Toward Country-led Development A Multi-Partner Evaluation of the Comprehensive Development Framework Findings from Six Country Case Studies BOLIVIA BURKINA FASO GHANA ROMANIA UGANDA VIETNAM The Country Case Studies in this volume contributed to the Multi-Partner Evaluation of the Comprehensive Development Framework. The evaluation has been supported and guided by the following organizations and countries, which form its Steering Committee. Bilateral Aid Agencies: Multilateral Aid Agencies: Governments: Canadian International TThe World Bank Bolivia Development Agency ­ Operations Evaluation Department Burkina Faso Danish International ­ Development Research Development Assistance, Group Ghana Ministry of Foreign Affairs African Development Bank Romania Japan Bank for International Cooperation Asian Development Bank Uganda Economic Cooperation European Commission Vietnam Bureau, Ministry of Foreign Affairs, Japan Organisation for Economic Co-operation and Ministry of Foreign Affairs, Development The Netherlands ­ OECD Development International Centre Nongovernmental Ministry of Foreign Affairs, ­ OECD DAC Network on Organization: Norway Development Evaluation Oxfam Great Britain Swedish International United Nations Development Authority Development Programme Swiss Agency for United Nations Economic Private Sector: Development and Commission for Africa Cooperation Pakistan, Packages Limited United Kingdom Romania, TOFAN Department for International Development United States Agency for International Development Toward Country-led Development A Multi-Partner Evaluation of the Comprehensive Development Framework Findings from Six Country Case Studies BOLIVIA BURKINA FASO GHANA ROMANIA UGANDA VIETNAM © 2003 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 Telephone 202-473-1000 Internet www.worldbank.org E-mail feedback@worldbank.org All rights reserved. 1 2 3 4 05 04 03 The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the organizations or institutions they represent, or of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denom- inations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. Copying and/or transmitting portions or all of this work without permis- sion may be a violation of applicable law. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ISBN: 0-8213-5669-0 e-ISBN: 0-8213-5670-4 Library of Congress Cataloging-in-Publication Data has been applied for. Contents iii Country Case Study Teams vii Abbreviations and Acronyms ix 1. Introduction 1 The Broad Development Context for the CDF Evaluation 1 Purpose of This Volume 1 Selection of Country Case Studies 2 Country Study Methodology 2 Main Messages from Country Case Studies 2 CDF Principles--Working Definitions 3 2. Bolivia 7 Methodology 7 Bolivia and the CDF: A Glass Half Full--Or Half Empty 7 Unique Country Characteristics 7 Long-Term, Holistic Development Framework 10 Country Ownership 12 Country-led Partnership 14 Results Orientation 17 Central Issues 19 3. Burkina Faso 21 Methodology 21 Burkina Faso: A Proud Development History--and Daunting Challenges 21 Long-Term, Holistic Development Framework 25 Country Ownership 26 Country-led Partnership 28 Results Orientation 30 The Way Forward 32 4. Ghana 35 Methodology 35 Background 35 Long-Term, Holistic Development Framework 38 Country Ownership 40 Country-led Partnership 40 Results Orientation 43 Conclusions 44 iv 5. Romania 47 Methodology 47 Romania: A Long and Rocky Transition 47 Long-Term, Holistic Development Framework 49 Country Ownership 50 Country-led Partnership 54 Results Orientation 55 CDF Interrelationships 57 A Proposal 57 6. Uganda 61 Methodology 61 Uganda: Rising from the Ashes 62 Long-Term, Holistic Development Framework 64 Country Ownership 65 Country-led Partnership 67 Results Orientation 70 Sequences and Reinforcing Relationships among CDF Principles 72 Budget Support, Aid Quality, and Fiduciary Assurance 72 7. Vietnam 75 Methodology 75 Vietnam and the CDF: Progress, Yes, but Challenges Remain 75 Long-Term, Holistic Development Framework 77 Country Ownership 78 Country-led Partnership 80 Results Orientation 83 Tensions 84 Annex: Evaluation Methodology 87 Endnotes 91 Boxes Box 2.1: Bolivia at a Glance 8 Box 2.2: Customs Service Reform: Tough Government Decisions, with 13 Help from International Agreements Box 2.3: Basket Funding Lessons from the Ombudsman's Office: 15 It Takes Time and Money to Reap the Benefits Box 2.4: The World Bank and "Planet Bolivia" 17 Box 3.1: Burkina Faso at a Glance 23 Box 3.2: Donors to World Bank: "Know Thyself" 29 Box 3.3: Achieving the Millennium Development Goals: 32 A Probability Scorecard Box 4.1: Ghana at a Glance 36 Box 4.2: The First Health SWAp 41 Box 4.3: A Country with Real Potential 45 Box 5.1: Romania at a Glance 49 v Box 5.2: Romania's Shared Vision--The 1999 CDF Consultations 51 Box 5.3: Jiu Valley Development: A Regional CDF Model 52 Box 6.1: Uganda at a Glance 63 Box 6.2: An Essential Element: The Joint Sector Review 68 Box 6.3: Grassroots Monitoring of Poverty-Reduction Progress 71 Box 7.1: Vietnam at a Glance 76 Box 7.2: The MDGs: Proof of Ownership 79 