The government requested the Bank’s support in
five specific areas: (a) a reconstruction credit to provide emergency for-
eign exchange, (b) the reactivation of the project pipeline, (c) assistance in
coordinating aid, (d) assistance in development planning and project prepa-
ration, and (e) the establishment of a resident mission in Kampala.
The Bank launched economic reconnaissance and appraisal missions
for the First Reconstruction Credit in September 1979. The initial US$55
million credit was in the form of a quick-disbursing Economic Recovery
21
Uganda: Post-Conflict Reconstruction
Table 3.1: Post-Conflict Lending Operations in Uganda
Loan/credit amt. Percent of
Project name Fiscal year (US$ million) total
Agriculture
Agricultural Rehabilitation 1983 66.1
Agricultural Development (IFAD II) 1985 9.1
Forestry Rehabilitation 1987 13.0
Southwest Agricultural Rehabilitation 1988 9.3
Sugar Rehabilitation 1988 24.9
Agricultural Sector Adjustment 1990 115.0
Livestock 1990 21.0
Agricultural Research and Training 1992 25.0
Cotton Sector Development 1994 14.0
Agricultural Extension 1996 15.8
Agricultural Sector Management 1996 17.9
Total, Agriculture Sector 331.1 14.4
Education
Education II 1983 32.0
Education IV 1988 22.0
Primary Education and Teacher Training 1993 52.6
Total, Education Sector 106.6 4.6
Environment
Environmental Management 1995 11.8
Lake Victoria Environment 1996 12.1
Total, Environment Sector 23.9 1.0
Health
Health I 1988 42.5
Sexually Transmitted Diseases 1994 50.0
District Health 1995 45.0
Total, Health Sector 137.5 6.0
Power and Energy
Petroleum Exploration Promotion 1985 4.5
Power II 1985 28.8
Power III 1991 125.0
Total, Power and Energy Sector 158.3 6.9
Reconstruction and Structural Adjustment
Reconstruction Credit 1980 72.5
Reconstruction Credit II 1982 70.0
Reconstruction Credit III 1984 16.7
Economic Recovery I 1987 65.0
Economic Recovery Credit I SFA 1989 12.8
Economic Recovery Credit II Supplement 1989 25.0
Economic Recovery II 1990 125.0
Structural Adjustment Credit I 1991 125.0
22
The Bank’s Role in Brief
Table 3.1: Post-Conflict Lending Operations in Uganda (cont.)
Loan/credit amt. Percent of
Project Name Fiscal year (US$ million) total
Reconstruction and Structural Adjustment (cont.)
Northern Reconstruction 1992 71.2
Structural Adjustment Credit II 1994 80.0
Structural Adjustment Credit III 1997 125.0
Total, Reconstruction and Structural 788.2 34.2
Adjustment Sector
Technical Assistance
Technical Assistance 1980 8.0
Technical Assistance II 1983 51.3
Technical Assistance III 1988 18.0
Economic and Financial Management 1992 29.0
Institutional Capacity Building 1995 36.4
Total, Technical Assistance Sector 142.7 6.2
Telecommunications
Posts and Telecommunications
Rehabilitation 1983 22.0
Telecommunications II 1989 52.3
Total, Telecommunications Sector 74.3 3.2
Transportation
Highways III 1984 51.3
Highways IV 1987 16.7
Railways I 1989 7.0
Transport Rehabilitation 1994 75.0
Total, Transportation Sector 150.0 6.5
Water Supply and Sanitation
Water Supply Engineering 1981 9.0
Water Supply and Sanitation
Rehabilitation 1984 28.0
Water Supply II 1990 60.0
Small Towns Water 1994 42.3
Total, Water Supply and Sanitation Sector 139.3 6.0
Other
Phosphate Engineering 1982 3.9
Public Enterprises 1988 15.0
Poverty and Social Adjustment Costs 1990 28.0
Enterprise Development 1991 65.6
Urban I 1991 28.7
Financial Sector Adjustment 1993 100.0
Private Sector Competitiveness 1995 12.3
Total, other sectors 253.5 11.0
TOTAL 2,305.4 100.0
23
Uganda: Post-Conflict Reconstruction
Loan designed to transfer resources rapidly to meet short-term foreign
exchange needs and to supply inputs for productive sectors. In addition,
the Bank began rehabilitation and technical assistance in other sectors,
such as agriculture, education, industry, mining, posts and telecommu-
nications, and water supply. It also convened a Consultative Group (CG)
meeting of donors and appointed a resident representative in July 1982.
The Reconstruction Credit was actually disbursed over three years
instead of the expected four months. There were several reasons for the
prolonged disbursement period. First, the Bank was overoptimistic in its
expectations about the reconstruction process, which proved to be far
lengthier and more complicated than initially anticipated. Second, po-
litical uncertainties also contributed to delays. Third, procurement and
disbursement procedures slowed progress considerably. Finally, although
the project was designed as an emergency operation, it evolved into a
policy loan, with attendant conditionality-related delays.
Despite the disbursement delays, the Reconstruction Credit did pro-
vide resources more quickly than traditional project lending. The first
and second Reconstruction Credits and the second SAL accounted for
almost all resource transfers in the first four years of the resumption of
Bank operations in Uganda. Moreover, early Bank involvement was
seen to have facilitated a fruitful country dialogue with respect to eco-
nomic policy and management issues.
Lending for Recovery, 1986–97
The Bank provided support to the Ugandan government’s Economic
Recovery Program (ERP) through a set of policy-based credits. The projects
were interrelated and supported the objectives
The entire portfolio from 1987 to 1997, of the reconstruction program. The First Eco-
which includes more than over 40
nomic Recovery Credit (ERC I), approved in
credits approaching US$2 billion,
focused on reconstruction and 1988, aimed to stabilize the economy and to
rehabilitation. This is not readily create a policy environment conducive to rapid
apparent from the project titles in the growth with an efficient use of resources from
portfolio, or from some of the project previous credits.
documentation. Only one credit was an The Second Economic Recovery Credit
Economic Recovery Loan, and many
(ERC II), approved over the three-year period
others made little reference to conflict
and its effects. 1990–92, supported the economic recovery pro-
gram through demand management—that is,
interest rate reform and fiscal measures designed to contain the budget
deficit. In addition, it included support for liberalization of trade policies
24
The Bank’s Role in Brief
and the revitalization of the private sector. Finally, it also aided the govern-
ment in reforming public sector management.
Subsequent Bank loans provided continued support for some elements
of the earlier ERC II project. Coffee sector reforms were continued under
the Agricultural Sector Adjustment Credit, approved in December 1990.
The first Structural Adjustment Credit (SAC), approved in 1992, also
made improvements in the trade regime, the Investment Code’s opera-
tional guidelines, and the Custodian’s Board’s operations. It supported
revenue mobilization by the public sector, as well as the reform of the
civil service. The Financial Sector Adjustment Credit picked up on themes
of interest rate reform and restructuring of the banking system. With
respect to the private sector, a Sugar Rehabilitation Project sought to
rehabilitate a sugar estate and to strengthen institutional capacity for
monitoring the performance of the industry.
Regarding the reconstruction program, the Bank also launched major
efforts to help rebuild basic infrastructure in Uganda through a variety of
projects, including Telecommunications II, Railways, Water Supply II,
and Power III.
Five technical assistance (TA) operations were approved in the period
after 1980: TA I, TA II, and TA III, the Economic and Financial Manage-
ment Project, and the Institutional Capacity Building Project. All of these
projects had as their central objective helping core economic ministries
and agencies design and implement a series of economic reforms. They
also strengthened the systems and capacities within the particular agen-
cies. Also on the public sector side, the Public Enterprises Project sought
to restore the country’s productive capacity while difficult reforms were
being implemented. This project overlapped with a follow-on Enterprise
Development Project.
Other Bank efforts in Uganda have focused on urgent social concerns
with economic or human capital consequences. These included the Alle-
viation of Poverty and Social Costs of Adjustment, which targeted pri-
mary education and vulnerable groups such as orphans, war widows,
and retrenched civil servants. The TA III operation encouraged Ugandans
living outside the country to return and help rebuild the nation. A number
of other projects, including the Northern Reconstruction Project, First
Health Project, and Education IV, helped to rebuild both social and physi-
cal infrastructure. Finally, the Urban I Project rehabilitated urban infra-
structure and institutions in Kampala, the nation’s capital.
25
Donors and the government appreciated the Bank’s role in aid coordination,
noting its close relationship with the IMF, its interest in partnership, and its
preparation of accurate planning data as key advantages. Coordination and
strategizing with respect to particular sectors or programs were less evident.
NGOs were less enthusiastic about the Bank’s coordination role, feeling that
the rhetoric of partnership and participation “ was moving faster than the
reality” and that the Bank was still not very transparent about its role. The
Bank improved this shortcoming through its well-used public information center in the Uganda
resident mission. However, several interlocutors, including some Bank staff, have called for a
more proactive information strategy addressing the Ugandan public.
The Bank’s Role in Aid Coordination
4
Since 1986, the Bank has organized nine CGs and has held regular
donor meetings, including the monthly donor meetings that take place
in the resident mission. Subgroups of the monthly meeting also work on
specific task areas. The Bank has also coordinated planning in recon-
struction activities through the preparation of Country Economic Memo-
randa (CEM), CASs, and other studies. The Bank’s coordination role has
undoubtedly been made easier by a shared view of policy imperatives
with the government and donor community.
The Bank’s Relationship with Donors, NGOs, and the
International Community
Donor Views
Respondents, especially the donors, were generally favorable toward
the Bank’s aid coordination activities. According to one donor, this role
was enhanced because the Bank was “here and positioned before many
of the bilateral donors.” The Bank’s ability to work closely with the
IMF, in particular, was recognized as a key advantage. These sentiments
were expressed by both government and donor sources. The Bank was
“very open and interested in partnership,” remarked one donor. Uganda
even set a precedent in coordination when, to provide an example of
best practice, it allowed Eritrea to attend the CG meeting in Paris.
Government Views
Government sources thought a useful aspect of the Bank’s coordination
role was the preparation of accurate planning data. For example, a popu-
lation census, including data on housing, schooling, and population growth,
27
Uganda: Post-Conflict Reconstruction
said one senior official, “gave a good start in the 1990s.” The coordinat-
ing value of Bank damage and needs assessment was proven by the
universal demand for preappraisal project documents and CEMs.
Beyond the macroeconomic framework, coordination with respect to
particular sectors or programs was not evident. This was contrary to the
popular view that the Bank’s comparative advantage was in its multi-
sector nature and potential ability to coordinate across sectors, as it did
when it engaged several donors in the demobilization and reintegration
program.
Guiding principles in sectors such as health and agriculture were also
lacking, according to donor sources. In addition, implementation suf-
fered from coordination problems. In the case of the Northern Uganda
Reconstruction Project, for example, confu-
Despite the Bank’s respected sion and delay were reported by the Bank
performance in aid coordination, it and cofinancing donors. Nationally, despite
was lacking an overall strategy for coordination structures, some replication of
reconstruction and prioritized sector
activities. As one Bank staff member
project initiatives was also noted, along with
explained, “ we didn’t know we were in “too much donor competition” in a “donor-
a reconstruction program until it was driven” environment. In sum, then, Bank
given that name later.” coordination did not extend to comprehen-
sive strategizing.
Bank–European Union Relations
One possible exception to the Bank’s effective relations with donors
was in its interactions with the European Union (EU). Here, a mutual lack
of detailed understanding was apparent There was little informal contact
outside of the monthly coordination meetings, although the two were
located only one staircase apart in the same office building. In light of
their proximity, one explanation given for their separation—that they
operate with different parts of the government (actually different depart-
ments of the finance ministry)—hardly seemed credible.
NGO Views
Not all of the Bank’s partners were quite so enthusiastic about the
Bank’s coordination role in Uganda, particularly some NGOs and U.N.
sources. Some felt the Bank’s leadership had focused the development
community on macroeconomic reforms at the expense of poverty alle-
viation programs. In addition, the Bank was criticized for employing a
“standard economic model approach” to coordination rather than adopt-
ing a specialized post-conflict approach.
28
The Bank’s Role in Aid Coordination
NGOs also felt that the rhetoric of partnership and participation “was
moving faster than the reality” and that the Bank was still secretive in
its role. This opinion was corroborated by widespread NGO views of
Bank activities. “Very few of our people really follow what the World
Bank is doing,” commented one Ugandan NGO, indicating the need for
greater dissemination of information on Bank activities. Moreover, the
U.N. system as a whole was difficult for these commentators to compre-
hend, such as when the United Nations Development Programme (UNDP)
also played a coordination role at times.
Information and Public Relations
Providing information about the Bank’s activities is, arguably, inte-
gral to its coordination role. Yet the level of understanding of the Bank’s
work in Uganda was generally low, even within the aid community.
This was partly a product of history. Traditionally, the Bank saw itself
in a confidential relationship with its perceived client, the government
of Uganda. But the ignorance was also a product of a more general
attitude within the aid community that “information was power” and
was to be shared only for the benefit of the sharing institution. Undoubt-
edly the Bank had done much to improve this shortcoming through its
well-used public information center at its resident mission. But this pas-
sive resource was not matched by any proactive information strategy
addressing the Ugandan public.
Even some Bank staff expressed the need for a concerted effort to take
the lead in making beneficiary groups more aware of their potential to
draw on Bank resources. Others felt communications and public rela-
tions skills might be an important consideration in selecting staff for the
resident mission. One major donor expressed the need for “more deliber-
ate exposure of what the Bank was doing.”
29
Despite good performance in reforming and rebuilding the economy, there
are several respects in which Bank involvement could have been
improved: insufficient attention to consensus building; excessive use of
conditionalities; and most important, a dysfunctional emphasis on
raising taxes. Nor did it fully utilize its coordination role to create an
overall strategy for reconstruction or a sector-by-sector plan. The role of
the Bank was key in assisting the rehabilitation of the sugar industry, in
removing the monopoly of the Coffee Board, and in rebuilding roads. But
the Bank did not always fulfill its potential comparative advantage, as in
the power sector, a key element in recovery.
5
The Bank’s Role in Stabilizing,
Reforming, and Rebuilding the Economy
The long period of social disturbance in Uganda profoundly changed
the endowments and structure of the economy, depleting it of human, finan-
cial, public, and social capital. Private agents retreated into informal and
subsistence activities. These processes are reversible only on a long-term
scale of perhaps two decades. In the interim, they impose constraints on
policy reform and service delivery. For Bank operations to be effective,
there is a need both to respect these constraints and to accelerate their relax-
ation. This section focuses on the effects of civil war on the Ugandan economy
and an evaluation of Bank lending operations in that context.
The Legacy of Civil War and Its Economic Implications
for Development Strategy
Although the typical image of civil war is one of physical destruction,
this is only one aspect of the real economic damage. In this
respect, a civil war cannot be compared to an earthquake that causes the
same amount of physical destruction. Indeed, a civil war alters both the
level and the structure of economic activity in ways that persist beyond
the war. To understand this, it helps to look at a few useful concepts.
Effects on the Different Types of Capital
Private Capital
The post-war period of a given country can be characterized by very
rapid growth or by continued rapid decline. The outcome depends on the
interplay of the credibility of the peace and the extent to which private
31
Uganda: Post-Conflict Reconstruction
agents have adjusted their portfolios during the war. Unless a peace
settlement is unusually robust, the rational expectation is that civil war
could likely reerupt. This presents a much riskier investment environ-
ment than during the prewar period.
What happens to capital stock during a war? If the capital stock could
be moved into and out of the country at will, it would drop abruptly and
catastrophically during the war and then partially recover after it. This
partial recovery would give rise to a “peace
The neglect of infrastructure investment dividend.” However, the capital stock can-
and maintenance so evident during civil not be moved into and out of the country at
wars can be thought of as part of a will. Even though regulations to prevent
portfolio strategy of capital flight. capital leaving the country can be largely
evaded, the major impediment to capital
flight during a civil war is that much of the capital stock is virtually
immobile once installed. A coffee plantation, for instance, cannot be
uprooted and replanted elsewhere; a building cannot be relocated. There-
fore, such investments are shifted out of the country gradually. The money
normally spent on maintenance and replacement is instead removed from
the country, so that the capital stock declines through depreciation.
Because the exodus of capital during a civil war is a gradual process, the
actual capital stock will be greater than that which investors would have
chosen were they free to remove capital at will. The war may end before or
after the capital stock has fully adjusted to its “preferred” wartime level.
During civil war, the growth rate of per capita GDP is typically
reduced by 2.2 percent per annum, implying that in most economies per
capita incomes will be falling.1 This annual loss of 2.2 percent consti-
tutes a leveling effect on economic activity—that is, as the war contin-
ues, the level of income gradually falls further and further behind what
it would have been during peacetime. The main process causing this
growth loss is the exodus of private capital. Other factors, such as eco-
nomic disruption and diversion, may also be at work.
Conflict Duration and Effect on Private Capital
If the war continues for a long time, to the extent that the capital stock
has been fully adjusted to the lower preferred wartime level, then peace,
even if fairly precarious, will produce a large peace dividend. The reason
for this is that the desired capital stock rises as a result of peace, and the
previous capital flight enables a rapid reallocation of portfolios to bring
the capital stock up to the newly desired level. After the typical 15-year
civil war, the growth rate of per capita GDP in the first five years of peace
is augmented by 5.9 percent each year above its underlying rate.
32
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
If, in contrast, the war is brief, then the capital stock is unlikely to have
adjusted very much. Still, capital stock may fall short of desirable levels in
the post-war period. Private agents will, of course, be pleased that the war is
over and may wish to hold a larger capital stock than during the war. But
because they did not have time to adjust their
actual capital stock, they continue to want to The focus of post-war policy should be
on restoring confidence in the
reduce it. In this case, despite the peace, capital
economy and in peace prospects, so
stock continues to decline. as to reverse the exodus of capital.
After a civil war lasting only one year, the Evidence suggests that by the early
economy continues to lose growth of 2.1 percent 1990s, around 70 percent of African
yearly during the first five years of peace, rela- private wealth was held outside the
tive to its underlying rate. Thus, the continent— a far higher proportion than
any other region.2 The portfolio choice
decline during the post-war period is indistin-
between domestic investment and
guishable from the decline during the war itself deposit abroad is indeed central to the
(2.2 percent). This is evidence that the major force African growth process.
causing a loss of growth during a civil war is the
exodus of private capital (which continues after a short war) rather than the
direct disruption and destruction brought about by the war.
If this focus on portfolio choices is correct, what are its implications?
First, paradoxically, the most difficult problems arise after short wars rather
than long ones. Even a half-hearted peace settlement after a very long war
is likely to be sufficient to increase the capital stock somewhat, so that the
early post-war years will be characterized by rapidly rising incomes. Con-
versely, after short civil wars, the likely scenario is continuing economic
decline, presumably making it more difficult to maintain the peace.
The Uganda Experience
How does the concept of the portfolio substitution apply to Uganda?
A key factor in the analysis above is the extent to which the capital stock
adjusted during the civil war. However, in Uganda, the duration of civil
conflict is not straightforward. By one definition, civil disturbance
began with Amin’s expulsion of the Asians and did not end until the
NRM government assumed power. By an alternative definition, there
was temporary peace from 1980 to 1982, followed by war from 1983 to
1986. By yet a further definition, peace was not restored until around
1989, when the north was largely secured.
Evidence from the behavior of private investment suggests that the
capital stock had not fully adjusted by 1986. Investment remained very
low until 1993. The growth rate certainly increased, but not to the spec-
tacular levels implied by the multicountry regression analysis cited above.
Even had the underlying growth rate for Ugandan per capita GDP been
33
Uganda: Post-Conflict Reconstruction
zero, the first 5 years after a 15-year disturbance would normally have
seen per capita growth rates of 5.9 percent. Uganda eventually reached
such growth rates, but only from 1994 onward. The period from 1986 to
1994 was therefore characterized by atypically slow growth for recov-
ery from prolonged civil war. Evidence indicates that the early years of
peace were atypically uncertain, delaying the return of private capital
(financial and human).
The initially low level of investor confidence was a key factor in the
post-war economy. By the end of the civil war in 1986, Uganda had the
worst risk rating of the 25 countries then rated by Institutional Investor
in Africa. A survey of potential and actual investors in Uganda con-
ducted in 1993–94 found that noncommercial risk was rated as the single
most important impediment to investment, well ahead of such factors as
infrastructure (World Bank 1994b). Within the area of noncommercial
risk, the two biggest fears were
Table 5.1: Private Investment in Uganda policy reversal and a resumption
Ush billion of civil conflict. Thus, the recov-
Year (1991 prices) ery in private investment was
1985 117.0 delayed until 1994 (table 5.1).
1986 111.4 Estimates of private capital
1987 137.1 inflows show a similar pattern.
1988 134.1 They were small and constant
1989 148.6
1990 172.7
before 1994–95, whereupon they
1991 193.4 almost doubled, growing further
1992 200.8 in 1995–96. These increases in
1993 228.6 private investment and private
1994 307.0 capital inflows coincided with a
1995 391.3
sharp improvement in the risk
Source: Background to the budget 1996– 97. rating for the economy, which rose
steadily from 5.2 (out of 100) in
March 1992 to 16.2 by October 1996. While this is a rapid recovery, it is
important to bear in mind that this remains a very poor risk rating, approxi-
mately on par with Nigeria. By comparison, Botswana and Mauritius have
ratings of around 50. While the time series are too short for econometric
analysis, the data are consistent with the hypothesis that the structural break
in the economy occurred not with the end of the civil war in 1986, but
around 1992, after which investor confidence rose and private investment
recovered.
The structural break around 1992 coincided with a major policy
change in the economy. As identified in Bank audit reports, the govern-
34
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
ment appeared to be ambivalent about the reform program before 1992.
This ambivalence had an impact on donor conditionality. During fiscal
1991–92, there were disagreements between the government and the
donors as to which of two exchange rates
should be applied to aid receipts. This caused While international wars tend to
serious delays in donor disbursements. The produce cohesion in the domestic
population, increasing trust and
government failed to adjust the budget in reducing opportunism, civil wars
time, resulting in a rapid spiral of inflation damage social capital and
to an annualized rate of 230 percent by introduce an expectation of
March 1992. This episode of inflation suffi- opportunistic behavior.
ciently scared the political leadership,
prompting a change in the economic management team and a new sense
of ownership of the reform program. The government’s actions led to an
upgrading of Uganda’s economy rating with investors. By 1994, Uganda
was among the most rapidly growing economies in the world. This can
be interpreted as partly a delayed peace dividend.
Human Capital
Although the discussion above has focused on private financial and
physical capital, a similar story could be told with respect to human
capital. From the early 1970s onward, there were very large losses of
educated Ugandans, a flow that began to reverse in the 1990s.
Social Capital
The effect on social capital—that is, social cohesion and trust among
the people—may be one of the major differences between civil and inter-
national wars.
For instance, it can pay for people to “cash in” on their accumulated
social capital, using it either to cushion consumption (the dissavings
effect) or for capital flight (the portfolio effect). If society switches from
a high-trust to a low-trust equilibrium, the cost of enforcing transactions
increases and the mode of transactions shifts from credit to cash, which
is a highly restrictive form of business.
If the society has moved to a low-trust equilibrium, the restoration of
social capital in the post-war era is problematic—that is, the low-trust
equilibrium might be locally stable. The rebuilding of standards in the
professions—doctors, lawyers, accountants, and civil servants—is likely
to be at the core of this process. Since these professions largely depend
on self-regulation, one strategy to rebuild trust is to internationalize
accreditation procedures.
35
Uganda: Post-Conflict Reconstruction
There were severe losses of social capital in Uganda. The legal system
deteriorated, increasing the difficulties of contract enforcement. For
example, banks considered Uganda to have acquired a culture of
nonrepayment. They could not make assets function as collateral because
they could not rely on court decisions. A survey of the banking system
conducted by the Bank of Uganda in 1995 found that the reform most
wanted was a fast-track legal procedure for loan recovery. Also, the con-
fused property rights following the expulsion of the Asians blighted pri-
vate investment and encouraged opportunistic legal claims.
There was a generalized collapse of standards in the professions. In
the legal profession, a 1995 survey found that only a quarter of lawyers
considered the judiciary fully independent of the executive (Widner 1997).
It also found that lawyers’ knowledge of the law was quite deficient. In
the civil service, the objective of service delivery had become secondary
to that of personal survival.
In the education sector, a survey of nonsalary primary education
releases from the central government conducted in 1996 found that only 36
percent of the money actually reached primary schools; the remainder was
retained by district-level administrations (Ablo and Reinikka, forthcoming).
In the health sector, a matching attempt to track expenditures broke
down because the basic financial information needed to trace expendi-
tures through to the health clinic level was not available. In the absence
of such information even for audit purposes, the use of resources in their
intended fashion appears to be voluntary. In the medical profession, there
has been widespread malpractice with respect to drugs and there is evi-
dence of inadequate knowledge. A study of prescription practices in
Mulago Hospital, the main hospital in the country, found an incidence
of misprescription of about 50 percent. Although sufficient drugs were
imported into the country for the population, the public health services
were chronically short of drugs because of their diversion to the private
practices of doctors working in the public system.
Public Capital
Public capital is not subject to the same process of portfolio substitu-
tion as private capital and, therefore, is likely to be less depleted by
social disturbance. There is also no equivalent to the repatriation pro-
cess. But donors do shift their portfolios, resulting in a suspension of aid
flows during civil war and a resumption after the war. These shifts are
somewhat analogous to private portfolio choices. Public capital is
depleted during war partly by destruction and partly by dissavings on
36
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
the part of the government. Typically the government will squeeze capi-
tal formation in order to fight the war. This creates an obvious need for
physical reconstruction in the post-war era.