Box 7.3: Lessons from the Poverty Working Group 81 Box 7.4: Consultative Group Meetings: The CDF in Microcosm 82 Country Case Study Teams vii Bolivia Team Ghana Team Romania Team Nils Boesen Nils Boesen John Eriksson Laura Kullenberg Anthony Killick B. Lynn Salinger Jose Antonio Peres A. Laura Kullenberg Dumitru Sandu Juan Carlos Requena Mirafe Marcos Abena Oduro Vietnam Team Burkina Faso Team Alf Morten Jerve Uganda Team Dominique Lallement Laura Kullenberg Kyran O'Sullivan John Eriksson Ray Mallon Patrick Plane Mirafe Marcos Keiko Nishino Kimseyinga Savadogo David Pedley Han Manh Tien Della McMillan Rosern Rwampororo Editorial Team Laura Kullenberg Linda Peterson Julia Ooro Yezena Yimer Abbreviations and Acronyms ix ADB Asian Development Bank GNI Gross National Income BEAP Business Environment Action Plan GPRS Ghanaian Poverty Reduction (Romania) Strategy (Ghana) BPRS Bolivia Poverty Reduction Strategy GTZ Deutsche Gesellschaft fur (Bolivia) Technische Zusammenarbeit CDF Comprehensive Development HIPC Heavily Indebted Poor Countries Framework HIV Human Immunodeficiency Virus CEM Country Economic Memorandum IDA International Development CG Consultative Group Association CIDA Canadian International IMF International Monetary Fund Development Agency IRP Institutional Reform Program CPRGS Comprehensive Poverty Reduction (Bolivia) and Growth Strategy (Vietnam) JBIC Japan Bank for International CSLP Cadre Stratégique de Lutte Contre Cooperation La Pauvreté (Burkina Faso) JSR Joint Staff Review CSO Civil Society Organization (Ghana) LIPDHD Letter d'Intention de Politique de (Uganda) Developpement Humain Durable Danida Danish International Development (Burkina Faso) Agency (Denmark) MDBS Multi-Donor Budget Support DFID Department for International (Ghana) Development (UK) MDGs Millennium Development Goals EBRD European Bank for Reconstruction MFPED Ministry of Finance, Planning and and Development Economic Development (Uganda) EU European Union MOFA Ministry of Foreign Affairs ESMAP Energy Sector Management MP Member of Parliament Assistance Program (World Bank) MTEF Medium-Term Expenditure FAO Food and Agricultural Organization Framework FOIA Freedom of Information Act MTES Medium-Term Economic Strategy (Romania) (Romania) GDP Gross Domestic Product NATO North Atlantic Treaty Organization x NDC National Democratic Congress PNDC Provisional National Defense (Ghana) Council (Ghana) NDPC National Development Planning PNGT Programme National de Gestion des Commission (Ghana) Terroirs (Burkina Faso) NGO Nongovernmental Organization PPL Popular Participation Law (Bolivia) Norad Norwegian Agency for Development PRSC Poverty Reduction Support Credit Cooperation PRSP Poverty Reduction Strategy Paper NPP New Patriotic Party (Ghana) PWG Poverty Working Group OED Operations Evaluation Department Sida Swedish International Development (World Bank) Authority OMS/ Organisation Mondial de Santé SWAp Sectorwide Approach WHO SWG Sector Working Group PAF Poverty Action Fund (Uganda) UK United Kingdom PAR Public Administration Reform (Vietnam) UN United Nations PEAP Poverty Eradication Action Plan UNDP United Nations Development (Uganda) Programme PEP Pre-Accession Economic Plan UNICEF United Nations Children's Fund (Romania) US United States PIU Project Implementation Unit USAID United States Agency for PMES Poverty Monitoring and Evaluation International Development Strategy (Burkina Faso) WHO/ World Health Organization OMS Introduction 1 The Broad Development Operations Evaluation Department (OED) to evaluate how the CDF approach was being Context for the CDF implemented. This resulted in a two-and-a-half- Evaluation year multi-partner evaluation effort. A broad array of stakeholders--representing donor and In early 1999, World Bank President James recipient countries, multilateral agencies, and Wolfensohn launched the Comprehensive civil society and private sector organizations-- Development Framework (CDF)--a new joined in designing and funding the evaluation. framework for how the World Bank should do The work was overseen and the findings were business with recipient countries and other endorsed by a 30-member Steering Committee development partners. Two basic ideas are at and a 5-member Management Group. The main the heart of the CDF: one, that poverty reduc- source of evidence for the evaluation comes from tion should be the fundamental goal of interna- six countries selected for in-depth case studies-- tional aid and, two, that the way aid is Bolivia, Burkina Faso, Ghana, Romania, delivered, not just its content, has an important Uganda, and Vietnam. influence on its effectiveness. Purpose of This Volume The CDF promotes four principles, each of which responds to past development assistance Collectively, the six country case studies pro- shortcomings and presents an approach for vide an unusually rich source of material on the improvement. First, development efforts should local dynamics of the aid business and the real- be rooted in a long-term, holistic perspective of a ities that countries face when they try to adopt country's needs, and should focus