Effects on the Structure of Economic Activity
The loss of private, human, public, and social capital during war
induces agents to reallocate their remaining resources between activi-
ties. The sectors most vulnerable to war are intensive in transactions
and capital. These include manufacturing; supply transactions services,
such as the distribution sector; or capital, such as the construction sector.
By contrast, the safest sector is subsistence agriculture, which does not
depend on transactions and uses few assets. Table 5.2 catagorizes activi-
ties on the schema above into war-vulnerable, war-invulnerable, and a
residual (unassigned activities) and shows how the shares of the three
catagories changed during and after the period of disturbance.
Table 5.2: The Composition of Ugandan GDP by War Vulnerability
Percent share of GDP at constant 1991 prices
1971 1986 1993-94
War-vulnerable 42.5 24.0 28.7
War-invulnerable 20.5 36.0 32.1
Unassigned activities 37.0 40.0 39.2
Source: Collier and Pradhan 1997.
There were large shifts between the war-vulnerable and war-invulner-
able sectors, whereas the share of unassigned activities stayed fairly con-
stant (suggesting that in aggregate these activities were neither
especially vulnerable nor safe). During the period of social disorder, subsis-
tence activities nearly doubled in share of GDP—an astonishing trend since,
counterfactually, there would have been a gradual decline in their share.
During the post-war period, the trend was reversed, and resources shifted
out of the subsistence sector into the war-vulnerable sectors. It is striking
however, how slowly this reversal occurred. After the first seven post-war
years, the structure of the economy still looked more like the immediate
post-war structure than the pre-war structure. Having grown by 15.5 per-
centage points of GDP during the war, the subsistence sector shed only 3.9
percentage points in the first seven years of peace. There was then some
acceleration, but even by 1995–96, subsistence was still 25 percent of GDP.
37
Uganda: Post-Conflict Reconstruction
What this means is that ten years after the end of the conflict, the economy
was still some way from restoring the pre-war structure. Again, this may
suggest that the period 1986–92 was characterized by atypically slow re-
construction for an economy that had suffered such a long period of distur-
bance. This, in turn, may reflect the intermittent
The Ugandan civil war reduced the nature of the disturbances, 1971–86, so that the
growth of the economy and changed its recovery resembled the slow pace associated with
very structure. a shorter period of war.
The loss of growth occurred primarily through
losses of private, social, and public capital. The economy lost approxi-
mately 2.2 percent of growth for 15 years and about 12 percent of GDP,
which was initially market activity that shifted into subsistence. Even ten
years after the end of the war, Uganda still had not fully made up these
losses, although it had probably recovered about half the lost ground. This
was not because Ugandan post-war policy has been particularly poor or
inconsistent. On the contrary, it has been considerably better than is likely
to be found in many post-war situations. Rather, it suggests that the relevant
time period for post-war recovery may be around two decades.
The Bank’s Recovery Lending Program
As of 1987, the Ugandan policy environment was very poor. A policy
reform program was needed, but the legacy of the prolonged social dis-
turbances constrained reforms in two ways. First, the deterioration in
the human and social capital of the public sector professions implied
that delivery of both reform and public services was going to be much
more difficult than elsewhere, and therefore less cost-effective. Second,
the retreat of the private sector into a kind of informalization and subsis-
tence implied that the revenue base for the government would be much
lower than elsewhere. Given the realistic timetable for post-war recon-
struction, such constraints last for perhaps two decades.
To accelerate development within this context, Bank lending needed
to focus not only directly on the objectives of policy reform and in-
creased service delivery but also on relaxing the long-term constraints
that would otherwise delay achievement of these objectives. This
implied that the critical path to reform and service delivery was the
enhancement of human capital and organizational efficiency in the pub-
lic sector. It also implied that the critical path to fiscal sustainability was
the reformalization and recommercialization of private activity through
the suspension of predatory taxation. The Bank’s Recovery Program lend-
ing portfolio can be evaluated on the basis of these considerations.
38
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
Three Bank Adjustment Lending Interventions
The Bank supported an abortive adjustment program in the Obote II
period. Although this was the first, briefly successful, stabilization pro-
gram in Africa, it fell apart in 1985 as a result of the spending pressures
brought on by the civil war and the limited commitment of the political
leadership. By 1986, few if any signs of this adjustment program remained.
The second episode of adjustment lending covered the period up to
March 1992. Government commitment was half-hearted, but neverthe-
less there was significant progress in some areas from 1989 to 1990. The
exchange rate was partially liberalized during this period, following a
major exercise in consensus building by the government. The single most
important mechanism of predatory taxation, the Coffee Marketing Board
monopoly, was removed. Asian property confiscated under Amin was
returned as a condition of SAL I. This restored the rule of law, gradually
removed the confusion over urban property rights that would otherwise
have become a binding constraint on investment, and provided a power-
ful signal to the international community that the government was seri-
ous in its commitment to recovery. In each of these reforms, the
government chose to incur substantial short-term political costs in the
hope of longer-term benefits.
The third episode, post-1992, is one of the most successful inter-
ventions of the Bank in development. The economy had grown rap-
idly, with private investment rising to levels at which growth is
sustainable. There has very likely been a substantial reduction in
poverty, as suggested by preliminary analysis of recent household
survey data. However, as the 1997 CAS stressed, the benefits of this
economic growth were not spread evenly across the country, with the
north in particular lagging behind.
Suggested Areas for Improvement in Bank Performance
Despite this good performance, there have been several respects in
which Bank involvement could have been improved. The three consid-
ered here are insufficient attention to consensus building, excessive use
of conditionalities, and, most important, a seriously dysfunctional
emphasis on raising taxation.
Consensus Building
Dialogue was largely confined to the Ministry of Finance and Plan-
ning. While there were advantages in this approach, as it was the most
competent ministry, it meant that other ministries failed to benefit from
the learning experience of working with Bank staff. Indeed, other
39
Uganda: Post-Conflict Reconstruction
ministries may have even felt excluded from the decision process and
increased their opposition. As a result, there was an overreliance on
presidential support for key reforms rather than engaging in the more
difficult task of building a more broadly based constituency.
Conditionalities
Although in the post-1992 period the government was committed to
the reform program, conditionality was used extensively. While this may
have forced through some policy changes, it had three disadvantages.
First, it tended to cast the role of the government in the eyes of the
outside world as one of niggling opposition to reform. Second, it may
have inclined reformers within the government to rely too much on Bank
and IMF pressure rather than on their own efforts to build a wider con-
stituency. Third, by spreading conditionality very widely it may have
weakened the government’s focus on the central reforms.
In private, the government itself sometimes invited conditionality as
a means of winning political arguments. However, it might have been
better had this process been made public and the purpose of the condi-
tionalities changed. Instead of using conditionality to accelerate the pace
of reforms, it might have been used to lock in the already completed
reforms. This might have done more to improve the country’s risk rat-
ing, given that the main deterrent to investment was fear of policy rever-
sal. Businesses were probably less impressed by promises that import
bans on beer would be removed within the next two years on pain of
donor penalties than had the government volunteered that it would main-
tain low inflation and a convertible currency to avoid such penalties.
There was arguably too much concentration on maximizing the pace of
reform across a wide front and insufficient recognition that the core task
was improving investor confidence.
Taxation
The changes that had taken place in the Ugandan economy as a result
of prolonged social disorder had both reduced tax revenue and eroded
the potential for taxation. The private sector had come to regard the
government as predatory, and so devised means of defending itself. Capital
flight, retreat into the untaxable subsistence sector, abandonment of
financial records in favor of cash-based transactions, and smuggling
were all well-established means of tax avoidance and evasion. Further,
the tax collection system depended on civil servants who had little
incentive to perform honestly, so payments by the private sector did not
necessarily reflect receipts of the public sector.
40
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
The Coffee Taxation Problem
This legacy of the war was compounded by two massive problems
with respect to the main tax base of the economy, both related to coffee.
Coffee had been almost the only activity that government had retained a
capacity to tax throughout the period of conflict. It had achieved this by
imposing a government monopoly on coffee marketing. As a result, the
government imposed astonishingly high true taxation (explicit export
taxes, plus the implicit taxes through overvaluation and inflated mar-
keting costs). This produced both evasion and avoidance. Smuggling
averaged around 27 percent of the crop. Resources were shifted from
coffee to subsistence crops and showed up in gradually declining levels
of production. This buildup of avoidance implied that the high taxation
strategy was unsustainable—it was snatching revenue now at the
expense of revenue in the future. The Bank, the IMF, and the government
recognized that tax rates on coffee would have to be reduced.
The second coffee-related problem was the 1989 crash in world coffee
prices following the breakup of the International Coffee Agreement. This
forced the government to remove coffee taxation. Hence, by 1990 the
government had lost what had been the major revenue source of its
predecessors. While the collapse of coffee taxation increased the need
for other revenue, it did not make it any easier to raise.
Both the Bank and the IMF attached considerable importance to a
rapid restoration of tax revenue and required that the share of revenue in
GDP increase by 1 percentage point each year. This condition, and the
consequent emphasis on “tax effort,” reflected a view of the post-war
recovery process that was insufficiently long term. It interpreted low tax
revenues as being the result of lack of political will and “aid depen-
dency” rather than constraints that were intrinsic to a post-war economy.
This may have been attributable to the presence of two timetables—one
for post-war recovery, and another for policy reform.
The timetable for policy reform could be accelerated somewhat by the
adept use of conditionalities, so that complete reform within a decade was
a feasible goal. However, the timetable for reconstruction depended not
solely on the pace of government policy changes but also on the pace of
private and social capital restoration. This, in turn, depended on the
actions of private agents and could take two decades.
The ostensible reason for raising the share of revenue was to reduce aid
dependence. The Bank, the IMF, and the government all acted on two
implicit assumptions: first, that the supply of aid would be greatly reduced
over a relatively short horizon, and second, that the efficient route to
41
Uganda: Post-Conflict Reconstruction
increasing tax revenue was a gradual, year-by-year increase in the share of
revenue in GDP. Neither of these assumptions was seriously scrutinized, yet
both were highly contentious.
The main future criteria for aid flows are likely to be economic need
and the policy environment. Aid is increasingly directed toward low-
income countries, since other countries can
Increased taxation became a central attract private capital flows. But, even in
part of the policy reform agenda, yet low-income countries, aid is ineffective
its detrimental effects upon the in poor policy environments.
reconstruction agenda were insuffi- Since Uganda is among the poorest
ciently appreciated. The end result
countries in the world, even good growth
was that the reform agenda was
accelerated by the very same performance for two decades will leave it
instrument that retarded the recon- eligible for large aid inflows on the
struction agenda. income criterion. It has now achieved
among the best policy environments of all
low-income countries. As long as there is no significant retrogression,
aid flows are likely to be substantial for the foreseeable future.
Low Taxation: The Better Option for Post-Conflict
Economies
Even were aid to decrease in the medium term, the fastest route to
self-reliance is unlikely to involve rapid increases in the tax burden.
There are two reasons why a phase of low taxation is appropriate. First,
the eventual tax revenue that can be generated from the economy
depends on the growth of the taxable base rather than actual tax
receipts. The rate of growth of the taxable base is reduced by current
taxation. Thus, there is a tradeoff between tax revenue now and tax
revenue in the future.
This is particularly true in a post-war economy because an important
part of rebuilding the taxable base involves the decisions of private agents
to reverse their previous tax-avoiding and evading strategies. Notably,
rural households must rejoin the market economy, and firms must switch
from the cash and informal enforcement economy to the financial records
and legal process economy. Taxation retards these conversions and so keeps
the taxable base narrow. Revenue targets can then be met only by imposing
highly distortionary tax rates on a narrow range of commodities.
The second reason why a period of low taxation is appropriate is that
if revenues rise in anticipation of a decrease in aid instead of in response
to such a decrease, the extra money raised is likely to augment govern-
ment expenditures. Since public expenditure is notoriously subject to a
42
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
ratchet effect, when aid is reduced the adjustment mechanism must still
come from increased taxation. The effect of the strategy of raising tax
revenue in anticipation of an aid reduction, then, is merely to lock the
economy into a higher level of public expenditure and taxation than it
would otherwise have chosen.
The effect of this tax-raising strategy was predictably that revenue
indeed rose in line with conditionality, but that this was not offset by any
aid reduction. On the contrary, as Ugandan policy and performance steadily
improved, donor flows were naturally augmented. The corollary was that
either the share of public expenditure in GDP would rise or the extra tax
revenue would be saved.
The share of public expenditure of GDP was arguably at a level
beyond the capacity of the government to deliver good value, as evi-
denced by the tracking survey of public expenditure noted above. Public
services were adequate, and there was a shortage of funds. However, the
real problem was weaknesses in the civil service and local government.
To the extent that the government restrained
public expenditure, its other options for us- An implication of the foregoing
ing tax receipts were the accumulation of discussion of the effects of civil war
reserves or lending to the banking system. is that a post-war economy has
Reserves were indeed rapidly accumu- atypically low marginal benefits of
public expenditure and atypically high
lated. By 1997, the Bank of Uganda held
marginal costs of taxation.
more than four months of import coverage.
Clearly, this was not a sustainable use of
higher tax receipts, and the remaining alternative—of lending to the
banking system—was begun. However, the Ugandan banking system in
its deregulated form is a very new and small creation, and its capacity
to on-lend money for investment with reasonable prospects of repayment
is likely to be very limited for several more years. Hence, the emphasis
on additional revenue-raising seems to have lacked a sound expendi-
ture-focused rationale.
The Five Negative Economic Costs of High Taxation
in Post-Conflict Economies
By contrast, the costs of the additional revenue have been all too evident.
Five costs in particular will be examined. First, a corollary of forcing an
increase in tax receipts in an economy in which there are few taxable activi-
ties or agents is that the major sources of additional revenue have come
from the application of very high tax rates to a few commodities. Most of
the extra revenue to date has come from the tax on petroleum. As a result,
43
Uganda: Post-Conflict Reconstruction
petroleum prices are now double the level of those in Kenya and Tanzania.
The structural transformation from civil war in Uganda is primarily about
reversing the massive retreat into the subsistence sector that occurred
between 1972 and 1986, so the high fuel tax is precisely pitched to retard
this restructuring. It is hard to believe that the marginal benefit of public
expenditure is high enough to offset this cost.
Second, the Uganda Revenue Authority (URA) finds it difficult to col-
lect taxes on corporate profits based on audited accounts because for
many companies audits are highly unreliable. The standards of the
accountancy profession deteriorated during the periods of social disor-
der. As a result, for many companies taxes are negotiated. Inevitably,
the information URA uses includes visible signs of profitability, notably
investment, so that what is ostensibly a profits tax has been, in part, a de
facto investment tax.
Third, the revenue from import taxes has increased considerably
because the effective rates of duty have risen as a result of devaluation.
When the official rate was highly overvalued, the effective import duty
was only a small fraction of the apparent rate, since duty was charged
only on the underestimated value, calculated by applying the official
exchange rate. While export taxes were removed as part of the liberal-
ization program, they were replaced by im-
The switch from export taxes to import port duties for revenue purposes. This has
taxes has defused the lobby for free proved highly dysfunctional.
trade (exporters), while enhancing the Although import duties are shifted through
power of the lobby for protection. The
general equilibrium effects to exporters, and
chance to remove export taxes while
leaving import taxes at their initial so are equivalent to an export tax (the Lerner
low levels was missed because of the Equivalence Theorem), this incidence is
excessive concern with raising highly nontransparent, unlike the export
revenue.3 taxes that they replaced. Conversely, the
protection afforded to particular import sub-
stitute producers is sufficiently transparent to induce strong lobbying.
Fourth, the coffee export tax, which had been removed in 1992, was
reintroduced in 1994 during the coffee boom. The IMF’s rationale for
this action was stabilization. However, this turned out to be unneces-
sary, as only a small amount of revenue was raised. The incidence of
the tax probably fell very heavily on private investment, since the
investment rate out of the coffee windfall appears to have been more
than 50 percent. The coffee tax was reduced in the 1995 budget and
removed in the 1996 budget (against the IMF’s advice). However, the
potentially negative signal to coffee growers may have significantly
44
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
reduced investment in the sector. Further, because of the considerable
idle coffee capacity as a result of the rundown during the civil war,
there was an unusually high supply response to the increase in the
coffee price. Hence, unlike in an economy without a recent history of
conflict, there was a high distortionary cost to the tax in the form of
forgone exports. The IMF failed to consider these structural and be-
havioral aspects of the tax, focusing upon a stabilization rationale
that, as it transpired, was unnecessary.
Finally, the value added tax introduced in July 1996 was set at over-
ambitious levels of collection rather than being phased in gradually to
build up compliance. The tax met massive
noncompliance. There was a strike by trad- Both the Bank and the IMF were
ers, and a large majority of the 12,000 firms continuous sources of pressure,
liable for the tax failed to register. In response including explicit conditionalities, for
to the strike, the turnover threshold was increased revenue. One interpretation
raised from Ush20m to Ush150m, but then of this is that both agencies failed to
take sufficiently into account the
lowered back to Ush50m because of pressure
distinctive characteristics and
for revenue. Even at this new level, nine constraints of a post-war recovery,
months after its introduction only around 30 focusing instead on the timetable that
percent of the firms still liable had registered. would have been more appropriate for
In each of these five cases, the thrust to- policy reform in an economy without
such an inheritance.
ward higher revenue collection came at con-
siderable cost. No attempt was made to
compare these costs against the marginal benefits of public expenditure.
As a result, the timetable by which the economy will eventually raise
enough revenue for self-sufficiency may have been retarded.
The Bank’s Role in Rebuilding Infrastructure and
Economic Sectors
Projects for 1988–89: Health, Sugar, Technical Assistance,
Public Enterprises, Telecommunications, Railways, and
Education
Following the ERC negotiated in 1987, the Bank launched four projects
in 1988. These were for health, sugar, TA, and public enterprises and
were followed in 1989 by projects for telecommunications, railways,
and education.
Of these seven lending operations, the two that have probably had
the most impact on the reconstruction process were TA and sugar. The
two projects together cost only US$43 million but provided very high
payoffs.
45
Uganda: Post-Conflict Reconstruction
The TA project built a competent team of economists within the core
ministries and put data collection systems in place. It gradually devel-
oped a team of technocrats with enough knowledge and understanding
of short-term changes in the economy to manage policy reforms while
maintaining macroeconomic stability. The sugar project was important
because it enabled the rehabilitation of
Two of the projects most appropriate Uganda’s foremost private business concern,
for Uganda’s post-war transition were the Madhvani Group. The return of
those that helped rebuild human
Madhvani to Uganda pioneered the influx
capital in the civil service and
restored private business confidence. of Asian skills and capital, which are now
helping to transform the economy.
Infrastructure Projects
Two infrastructure projects for railways and telecommunications each
had a component of organizational improvement, as well as the pur-
chase of equipment. From the framework developed above, the empha-
sis of initial public sector projects should have been on organizational
redesign and human capital enhancement. For example, the railways
had a monopoly in the transportation of coffee. Not only was this
directly supportive of inefficiency in the railways but it also had serious
implications for the efficiency of road haulage. Once the monopoly was
removed in 1993, the scale of business in road haulage increased suffi-
ciently to induce entry to the activity, breaking a cartel that had kept
costs high (Collier 1997). The public enterprises project appropriately
focused on institutional reform in the public sector.
Social Projects
The two social projects, health and education, may have been prema-
ture. These are considered at greater length in the following chapter.
Public Enterprise Projects
The Public Enterprise Project (PEP) emphasized reorganization. Three
years before it was completed, a second public enterprise was started
that emphasized retrenchment. OED criticized these projects for having
somewhat conflicting goals. By the mid-1990s, public enterprise deficits
had grown to be a major claim on the budget and the government was
expressing alarm. The extent to which public enterprises were radically
dysfunctional as a result of the collapse of public sector standards of
behavior was probably insufficiently appreciated.
46
The Bank’s Role in Stabilizing, Reforming, and Rebuilding the Economy
Projects for 1990: Social Costs of Adjustment, Water
Supply, Agricultural Adjustment, and Livestock
The Program for Alleviation of Poverty and Social Costs of Adjust-
ments (PAPSCA) project was arguably politically necessary. The reform
program was under considerable political attack from various vested
interests, and the needs of “vulnerable groups,” including women, needed
to be accommodated. Other than this it is doubtful whether PAPSCA was
a priority. At that particular stage in Uganda’s development, the key
poverty reduction process was the reintegration of the rural economy
into the market. The delivery of public services to the poor depended on
the prior reform of the delivery system.
The water project followed the pattern of the earlier projects with
some element of institutional reform, but reform could have been a higher
priority. The agricultural project focused on institutional reform in the
coffee sector. It enabled the dismantling of the monopoly of the Coffee
Marketing Board, arguably the single most important reform in the
entire decade of policy change. Since control of coffee had traditionally
been the most important instrument of public sector predatory action,
the liberalization of marketing was a major political change.
The livestock project involved both institutional reform in the Ministry
of Agriculture and an attempt to revive extension services. Livestock is an
appropriate focus in a post-war recovery since it facilitates domestic asset
reaccumulation. Agriculture production provided the vehicle for the many
series of discrete devaluations that were effected before the floating of the
shilling in 1992. Before 1992, the shilling was being devalued to provide
room for an increase in the producer prices for exports, following recom-
mendations by the Agricultural Secretariat based on its study of the impact
of inflation on the production cost of main exports.
Projects for 1991: Urban Infrastructure, Power,
and Enterprise Development
The urban infrastructure project was timely, because by 1991 Uganda
was starting to reurbanize and this process subsequently accelerated.
Reurbanization was functional, reflecting the recommercialization of
the economy. By 1997, urban infrastructure had barely kept pace with
increased needs.
The power project was arguably too late and insufficiently focused
on institutional reform. It was also beset by severe implementation
difficulties. And, by 1997, there were frequent power cuts. As a geo-
47
Uganda: Post-Conflict Reconstruction
graphically landlocked country, Uganda did not have recourse to
mobile power stations, and so a more cautious approach should have
been taken toward power provision. Since there are large leaks from
the system, there was also scope for ameliorating the supply problem
with institutional reorganization.
The enterprise development loan included elements that rapidly
became dated. Specifically, the provision of a foreign exchange Fund for
Technology and Management Contracts became unnecessary with the
liberalization of the foreign exchange market as the project started.
However, its main purpose was to rehabilitate public enterprises with a
view to privatization.
Project for 1992: Northern Uganda Reconstruction
Widely regarded as too little too late, the Northern Uganda Recon-
struction Project (NURP) is discussed in detail in Chapter 7. While the
current serious fighting in the north would have occurred regardless of
development efforts, there is no doubt that
Overall, there was probably insuffi- the north started as the most disaffected
cient attention to capacity building region and has been conscious of being left
and incentive design in the ministries further behind. Opposition candidates
and public enterprises that were vital received a majority of votes in the north dur-
for service delivery. Such efforts could
ing the 1996 election. The north is now prob-
have been started across the board
during a window of opportunity in the ably too dangerous for large-scale projects,
late 1980s. whereas from 1989 until 1996 there was a
window of opportunity.
48
Bank performance was relatively poor in the social sectors, particularly in
strengthening health and education institutions. Investment in the social
sectors was often premature, implemented through weak ministries and with
inadequate forethought as to the nature of the services the Bank was trying
to support. The post-conflict period required major health sector reforms that
fell short of needs. Education investment was equally disappointing, with
only an estimated 37 percent of funds reaching schools. There was too much
dependence on the existing bureaucracy and not enough use made of NGOs.
6
The Bank’s Role in Rebuilding Human
and Social Capital
In the post-1986 reconstruction era, rebuilding institutional capacity
was focused on the economic rather than human development sectors.
Early policies and projects were not directly targeted at vulnerable groups
such as the rural poor or households headed by women. Despite a low
urbanization rate of about 15 percent, Bank reach into rural areas was
limited. Northern districts, like rural areas generally, were neglected.
Even when NURP was designed to redress this balance, implementation
was less successful than in many other rural areas.
Human Development
The two social projects in health and education may have been prema-
ture. Both were attempting to deliver services through very weak minis-
tries, perhaps with insufficient attention to improvements in the incentives
for service delivery. Given the collapse of civil service standards, the key
reform in social services probably served to give the frontline service pro-
viders—schools and clinics, budgets, and their staff, incentives linked to
monitorable performance measures. Without this, attempting to push money
through the ministries has been demonstrably ineffective. Overall, the
concern with “projectizing” social sector reconstruction has overlooked
the need and opportunity to reappraise and redirect policy along lines
relevant to the 1990s.
When attempted, social sector results on the ground were “modest to
negligible,” as revealed in the Bank-funded tracking study of 1996 (Ablo
and Reinikka, forthcoming). PAPSCA was the first Bank project in Uganda
to address vulnerable groups, including women, but project conception
and performance were generally rated as poor.
49
Uganda: Post-Conflict Reconstruction
Health
The Bank’s first health project was intended to rehabilitate a select
number of hospitals and health centers, promote preventive health pro-
grams, and ensure the long-term sustainability of health care delivery
systems. The long period of conflict had resulted in the nearly total col-
lapse of the health sector, and more recently human immunodeficiency
virus (HIV) spread quickly, and acquired immunodeficiency syndrome
(AIDS) became pandemic in the country. Despite these needs, early health
spending suffered from a lack of investment in institutional capacity, and,
according to one NGO, concentration on rehabilitating hospitals, which
“was badly placed investment pandering to people’s notion of returning
the health system to its former glory.” Despite continuing donor invest-
ment, independent research has exposed the continuing absence of any
substantial public health service. It has largely become an “empty ves-
sel,” a vehicle for the private business activity of health workers selling
services and drugs on the side.
Nevertheless, there has been some overall improvement in health sta-
tus in recent years as indicated by the Demographic and Health Survey
of 1995. Compared to the situation in 1991, infant mortality fell from
122 to 97 deaths per thousand live births.