on results rather CDF principles in earnest. In order to make this than inputs. All assistance should be based on material more accessible (full case studies are country-owned strategies and partnership relation- over 70 pages), OED has summarized each ships among development actors, led by the country. study to about 15 pages and gathered all six While these four principles themselves are not summaries into this volume, which is intended new, bringing them together in a unified frame- to complement the main synthesis report for work and promoting them vigorously in the the CDF evaluation, Toward Country-Led global arena is an important--indeed, unprece- Development: A Multi-Partner Evaluation of the dented--innovation at the World Bank. Comprehensive Development Framework. This volume can also be used alone, to enrich the In late 1999 the Committee on Development discussion of development assistance in a Effectiveness of the Bank's Board asked the particular case study country or as a basis for 2 comparing country experiences. The CDF asked to serve as a country reference group. Evaluation Synthesis Report and the full Their job was to guide the evaluation process Country Case Studies can be found on the OED and provide feedback on design choices and Web site at www.worldbank.org/evaluation/cdf. early findings. Each mission concluded with a workshop to allow country-level stakeholders an opportunity to comment on the findings before a Selection of Country draft was offered to an international audience for feedback. For more detail on country selection Case Studies and methodology see the annex to this volume. Since the main purpose of the CDF evaluation was to look at what had happened since the Main Messages from CDF was launched, priority was given to inter- ested countries where pilot implementation of Country Case Studies the CDF was the most advanced (on the grounds that these countries would offer the Evidence from the case studies shows progress in greatest potential for learning). Consideration implementing the CDF principles, particularly was also given to Regional balance. One where one or more of the principles have been non­CDF pilot country (Burkina Faso) was applied over a number of years. chosen as a control. But these positive changes are also uneven and fragile. This is not surprising, given that the Country Study process of change is still young and that imple- menting the principles requires changes in Methodology entrenched behaviors and institutional practices on the part of both donors and recipient coun- Interdisciplinary teams of international and tries--something not easily or quickly done. The national evaluators spent several weeks in each evaluation concluded that continuous political country. A variety of methods were used to col- leadership and sustained will by all development lect data, including literature reviews, focus actors are needed if today's inefficient aid prac- group meetings, structured interviews, and field tices are to be transformed under the CDF visits to regions and districts. Each team exam- approach. Common findings across case studies ined the extent to which CDF (and CDF-like) call for several areas of immediate change. principles had been implemented, the factors that helped and hindered progress, and the Despite some improvement, especially with impact on behaviors and outcomes. The teams regard to strategic alignment, there is still sig- drew from evidence already available to chart nificant scope for donors to harmonize policies, changes in behaviors, processes, and transaction procedures, and practices with each other and costs and to identify emerging problems and with partner governments. Ultimately, what is risks. Surveys of government-donor relations, also needed to harmonize the current frag- with a focus on the cost of doing business, were mented situation is strong country leadership. carried out in five of the six country studies. In each country, a small group of advisors was Donors must be willing to change their own selected from the various stakeholder groups and internal procedures and culture to better accept and foster country leadership. These transaction costs, then the financial manage- 3 changes include: ment systems of recipient countries must be sound. Increasing the responsibility and authority of donor field offices CDF Principles-- Furthering institutional reform to shift all actors from a disbursement culture to a learn- Working Definitions ing and results culture The following working definitions are drawn Implementing new incentives systems to from President Wolfensohn's January 19, 1999, reward achievement of outcomes instead of speech launching the CDF, and documents sub- success in spending money sequently issued by the CDF Secretariat.1 Because of space constraints these definitions Making longer-term commitments and will not be repeated in each country case study increasing the predictability and reliability of summary, but are placed here for reference. donor funding. Long-Term, Holistic Development Framework Reciprocal actions by recipient