Universal comments on the weakness But this may be attributed to a combination
of the health and education ministries of several factors, including improvement
reflected a lack of prior institutional in economic conditions, housing, access to
support to manage ambitious
programs. Some sources suggested
safe water, and sanitation services. More-
funds were swallowed up by over, despite the positive trend, the health of
administrative costs and did not Ugandans is still rated poor compared with
improve capacity to deliver services. other countries in East Africa. For example,
in Kenya—frequently used as a comparative
yardstick by Ugandan professionals—infant mortality is now estimated
at 67 per thousand births.
Education
Education assistance, which was pursued through PAPSCA, NURP,
and specific education projects (Education III, IV, and V), achieved equally
modest results on the ground. These projects focused on concrete goals:
exercise and textbooks, building materials and construction of class-
rooms and staff rooms, and some training of teachers.
Incremental gains were not apparent; rather, there was more of a
“stop and go” approach through the various credits. Sources confirmed
that “there was not enough pre-study, often nothing on the ground to
50
The Bank’s Role in Rebuilding Human and Social Capital
rebuild,” and that a standardized approach was employed, dictating
rather than reflecting community needs. Under PAPSCA, for example,
government education sources criticized the selection of 4,200 classrooms
in 10 poor districts for renovation as “creating local apathy and depen-
dency.” Moreover, the government and Bank focus on primary educa-
tion, which drew in all the donors, “was akin to throwing a stone into
the bush to see what happens.”
Educational attainment remains low. Less than 50 percent of adult
women are literate, and overall literacy levels are about 61 percent. The
gross enrollment rate is about 68 percent (a
lower rate than in 1986) and may be artifi- There appears to have been little
cially inflated by the large numbers currently overall strategy guiding recovery in
enrolled in Primary One following the the education sector.
introduction of universal public education
(UPE). There are serious problems of wastage, with about 70 percent of
children failing to complete the primary cycle. Only three-quarters of
the candidates pass the primary leaving examination, and few students
progress to secondary education.
Although government spending on education has shown real increases
(18 percent for the period 1992–93 to 1995–96), it is still quite modest,
with primary education receiving about 1 percent of the gross national
product (GNP), compared, for example, with Ghana, at 6 or 7 percent.
There is a gross insufficiency in trained teachers throughout the country,
a problem exacerbated by the introduction of UPE, compelling the gov-
ernment to recruit untrained teachers to deal with the shortage.
At schools visited during this study, the assessment team observed
that the government declaration of UPE with compulsory enrollment
had resulted in overcrowded classrooms. Ambitious building plans, with
community involvement as a precondition, were under way, but without
reliable data on potential school populations. Local sources suggested
that finished schools were as much the result of local politicians needing
to show product as a planned response to need. Lack of trained teachers
overshadowed the effectiveness and sustainability of these initiatives.
Lack of food, clean water, and latrines for children was an additional
serious problem, especially where drought and food shortage had
reduced families’ ability to provide their own supply. Private schools,
often established by NGO projects, suffered similar problems and added
to the impression of a random building program.
NGOs felt strongly that more Bank resources (and any dividend from
debt relief initiatives) should go into the social sectors. But an effective
51
Uganda: Post-Conflict Reconstruction
implementation plan would clearly be desirable before investing more
resources in this direction. One solution proposed would involve more
active participation of NGOs in planning and supervising implementa-
tion in the districts.
Governance, Transparency, and Civil Service Reform
Respondents generally felt that the Bank was effective in helping to
rebuild key public sector institutional capacity, especially in the central
economic ministries. As the expulsion of the Asians under the Amin
regime had depleted commercial expertise, so years of conflict had driven
educated Ugandans away, removing the base for an effective civil service.
The attempted reconstruction under Obote II had failed because, said one
observer, “the center wasn’t strong enough.” By the time of the NRA
succession to power, he added, “the entire machinery of government had
collapsed.” With no experience in government,
Two governance issues tackled under it was conceded the NRA was bound to make
reconstruction were the restoration of mistakes.
the legal system and the return of
property to the expelled Asian
With the government described as initially
Ugandans. The Bank’s attachment of confused about policy orientation, the Bank had
conditionality to the latter objective to temper its approach. However, strong gov-
was widely recognized as critical. ernment ownership, particularly after 1992, was
key to the reconstruction program, and the gov-
ernment was reported to have given “good access on governance issues.”
At the center, the Ministry of Finance and Economic Planning and the
Bank of Uganda were described as the two most effective instruments of
governance, largely because the Bank and IMF put a lot of money into
them. By contrast, the sectoral and local government ministries did not
have the same investment or capacity.
Transparency in governance, particularly in relation to the work of
the Finance Ministry, was described as high. According to Transparency
International, Uganda is the second most open country in Africa. But
poor governance, according to a number of sources, was still widespread
in the form of corrupt practices.
Civil service reform was central to restoring governance. Bank efforts
to reform the civil service (through TA I, TA II, and TA III and ERC I and
ERC II), supplemented by other donors (notably Britain’s Overseas
Development Association, or ODA), initially had a significant impact in
the reduction in size of the civil service as well as pay levels. Recently
these efforts appear to have lost momentum.
52
The Bank’s Role in Rebuilding Human and Social Capital
Decentralization, incorporating political, administrative, and finan-
cial powers, was started in 1992 and “completed” in 1996 with the
creation of 35 districts. At the same time, delay in decentralization has
created an absence of resources for governance measures at the local
level. Districts visited by the assessment team complained of resources
held or still controlled at a central level. The tracking study of social
sector spending clearly records the lack of fund transfers to district health
and education authorities. Where funds had been transferred, severe
accountability problems have been reported.
Participation and Civil Society
Reports suggested that in the earlier years of the reconstruction pro-
gram, the Bank had not worked in a very participatory manner to in-
volve beneficiaries or various sections of civil society. Some Ugandan
NGOs reported virtual ignorance of the Bank’s activities, and interna-
tional sources stressed the need for the Bank to conduct extensive consul-
tation with beneficiaries in project planning.
At the time of the OED mission, the resident mission did not appear to
have a proactive information strategy—utilizing local media, for example.
There were many indications that the Bank was intent on stepping up par-
ticipation with the preparation of the 1997 CAS. But recent attempts to
involve NGOs in the preparation of a nutrition and early child develop-
ment project had been a “fiasco” according to the NGOs. Their views were
solicited, and then ignored by a series of expert missions from Washington.
53
NURP aimed to redress the imbalance in recovery assistance available to
northern districts. It has been criticized for achieving few results on the
ground and generally characterized as “ too little, too late.” Its successor will
need to address the complexities of operating, at least in part, in a conflict
zone, with the likelihood of parallel relief and rehabilitation activities. De-
mobilization in Uganda was initially delayed for the wrong reasons
(microsecurity), and then begun at an inappropriate time considering
macrosecurity factors. While the Bank might have counseled greater caution, the decision was
political and remains with the government. Regarding process, demobilization was clearly a
success. Evaluation reports of operations in Uganda reveal several lessons on operating in a
post-conflict context, including the importance of establishing a dialogue, preserving flexibility,
avoiding unrealistic expectations, considering structural constraints, keeping procurement and
disbursement procedures consistent with the post-conflict context, and ensuring coordination
consistency in the Bank’s strategy.
Special Issues
7
The Bank’s Role in Ongoing Conflict: Northern Uganda
Reconstruction Project
The 1992–1997 NURP was a regionally targeted, multisectoral project
covering ten northern projects. Still undergoing implementation at the
time of the post-conflict evaluation, it was intended to redress the imbal-
ance in recovery assistance available to the northern districts. Recog-
nized by the government as “disadvantaged from the beginning,” these
districts suffered a double penalty. They were rural (areas generally
neglected by early assistance) and they were associated with the losing
side of the conflict.
Early Implementation Problems
Government sources were critical of the “two years lost” preparing
NURP. Such projects should head off mistrust and conflict “when the
clouds are forming,” not during the downpour. Like PAPSCA before it,
NURP was multisectoral, comprising seven different sectors. Implemen-
tation therefore required intensive management. This was an ambitious
objective for the missions, possibly indicating the need for in-country
management. The use of NGOs for implementation was largely restricted
to the water sector.
NURP was plagued by implementation problems from the outset. When
the second task manager took over, only US$4 million had been dis-
bursed, and the management structure had to be simplified to improve
results on the ground. Clearly, management by missions was more of a
55
Uganda: Post-Conflict Reconstruction
problem for a project dealing with some of the most remote and poorest
parts of Uganda. Moreover, NURP had to deal with the shift to decen-
tralization after its inception. Renewed violence in some areas served by
NURP further complicated and impeded progress.
Project Shortcomings
Generally, criticism of NURP has related to the lack of tangible progress
on the ground. Procurement procedures were universally recognized as a
“nightmare” and “very frustrating.” The use
of international competitive bidding (ICB)
Despite claims by the government that seemed particularly inappropriate, consid-
the NURP has done much to soothe ering that provisions for streamlined proce-
hatred in areas of conflict, a sense of
dures were available had the Operational
grievance persists. Some commenta-
tors suggested that earlier and greater Directives (OD) for emergency lending
input in the north by both the (through OD 8.50) been used.
government and the Bank might have District authorities and sources were very
helped to reduce conflict, or at least to critical of delays in implementation, espe-
calm a disaffected population.
cially at the outset. Although concern was
focused on current drought conditions,
people in Pallisa, for example, recognized that their vulnerability was
a long-standing result of a conflict that subsequent assistance had not
removed. NURP had not addressed local concerns such as feeder roads,
water supplies, or cassava crops. There were also complaints about
lack of information about project requirements for local inputs.
During the evaluation, a Parliamentary Committee on the north
discussed NURP. Some Parliamentarians stated that it had failed in its
objectives and should be replaced by a new initiative, addressing three
separate regional groupings of districts in the north with distinct needs.
They concluded that NURP had failed for three reasons: (a) internal
corruption, made easier by a top-down approach whereby the local
communities cannot participate; (b) the escalation of conflict, halting
infrastructure work; and (c) the small loan scheme, which failed be-
cause many families were not suitable targets for soft loans.
Tendering methods had received widespread criticism, including press
coverage of contract awards to “known incompetent firms.” NGOs should
have been used more, the report claimed, and WorldVision was praised
for its results in building schools (although it was admitted that these
had remained “skeletons.”) Overall, the report concluded that the inten-
tions of NURP were good. But it was too much of an omnibus program,
with too much infighting over too little money.
56
Special Issues
Project Successes
One success recorded by the Parliamentarians was the rapid basic
teacher training program, the Northern Integrated Teacher Education
Program (NITEP). However, this had been too expensive to replicate on
a wider scale. Highway construction between Lira and Soroti was also
impressive, although less so on the main highway north to Gulu.
NGO and Donor Views
NGOs were among the most vociferous critics of NURP. “NURP was
a great idea,” said one established NGO, “but the implementation was
shoddy . . . with corruption and slowness, it was a real tragedy in many
ways.” The use of NGOs in implementation was largely restricted to the
water sector.
Donors gave NURP a more balanced verdict of “successes and fail-
ures.” One serious shortcoming noted by a major donor was the location
of NURP in the Prime Minister’s Office, which had limited implementa-
tion capacity. Decisionmaking was also distorted by patronage and po-
litical concerns. Despite some good achievements, the resulting value
for the money invested was inadequate. Described by one donor as a
“bit of a lost soul,” ownership of NURP was also questioned by a num-
ber of sources.
Coordination of activities under NURP became increasingly prob-
lematic for the Bank, especially as the growing conflict was drawing in
more donors and aid agencies provided emergency relief in the region.
Existing Bank staff working on NURP were concerned about donors
“tripping over one another” but did not have the capacity to address this
supervening demand.
Various sources discussed the possibility that through its failings, NURP
may have contributed to the escalation of the conflict. “Continuing pov-
erty drove some people into the hands of the rebels,” argued one Parlia-
mentarian, adding that “thousands of soldiers up there who [formerly]
depended on the State were sent back home with no provision.” The lack
of provision in NURP for demobilized soldiers was strongly criticized by
those closely involved in administering the process. This was a criticism
leveled at the donor community as a whole, not just the Bank.
Government Views
Government sources indicated that the lessons learned from NURP
included the need for fast-disbursing programs. A more detailed assess-
ment of the conflict and its causes, with a design intended to dissuade
people from fighting, was also required. Finally, the Bank needed to
57
Uganda: Post-Conflict Reconstruction
demonstrate greater flexibility in adjusting its targets and criteria should
component objectives prove elusive.
Bank Views
Bank sources judged implementation to be generally successful. How-
ever, it was designed before decentralization—without the features of a
demand-driven project or participatory
The successor project will have to operation—which proved to be a major
address the twin problems of poverty shortcoming. Without major redesign, a sub-
and conflict. The project will also project component to facilitate implementa-
need to deal with the complexities of tion was added post-decentralization. The
operating, at least partly, in a conflict other main problems with respect to imple-
zone, with the likelihood of parallel
relief and rehabilitation activities.
mentation were (a) inadequate and untimely
release of counterpart funds; (b) renewed
conflict that affected progress in primary
school rehabilitation, teacher training, and borehole drilling, particu-
larly in Gulu and Kitgum; (c) late submission of accounts and documen-
tation (which remains an issue in several districts), despite improvement
in financial management; and (d) procurement problems.
An audit would be useful to comment on the diversity and disparity of
views concerning NURP, particularly as preparation of a successor to
NURP is under way.
In addition, it will have to adjust to the government response to the
conflict, currently a focus of debate. Some of the government’s more
strident “vocabulary of war” was criticized by a more conciliatory “con-
flict resolution” school. If the latter approach were adopted, the project
would have a very big part to play. If the former, the project would find
itself in a difficult policy context, where total victory (unlikely accord-
ing to informed sources) might be a precondition for full-scale imple-
mentation. Some increased use of NGOs for implementation of the project
is indicated, although this may be limited by the capacity of local NGOs.
Security, Demobilization, and Reintegration
Despite continuing insecurity following the NRA victory in 1986,
Bank staff thought that the government had quickly established the pre-
conditions for reconstruction activities by promoting an attitude of rec-
onciliation and by not using their victory for retaliation. Soldiers were
well disciplined, the courts were not used to oppress opponents; instead,
the government tried to draw the opposition into government. Staff
reported “normal competition and jealousy, but no bitterness or cru-
58
Special Issues
elty.” Above all the government was able to maintain security, in con-
trast to previous changes of government that were marked by disorder
and looting. Looking back over these past ten years, even fierce political
opponents of the government were able to recognize the stability brought
by a decade of NRM domination.
The Ugandan demobilization was delayed until late 1992, partly
because of a continuing need for an army to deter rebels and partly
because of fears that soldiers would be more of a problem once demo-
bilized than while on the payroll. Specifically, demobilized soldiers
might become disaffected and join new rebel groups or resort to crime.
These potential risks of demobilization affected state, societal, and
individual security.
The effect of the demobilization on state security is difficult to
assess. Subsequent to the demobilization, problems of state security
have considerably increased. Within Uganda there are currently two
serious armed conflicts, as well as two major military conflicts in neigh-
boring countries. The extent to which demobilized soldiers are impli-
cated in these conflicts is unknown.
Demobilization is perhaps less likely to have reduced the army’s
deterrent effect. In the year following demobilization (1993–94), the share
of defense expenditures in the budget fell from 18.8 to 18.1 percent, but,
it increased to 19.9 percent in 1995–96. In effect, the army became con-
siderably more capital-intensive and consequently probably better
equipped to contain guerrilla movements. In this sense, demobilization
was more analogous to civil service retrenchments than to demilitariza-
tion. However, it has been suggested by those concerned with defense
capabilities in northern Uganda that demobilization was inadequately
synchronized with the reequiping of the army, giving rise to a time of
weakened capacity to suppress rebellion.
The Bank’s Role
Clearly, the Bank is not well positioned to make the political and mili-
tary calculations and judgments necessary for decisions about state secu-
rity. Programs that may have an impact on state security are not among
the Bank’s comparative advantages. Also, the government may have views
and information that override societal security considerations.
The Bank can bring some competence to the issue of societal and
individual security, since the effect of demobilization is amenable to
quantification. In Uganda, soldiers to be demobilized were first included
in a census that recorded their socioeconomic characteristics, and then
59
Uganda: Post-Conflict Reconstruction
transported back to their home districts. They were required to remain
in their home districts in order to collect subsequent installments of
demobilization payments. A follow-up sample survey of the demobilized
found that only around 2 percent had chosen to leave the district to
which they had been transported. The incidence of demobilization var-
ied considerably among districts, so that by combining this information
with time series on district-level crime, it is possible to measure the
effect of demobilization on crime.
Options for Demobilized Soldiers
Much of the recruitment into the army had occurred while the NRA
was itself a guerrilla movement, attracting teenagers with little or no
education. The census found that the soldiers to be demobilized were
largely unskilled and uneducated. This precluded options that involved
transferring the demobilized into wage employment. For example, the
police force required a far higher standard of education than was fea-
sible for demobilized soldiers.
The only realistic option was to reintegrate the demobilized into the kin-
group support systems and smallholder farming activities from which they
had been drawn. The ability of the demobilized to reenter the rural economy
thus depended upon their rights within their kin-group and whether they
had access to land. This question was asked as part of the census of soldiers
to be demobilized. Again, responses differed considerably among districts.
A large majority, 88 percent, claimed to have access to land. The remain-
ing 12 percent without access came disproportionately from districts with
high population densities and consequent land scarcity.
Before, during, and after demobilization, the expatriate community
was extremely concerned that demobilization would worsen crime. This
was reinforced by instances in which demobilized soldiers were thought
to have been implicated in incidents. However, an econometric analysis
revealed a rather different picture (Azam and others 1994).
In the first three months following the demobilization, the process
had two significant effects on crime. The demobilization of soldiers with-
out access to land significantly increased crime. About 6 percent of
demobilized soldiers without access to land committed a recorded crime
within three months of demobilization (or 3 percent committed two
recorded crimes). However, the demobilization of those with access to
land significantly reduced crime. On average, each soldier with access
to land reduced crime by just as much as each soldier without land
increased it. Although this may seem unlikely, it was supported by evi-
60
Special Issues
dence from interviews with village leaders: they judged that the pres-
ence of men with military training had discouraged crime by others.
Because most soldiers had access to land, the net effect of these two
opposing influences was for crime to decrease. Each of these effects held
only in the first three months following the demobilization.
In the following nine months, neither the numbers of demobilized sol-
diers without access to land nor those with access to land had any discern-
ible impact on district crime rates. The implication is that the demobilized,
even the landless, were quickly integrated into their local communities and
became indistinguishable from the rest of the population.
The Demobilization Project
These facts show that three mistakes were made with respect to the
demobilization project. First, demobilization was delayed because of
unnecessary fears about the difficulties of reintegrating the demobilized
into their communities. Schemes for special training and attempts to
establish soldiers in small-scale enterprises ran the danger of creating
unrealistic expectations, while exaggerating the difficulties of reentry to
agriculture. Second, the reasonably obvious problem of lack of access to
land was not addressed. Either access to land should have been negoti-
ated for those without it before demobilization or those without access
could have been retained in the army. Finally, the demobilization scheme
included a package of both equipment and money for the demobilized.
The equipment part of the package, such as roofing sheets, arrived very
late and posed a logistical problem. The choice of equipment over money,
although not ideal, was dictated by the resources available, especially
from the U.S. Agency for International Development (USAID), which
could provide materials but not money. Project planners assumed this
would be monetized.4 The installment nature of the payments was prob-
ably a good arrangement, helping to anchor people in their districts and
giving them a temporary lifeline.
The Veterans’ Assistance Board (VAB) felt that the Bank and the
donor community, by failing to support economic reintegration mea-
sures, had not followed through on a useful demobilization initiative.
For many of the ex-soldiers in the north, this may have created an army
of discontents ready to operate outside the law and even join the rebels.
NURP too was criticized for having no provision for ex-combatants. It
was the choice of project planners with limited resources to demobilize
greater numbers rather than demobilize lesser numbers with more
resources for economic integration.
61
Uganda: Post-Conflict Reconstruction
This shortcoming was noted at the end of phase I of the demobiliza-
tion (the first 22,000), and a provision was built into phase II. Gener-
ally, VAB was complimentary about the Bank’s performance, noting
that after some teething problems at the beginning, funds flowed, “things
worked well with the resident mission,” and the program “was privi-
leged to have serious people.”
Throughout phases I and II, the Bank’s con-
While the impetus for demobilization tribution was to support operating costs. For
came from the Bank’s desire to see phase III, funds of US$250,000 were expected
defense expenditures drop, it found a through SAC II through a technical arrange-
positive reception with the Ugandan
ment with the Ministry of Finance. This was
President, who was keen to confirm a
peaceful outcome for Uganda’s years of forthcoming, and VAB, which was accus-
conflict and ready to reap a peace tomed to dealing directly with the Bank, did
dividend. In this context, the Bank was not know how to access the SAC: “Having
certainly successful in mobilizing donor written many times, it had given up.” An
resources in support of the program. evaluation of phase III is under way, and
should uncover this problem.
Given that the army was now having to remobilize in part to deal
with problems in the north, the possibility of premature demobilization
was raised by many respondents. Making the Bank a scapegoat for this
overlooks the readiness of the government to proceed with demobiliza-
tion, a view that is easy to criticize in hindsight but probably made
sense at the time. For the government, it was a difficult choice between
taking risks with national security and with national finances.
Bank Evaluations and Findings
A brief summary of past self-evaluations and independent evalua-
tions in Uganda provides a perspective on the main problems in design-
ing, implementing, monitoring, and evaluating projects in the
post-conflict phase. This summary should be compared with previous
sections, which explain the special characteristics of the post-conflict
context and discuss some of the main assumptions in the portfolio.
Reconstruction after a prolonged war is a formidable task. The main
challenge is to change the “rational project model” that works under
normal conditions. The post-conflict phase has vastly different require-
ments from the mainstream development universe. Thus, holding on to
the conventional project wisdom and assumptions does not work.
An important issue raised by the audit of the First Reconstruction
Credit was that, with the economy in ruins and the government trying to
rebuild its own capacity, Uganda was clearly unlikely to make optimum
62
Special Issues
use of such assistance. Under these circumstances, the criteria for evalu-
ating project success may be rather different from those normally
applied. The audit of the Second Economic Recovery Credit highlights
some of the challenges posed during the evaluation process. The evalu-
ation noted that the “assessment of ERC II’s outcome is complicated by
the fact that Uganda has had a number of adjustment operations that
overlapped with this credit. Attribution of outcomes to one particular
credit is therefore difficult.”
Key Lessons
Past evaluations raise some general issues and identify key lessons, including the
importance of:
n Establishing a dialogue
n Preserving flexibility
n Avoiding unrealistic expectations
n Considering structural constraints
n Keeping procurement and disbursement procedures consistent with the post-conflict
context
n Ensuring consistency of coordination in the Bank’s strategy.
Establishing a Dialogue
The ERC I sheds light on the importance of opening an avenue for
discussion of recovery policy. Although the direct impact of the credit
was limited, it provided a valuable opportunity to initiate a dialogue on
a wide range of economic policies and management decisions that have
been carried forward into subsequent operations, as well as ESW.
Preserving Flexibility
Past evaluations point to the need for greater built-in flexibility in project
design in cases where a country’s institutional environment is unsettled
and where the prospects that project benefits will be sustainable are more
uncertain. A flexible project design should in-
corporate both “blueprint” and “process” Project preparation, implementation,
and evaluation must take into account
elements with more emphasis on the latter.
the unique post-conflict context.
Several components may usefully be left
underdesigned at appraisal.
Avoiding Unrealistic Expectations
The realization that the process of rehabilitation is complex and lengthy
often comes too late in the project process. Thus, expectations are unre-
alistic, and good intentions lead to a clustering of too many components
63
Uganda: Post-Conflict Reconstruction
under one umbrella. Expectations were not realistic in either ERC I or in
PAPSCA. Implementation proved difficult from the start in both cases.
Considering Structural Constraints
ERC I was not well targeted and, by focusing on the provision of for-
eign exchange without addressing other equally important constraints,
was overoptimistic. An audit cautioned about the tradeoffs between the
dual objectives of rapid disbursement and policy or institutional reforms
pursued within the same operations. It also cautioned against the use of
conditionality that is overly broad in scope.
Weak targeting and broad conditional- In ERC I, while it appears that the credit con-
ity were a reflection of inadequate tributed to increased capacity utilization in
attention to structural socioeconomic target industries and expansion in agricul-
and political limitations in post- tural output, production was critically con-
conflict recovery.
strained not only by the scarcity of foreign
exchange but also by many other factors such
as the shortage of local transport, services, finance, and manpower that
the credit was not designed to address.
In PAPSCA, remote and scattered locations of activities as well as the
time-consuming nature of mobilizing dispirited communities have
increased the cost of implementation. In addition, because of centralized
procurement, implementing agencies received construction materials in
bulk, leading to logistical problems of delivery, storage, and safe-guard-
ing that were not addressed in project design or costs.
Keeping Procurement and Disbursement Procedures
Consistent with the Post-Conflict Context
In many operations (ERC I, ERC II, PAPSCA, NURP), disbursements
were slower than expected because of the unforeseen political changes
and also because the procurement procedures
The experience here points to the need chosen for the credits were not conducive to
for using flexible procurement
arrangements in community-based
rapid disbursement. In both PAPSCA and
projects. NURP, cumbersome procurement and dis-
bursement procedures led to untimely deliv-
ery of goods and replenishment of funds.
Project management and disbursement were highly centralized, with
implementing agencies having no independent access to funds. Construc-
tion materials were packaged into larger contracts and procured through
ICB. In some cases it took 18 months after credit effectiveness for com-
munities to receive materials.
64
Special Issues
Ensuring Consistency in Bank Programs
Project experience in the PEP and the Enterprise Development Project
(EDP) underlines the need for better coordination of the Bank’s CAS.
PEP’s emphasis on strengthening parastatals was at odds with the EDP
objectives of parastatal reform and parallel adjustment programs, which
aimed to impose hard budget constraints on the operations of parastatals.
The approval of EDP three years before the planned closing of PEP
created substantial confusion.
Supporting Priority Areas
Power is a priority area in post-conflict recovery, but power operations
fell short of responding to critical needs. The Second Power Project
approved in 1985 was the Bank’s first operation in Uganda’s energy sector
in more than 20 years. It took eight years from project signing to credit
closure, compared with an earlier estimate of five years. The project’s
design underestimated the impact of the unstable situation and the needs
faced by the implementing agencies in ensuring day-to-day operation of the
power system and competing with the requirements of project implementa-
tion. The Second Power Project did not achieve its objective to improve the
Uganda Electricity Board’s (UEB’s) financial performance and operational
efficiency. Although the number of staff UEB employed doubled during project
implementation, staff productivity, measured as the number of connections
per staff employed, remained significantly below initial targets.
OED Evaluation of Economic Recovery Credits
To What Extent Did the OED Audits of the Two Economic
Recovery Credits Focus on the Important Issues?
The audit of ERC I rated it “unsatisfactory.” It made three key points.
The first was to stress the unusual importance of TA in the context of a
“nascent” government. This emphasis was, in our view, correct. In a
post-war context, with the civil service in disarray, the rebuilding of
professional capabilities was the essential prelude to subsequent success.
The second point was the greater need to devise better methods of
disbursing program aid than to condition it on a prior flow of import
documentation. This was also an appropriate concern. The civil service
was not in a position to deliver the documentation required by the
donors in a sufficiently orderly fashion to ensure that program aid would
be a stabilizing influence. Even when the documentation required was
available, it provided only the facade of a genuine aid audit (since
65
Uganda: Post-Conflict Reconstruction
import documentation need not be generated by aid dollars). There were
better and simpler ways of auditing the use of program aid.
The third point was the importance of a rapid increase in tax revenue
to offset a decline in aid. Numerous strongly phrased remarks on “sustain-
ability” echoed the IMF’s priorities. These assessments were ill-judged in
three ways. First, as predictions they were inaccurate: the net aid flow to
Uganda substantially increased in subsequent years and shows no signs of
diminishing. Second, they failed to appreciate the distinctive features of a
post-war recovery. Rather, the audit analyzed the ERC I as if it had been
a standard stabilization problem. The time period for fiscal recovery in a
post-war environment should be determined by the need to restore the
taxable base of the economy, and this requires a prolonged period of low
taxation. The attempt to raise tax receipts rapidly actually delays the
attainment of sustainability. Third, the sentiments were ill-judged pre-
cisely because they appear to be so reasonable. An audience is predis-
posed to accept that large aid-financed fiscal deficits must be reduced as
rapidly as possible and that failure to do so reflects badly upon the recal-
citrance of a government to make the requisite “tax effort.” The language
of “sustainability,” “resource mobilization,” and “effort” subtly implies
the fecklessness of government. While in many situations this imagery
may be appropriate, in the context of the Ugandan post-war recovery it
was both erroneous and dysfunctional. Its usage, and its endorsement by
the audit, reinforced a simple, potent diagnosis of the recovery problem
that was the seedbed of serious errors in fiscal policy.
The audit of ERC II correctly targeted the slow pace of civil service and
parastatal reform, consistent with our own prioritization of the reform of
public institutions. However, its main point was
The two OED audits had in common a again to stress the theme of fiscal sustainabil-
time period for a post-war recovery ity. Overall, it downgraded ERC II to “margin-
that was unreasonable, although it ally unsatisfactory.”
would have been appropriate for a
This, in turn, led both to negative evalu-
stabilization and liberalization program
in a peacetime economy. As a result, ations for the two ERCs and to a policy
the audits treated the fiscal deficit as inference that sounded like “motherhood
a major problem instead of as an and apple pie,” but was inappropriate. One
appropriate response to post-war measure of these errors is that the Ugandan
characteristics. recovery is now rated as a model for the
continent. This recovery, already spread
over a decade and only approximately halfway back to the pre-war
Ugandan economy, was financed by two credits rated “unsatisfactory”
and “marginally unsatisfactory.”
66
Close collaboration between the government and the international commu-
nity, along with strong government ownership, has given Uganda’s recovery
the reputation of a “ model reconstruction.” While the Bank has demonstrated
generally good performance in reforming and rebuilding the economy, sev-
eral areas for improvement have been identified. Bank performance in the
social sectors has been relatively poor and should be improved through the
use of effective partnerships with NGOs that are better equipped to imple-
ment social programs. Regarding Bank processes and institutional arrangements, project design
did not fully reflect the need in Uganda’s unsettled institutional environment for a flexible,
process-oriented design. A final shortcoming of Bank projects was that, where they were not
sequential, many were too short to address the long recovery process.
8
Conclusions: Post-Conflict Lessons
from the Uganda Experience
The Post-Conflict Context in Uganda and Implications for
Bank Activities
Differing Post-Conflict Time Tables
There are at least two different timetables in post-war recovery, each
well-manifested in Uganda. The first is the real-time duration of the
recovery process, which typically requires at least two decades of sus-
tained effort, with the risk of war a recurrent phenomenon. The second
timetable is set by donors’ considerations, which may be guided by dif-
ferent objectives, and manifests itself through programs that do not nec-
essarily take into consideration the very long recovery process.
Governments have to observe both these timetables, weighing the politi-
cal expediencies of short-term measures to provide security and boost
confidence, as well as the steps necessary for long-term recovery.
Implications for Bank Activities and Policy Reform
Prolonged social disturbances imposed potential constraints on both
delivery of public services and tax revenue. Given the realistic timetable
for post-war reconstruction, these constraints would be with the economy
for perhaps two decades.
67
Uganda: Post-Conflict Reconstruction
Lessons Learned
n The costs of failing to recognize the duration of real recovery are substantial. Recovery
requires incremental planning, careful and realistic policy reforms, and consideration
of post-war constraints. In project terms, “ process” may need greater recognition than
a “ blueprint,” preserving flexibility with respect to implementation and targets.
n Bank lending to accelerate development needs to focus not only directly on the objectives of
policy reform and increased service delivery but also on relaxing the long-term constraints
that would otherwise delay realization of these objectives. There were two main practical
implications. First, the critical path to reform and service delivery was the enhancement of
human capital and organizational efficiency in the public sector. Second, the critical path to
fiscal sustainability was through the reformalization and recommercialization of private
activity through the suspension of predatory taxation. These two considerations should
guide activities and evaluation in the second decade of recovery.
The Bank’s Overall Role and Comparative Advantage
Comparative Strengths and Weaknesses
The role of the Bank was key in the institutional strengthening of the
Ministry of Finance, the Ministry of Planning and Economic Develop-
ment, and the Central Bank. It was also pivotal in removing the monopoly
of the Coffee Board, assisting in sugar rehabilitation, and rebuilding roads.
However, the Bank did not always fulfill its potential comparative advan-
tage in areas such as the power sector—a key element in recovery. Nor
did it fully use its coordination role by creating an overall strategy for
reconstruction or a sector-by-sector plan. The Bank lagged behind in the
social sectors, particularly in the strengthening of health and education
institutions, where partnerships with NGOs were not fully used.
Project Design
The more unsettled a country’s institutional environment, the greater
the need for a flexible, process-oriented project design. Project design
did not fully reflect this need, especially in relation to the social sectors,
where education and health ministries were too weak to accommodate
spending and supervening events such as decentralization and renewed
conflict changed priorities. Moreover, where projects were not sequen-
tial, many Bank operations were too short to address the projected length
of recovery.
Staff Resources
Changes in key resident mission staff, task managers, and govern-
ment staff were inevitably disruptive at times. Bank staff were not gener-
68
Conclusions: Post-Conflict Lessons from the Uganda Experience
ally familiar with working in conflict countries or with the international
relief and rehabilitation system. Staff resources and time were concen-
trated in Kampala, often with little understanding of contact with inter-
national and local NGOs.
Lessons Learned
n The Bank should focus on the areas in which it has a comparative advantage, such as
defining an overall reconstruction program, defining the macroeconomic framework,
and coordinating donors internationally through CG meetings, and locally through
donors’ meetings. In areas of comparative disadvantage, such as projects in the social
sector, more use should be made of other agencies or NGOs that are better prepared to
implement them. If attempted, social sector projects require much greater in-country
partnership with these agencies.
n Flexibility and an accommodation of the length of the recovery period are important
elements in design. Furthermore, a flexible project design demands strong project man-
agement. There are no substitutes for ownership on the borrower’s side, and in-depth
country knowledge and careful supervision on the Bank’s side.
n Continuity and cohesiveness of the Bank’s country team and the borrower’s project
management team enhance learning and mutual trust, increasing prospects for the
project’s success. Staff specialized in “ war-to-peace” transition countries and famil-
iar with the international relief and rehabilitation system and NGOs will enhance
partnerships.
The Bank’s Role in Aid Coordination
Coordination
The Bank has been effective in coordinating the donors, although
less so with the EU and the NGO community. Differences with the
NGOs are partly ideological but also result from a gap in communica-
tion and mutual understanding. Conflict in the north of the country
will complicate the coordination picture, because it will draw in extra
elements of international relief and rehabilitation. The Bank did not
fully convert its coordination function into an overall strategic leader-
ship role or into sectoral strategies.
Information
In the politically charged atmosphere of post-conflict societies, the
need for the Bank to assert its neutrality and professionalism depends
upon clear communication of its purpose. If the Bank can demonstrate
an open and positive attitude toward information dissemination, it will
set a powerful example for government and donors alike.
69
Uganda: Post-Conflict Reconstruction
Lessons Learned
n The resident mission needs to improve coordination, particularly with the EU and
NGOs. It should establish effective partnerships with NGOs that have the capacity
to implement social programs. It also needs to actively promote public understand-
ing of the Bank’s role and activities.
n The Bank should be more proactive in publicizing its activities as a logical
extension of its coordination role.
The Bank’s Role in Stabilizing, Reforming,
and Rebuilding the Economy
Recovery Program Lending and Rebuilding Private Capital
Uganda conforms to the record of long-term, post-conflict countries
requiring two decades or more to reverse the depletion of financial,
human, social, and public capital. In the interim, these processes impose
constraints on policy reform and service delivery. The most destructive
economic effect of the civil war was the flight of private capital. If
reversed, it will yield a large peace dividend, but this is dependent on
policy choices. The 1986–94 period experienced atypically slow growth.
Only following major policy changes in 1992 did investor confidence
rise and private investment recover. Although not subject to the same
flight as private capital, public capital suffered depletion through
destruction and a government squeeze on savings to fight the war.
Despite good performance in reforming and rebuilding the economy,
Bank involvement could have been improved by giving greater attention
to consensus building; reducing use of conditionalities; and, most impor-
tant, rejecting a seriously dysfunctional emphasis on raising taxation.
Lessons Learned
n The Bank needs to focus on the restoration of all aspects of private capital—
financial, physical, human, and social— as well as of public capital. Of the three
improvements noted above, a low level of taxation over the recovery period in a
post-conflict country such as Uganda is critical. Even where aid is set to decrease
in the medium term, the fastest route to self-reliance is unlikely to involve rapid
increases in the tax burden in the short term.
n The Bank’s role in rebuilding economic and infrastructure sectors remains critical.
Technical assistance in these sectors, as well as the power sector, are clear areas
of comparative advantage for the Bank, where past experience should afford the
resources to produce timely and critical inputs.
70
Conclusions: Post-Conflict Lessons from the Uganda Experience
The Bank’s Role in Rebuilding Infrastructure
and Economic Sectors
Of the Bank projects launched subsequent to the ERC, the TA and
sugar projects had the most impact. Together, these two projects cost
only US$43 million. Yet they were precisely appropriate for a post-war
transition because they rebuilt human capital in the civil service and
strengthened private business confidence. The two infrastructure projects
in railways and telecommunications should have had more emphasis
on organizational redesign and human capital improvement. The PEP
appropriately focused on institutional reform in the public sector. The
agricultural project helped to facilitate crucial reforms, enabling the
dismantling of the monopoly of the Coffee Marketing Board.
The urban infrastructure project was timely because by 1991 Uganda
was starting to reurbanize, and this process subsequently accelerated.
The power project was arguably too late and insufficiently focused on
institutional reform. It also experienced severe implementation difficul-
ties. The enterprise development loan included elements that rapidly
became dated.
The Bank’s Role in Rebuilding Human and Social Capital
Social Sectors
Investment in the social sectors was often premature, implemented
through weak ministries, with inadequate forethought as to the nature
of the services the Bank was trying to support. The post-conflict period
required health sector reform that took into account the inevitable role
of the private sector and user funding for public services. Education
investment was equally disappointing—only an estimated 37 percent of
funds reached schools. There was too much dependence on the bureau-
cracy behaving efficiently and honestly, and not enough use made of
NGOs.
Social Capital, Corruption, Participation,
and Transparency
Losses of social capital in Uganda have been as destructive as losses
of other forms of capital. Standards collapsed, notably in the legal and
banking professions and the civil service, and these have yet to be fully
restored. The abuse of public funds for personal gain remains a prob-
lem across all sectors in Uganda. There is surprising readiness among
public officials to have an open discussion of this matter and a free
media that is keen to play a watchdog role.
71
Uganda: Post-Conflict Reconstruction
Lessons Learned
n In planning recovery, it should be acknowledged that the public sector may perform
badly or that it may not be the best choice for providing a particular service to the
public. The Bank should encourage more strategic planning to determine what kind
of sustainable health and education services are envisaged in the long term. The
key reform in social services provision, given the collapse of standards in the civil
service, is probably to give the frontline service provider— such as schools and
clinics, their budgets, and their staff— incentives linked to monitorable performance
measures. NGOs offer critical partnerships in achieving these objectives. There
should be a greater focus on education, especially vocational training, teacher
training, and training of trainers. Educational curriculums should include cross-
cutting themes, emphasizing peaceful settlement of disputes and reinforcing
behavior that emphasizes mutual respect and discourages random violence.
n The rebuilding of standards in the professions— for doctors, lawyers, accountants, and
civil servants— is likely to be at the core of restoring social capital. Since these
professions largely depend on self-regulation, one strategy is to internationalize
accreditation procedures. This may be a priority for the World Bank Development
Institute, which has done some innovative work on governance. The Bank could set high
standards by promoting effective participation and transparency in its own activities.
The Bank’s Role in Ongoing Conflict: Northern Uganda
Reconstruction Project
The North and Further Conflict
Opinions on the effectiveness of NURP vary greatly, and an audit
would be useful. To some extent, NURP has been the scapegoat for trouble
and disaffection in the north. However, it has suffered from implementa-
tion problems that were partly of the Bank’s making (largely in the
design), partly the result of financial mismanagement, and partly a product
of circumstance. Managing a project with such changeable circumstances
“by mission” has been demanding.
Lessons Learned
n Any successor project to NURP would be usefully guided by an audit. In any event,
it will have to take account of both continuing conflict and the need to engage with
the broadening international relief and rehabilitation system. Coordination with
donors in both planning and implementation will be critical. The project will also
have to incorporate changes brought about by decentralized government. More use of
NGOs, especially local organizations, should be explored.
72
Conclusions: Post-Conflict Lessons from the Uganda Experience
Special Issues
Demobilization
A distinction should be drawn between the political decision to demo-
bilize and the actual process of planning and execution. While it had
good intentions, demobilization was initially delayed for the wrong rea-
sons (societal and individual security), and then implemented at an
inappropriate time considering state security factors. Because of renewed
conflict in the north and border tensions, about one-third of the demobi-
lized soldiers have been recruited again. The Bank might have coun-
seled greater caution, but the decision was political and remained with
the government. Regarding process, demobilization was clearly a suc-
cess, even though the Bank failed to get donor support for more eco-
nomic measures to assist the reintegration of ex-combatants.
Procurement and Disbursement
Procedures for procurement and disbursement have been major stum-
bling blocks in the recovery process across sectors.
Monitoring and Evaluation
Where conducted, monitoring and evaluation generally provided valu-
able information on program and project effectiveness. In particular, the
Bank-financed tracking study on primary health and education spending
confirmed widespread donor fears about the failure of social sector funding.
Evaluation of innovative areas of work has been useful in the demobiliza-
tion process. However, the post facto evaluation of the ERCs failed to as-
sume the correct time period for post-conflict recovery, overstressing fiscal
sustainability.
Lessons Learned
n The current need to demobilize or remobilize depends very much on the progress of
conflict in the north. State security rather than societal and individual security will likely
be the basis for such a decision. In this context, the Bank does not have a comparative
advantage. Evaluation of phase III of the past demobilization should produce key lessons
for any future operations. The Bank’s involvement in future demobilization programs
depends upon a critical assessment of its priorities, its in-country role, and the ability of
other partners in the international system to provide support.
n Emergency procedures should be invoked where appropriate, but even these, as they
stand, may not meet urgent reconstruction needs for vulnerable groups, for example.
There is a need for major changes in procurement and disbursement procedures and for
intense training of both Bank and government staff in post-conflict areas.
n Systems for monitoring and evaluation should be established, particularly to track
investments and to follow up on developments in the areas where the Bank has
limited experience. OED evaluations in post-conflict countries need to accommo-
date the special characteristics of recovery.
73
Annex
List of Persons Interviewed
Government of Uganda NGOs and Other
Organizations
Central Bank
Charles Kikonyogo, Governor British High Commission (ODA)
Louis Austin Kasekende, Petra Byrde, First Secretary (Aid)
Director, Research and Policy
CARE
Ministry of Education Nick Ritchie, Country Director
Patrick Makumbi, Director,
Project Implementation Unit DANIDA
Noreda Kiremire, Deputy Daniel S. Iga, Program Officer
Director, Project Implementa- European Union (EU)
tion Unit John Croswaithe, Counselor
Ministry of Finance (Economics)
E. Tumusiime-Mutebile, Perma- Marja Laine
nent Secretary/Treasurer NURP
Parliament Ernest Oloya, Coordinator
The Honorable Norbert Mao Margaret Ajju
Prime Minister’s Office Oxfam
Peter Uchanda, Permanent Antony Burdon, Country
Secretary Representative
Veteran’s Assistance Board Save the Children Fund (U.K.)
Maj. Gen. (Rt.) Emilio Mondo, Richard Mawer, Field Director
Executive Secretary SNV (Holland)
Capt. Francis Apiko (Rt.), Evelyn Nyakoojo, Program
Program Manager Officer
75
Uganda: Post-Conflict Reconstruction
Ugandan Red Cross WorldVision
Peter Oryema, Secretary-General Moses Dombo, Policy Advocacy
Edward Mubirau, Program
UNDP Officer
Eugene Owusu, Resident Econo-
mist Field Visit: Luwero
Lawrence Nkooto Bategeka, Chairman, District Council
National Program Officer District Inspector of Education
District Planning Officer
UNICEF
Kathleen Cravero, Representa- Field Visit: Pallisa
tive Chairman of the Council
Keith Wright, Chief, CCA Masaba Justine, RCV’s Office
Program G. Kayongo, Chief Administrator
Omaido Enoch, DTLO
USAID Mr. Shaine, Chairman, Educa-
Patrick Fine tion Committee
World Bank
Marc Baird
John Clark
Nat Colletta
Sabine Cornelius
Randolph Harris
Yitzhak Kamhi
Petter Langseth
Peter Miovic
Ritva Reinikka
John Riverson
Iraj Talai
Gaiv Tata
76
Endnotes
1. Collier (1996) analyzes the growth effects of civil war using the Penn
World Tables data set.
2. Collier, Hoeffler, and Pattillo (1997) analyze capital flight from Africa
during the period 1970–90 and compare it with data on domestically owned,
private fixed capital stock in Africa (estimated by totalling investment and
allowing for depreciation).
3. Collier (1997) elaborates on this argument.
4. After the phase II evaluation, this element was monetized and transferred
as part of the overall cash safety net.
77
Bibliography
Ablo, Emmanuel, and Ritva Reinikka. Forthcoming. “Do Budgets
Really Matter? Evidence from Uganda.” Washington, D.C.: World
Bank.
Azam, Jean-Paul, David Bevan, Paul Collier, Stefan Dercom, Jan
Gunning, and Sanjay Pradhan. 1994. Some Economic Conse-
quences of the Transition from Civil War to Peace. Policy Research
Working Paper 1392. Washington, D.C.: World Bank.
Brett, E.A. 1993. Providing for the Rural Poor: Institutional Decay
and Transformation in Uganda. Uganda: Fountain.
Colletta, Nat, Markus Kostner, and Ingo Wiederhofer. 1996a. Case
Studies in the Transition from War to Peace: The Demobilization
and Reintegration of Ex-combatants in Ethiopia, Namibia, and
Uganda. World Bank Discussion Paper 331. Washington, D.C.
Colletta, Nat, Markus Kostner, and Ingo Wiederhofer. 1996b. The
Transition from War to Peace: Directions in Development. Wash-
ington, D.C.: World Bank.
Collier, P. 1997. “Ugandan Trade Policy: Liberalisation in an Environ-
ment of Limited Credibility.” In C. Milner, ed., forthcoming, World
Trade Policy. Oxford, U.K.: Blackwell.
Collier, P. 1996. “On the Economic Consequences of Civil War.”
Centre for the Study of African Economies, Oxford, U.K. Photo-
copy.
Collier, P., and A. Hoeffler. 1996. “On the Economic Causes of Civil
War.” Centre for the Study of African Economies, Oxford, U.K.
Photocopy.
79
Uganda: Post-Conflict Reconstruction
Collier, P., and S. Pradhan. 1997. “The Ugandan Economic Transition
from War to Peace.” In B. Hansen and M. Twaddle, eds., Develop-
ing Uganda. London: James Currey.
Collier, P., A. Hoeffler, and C. Pattillo. 1997. “Portfolio Choice and
Private Capital in Africa.” Centre for the Study of African Econo-
mies, Oxford, U.K. Photocopy.
Kapoor, Kapil. 1995. Restructuring Uganda’s Debt: The Commercial
Debt Buy-back Operation. Policy Research Working Paper 1409.
Washington, D.C.: World Bank.
Langseth, P., and Stuert. 1997. Uganda: Landmarks in Rebuilding a
Nation. Uganda: Fountain.
Langseth, P., and J. Mugaju. 1996. Post-conflict Uganda: Towards an
Effective Civil Service. Uganda: Fountain.
Macrae, J., and others. 1995. A Healthy Peace? Rehabilitation and
Development of the Health Sector in a Post-conflict Situation: The
Case of Uganda. London: London School of Hygiene and Tropical
Medicine.
Mutibwa, P. 1992. Uganda Since Independence. Uganda: Fountain.
Villadsen, S., and F. Lubanga, eds. 1996. Democratic Decentralisation
in Uganda. Uganda: Fountain.
World Bank. 1995. Uganda: The Challenge of Growth and Poverty
Reduction. Report 14313. Washington, D.C.
———. 1994b. “Survey of Investor Attitudes to East Africa.” Africa
Technical Department, Washington, D.C. Photocopy.
———. 1994a. Republic of Uganda: The Role of Nongovernmental
Organizations and Community-Based Groups in Poverty Allevia-
tion. Report 12262-UG. Washington, D.C.
———. 1993. Uganda: Growing Out of Poverty. Report 12029.
Washington, D.C.
Widner, J. 1997. “The Courts as Restraints.” In P. Collier and C.
Pattillo, eds., Risk and Agencies of Restraint: Reducing the Risks of
African Investment. London: Macmillan.
80
Operations Evaluation Department
Publications
The Operations Evaluation Department (OED), an independent
evaluation unit reporting to the World Bank’s Executive Directors,
rates the development impact and performance of all the Bank’s com-
pleted lending operations. Results and recommendations are reported
to the Executive Directors and fed back into the design and imple-
mentation of new policies and projects. In addition to the individual
operations and country assistance programs, OED evaluates the Bank’s
policies and processes.
Summaries of studies and the full text of the Précis and Lessons & Prac-
tices can be read on the Internet at http://www.worldbank.org/html/oed
How to Order OED Publications
Operations evaluation studies, World Bank discussion papers, and all
other documents are available from the World Bank InfoShop.
Documents listed with a stock number and price code may be
obtained through the World Bank’s mail order service or from its InfoS-
hop in downtown Washington, D.C. For information on all other docu-
ments, contact the World Bank InfoShop.
For more information about this study or OED’s other evaluation
work, please contact Elizabeth Campbell-Pagé or the OED Help Desk.
Operations Evaluation Department
Partnerships & Knowledge Programs (OEDPK)
E-mail: ecampbellpage@worldbank.org
E-mail: eline@worldbank.org
Telephone: (202) 473-4497
Facsimile: (202) 522-3200
Ordering World Bank Publications
Customers in the United States and in territories not served by any of
the Bank’s publication distributors may send publication orders to:
The World Bank
P.O. Box 960
Herndon, VA 20172-0960
Fax: (703) 661-1501
Telephone: (703) 661-1580
The address for the World Bank publication database on the Internet
is: http://www.worldbank.org (select publications/project info).
E-mail: pic@worldbank.org
Fax number: (202) 522-1500
Telephone number: (202) 458-5454
The World Bank InfoShop serves walk-in customers only. The InfoS-
hop is located at:
701 18th Street, NW
Washington, DC 20433, USA
All other customers must place their orders through their local
distributors.
Ordering by e-mail
If you have an established account with the World Bank, you may
transmit your order by electronic mail on the Internet to:
books@worldbank.org. Please include your account number; billing and
shipping addresses; and the title, order number, quantity, and unit price
for each item.