countries are 1. Design of a 15-to-20 year vision statement also critical. They include: containing monitorable development goals that will: Place responsibility for aid management at a a. Take account of broad aspirations of high level of government and provide that the population. function with the authority and resources b. Include sustainable poverty reduction needed to demonstrate leadership over the as overarching goal and related sub- various operating ministries. goals aligned with the Millennium Development Goals. Where needed, give priority to overarching public sector reform to provide the environ- 2. Formulation of a coherent medium-term ment in which CDF principles can flourish. (3-to-5 year) strategy for making progress toward vision goals, specifically addressing Ensure that donor efforts to consult with need for: country stakeholder groups are undertaken in a. Balance among macroeconomic and a broad and transparent (rather than ad hoc financial issues and structural and and selective) manner, with the government social concerns in the lead. b. Priorities in the face of capacity and hard budget constraints; and time- If harmonization is to mean adopting recipi- bound, concrete actions, with atten- ent country procedures, then those proce- tion to phasing and sequencing. dures must be sound and within a range of international good practice. Country Ownership 1. Identification of development goals and If donor budget support is to be forthcoming, formulation of strategy by country, not with its implied reduction in aid delivery by donors. 4 2. Regular, broad-based stakeholder partici- governments) activities with the pation, under government leadership, country's development strategy. including civil society, the private sector, 2. Relations among government, develop- local governments, and parliaments, ment agencies, and other stakeholders, with sustained public support from top marked by: political leadership and intellectual a. Mutual trust, consultation, and conviction by key policymakers, and transparency b. Assumption of mutual strong links to institutions. accountabilities and review of partners' performance Country-led Partnership c. Demand-led support for strengthening 1. Government leadership in management government management and and coordination of development partners coordination capacity. and aid resources, including: a. Consultative Groups, Aid Roundtables, Results Orientation and other coordination mechanisms 1. Designing programs with evaluable objec- b. Analytical and diagnostic work tives that contribute to development c. Alignment of external support-- framework goals, and developing inter- including lending, grants, analytical mediate indicators toward these goals and diagnostic work, and capacity 2. Monitoring and regularly sharing progress, building--with the country's development strategy and donor with accountability for outcomes and comparative advantage goals, not just inputs d. Harmonization of development 3. Creating and enabling capacities to agency procedures and practices generate, monitor, and use results e. Alignment of internal partners' (civil information to improve performance in society, the private sector, local achieving goals and accountability. World Bank C H A P T E R T W O Bolivia 7 Methodology been marked by ups and downs in the 1990s, more recently the planning and management of international support has generally evolved The Bolivian study team was led by Nils Boesen, toward CDF principles--between 1997 and consultant, Process & Change Consultancy, and 1999 Bolivia had developed an aid framework included Laura Kullenberg of the World Bank's that largely anticipated the CDF principles. Operations Evaluation Department; and Jose International aid plays an extremely important Antonio Peres and Juan Carlos Requena, role in Bolivia: half of the country's public national consultants. investment has been supported by aid. The study was launched in October 2001 with a preparatory mission in Bolivia, followed by Unique Country fieldwork from January 21 to February 11, 2002, to collect the majority of the data for the evalu-Characteristics ation. The team benefited from the contribu- tions of a national reference group that guided Twenty years of democracy and 16 years of struc- the evaluation and provided advice during the tural reforms have changed Bolivia from an final stage of report drafting. The study's con- unstable country with a strong military presence clusions rest on a number of information into one with a free-market economy and sources: interviews with leaders and selected improved social indicators. Bolivia's reform staff of government and donor agencies and record is particularly impressive given its nongovernmental organizations (NGOs); rele- 1984­85 starting point: a budget deficit equal- vant reports and other documents; and survey ing one-quarter of the nation's gross domestic results from respondents engaged in develop- product (GDP), an inflation rate of 23.4 per- ment work in the country. The team held a cent, and a balance of payments crisis that made number of focus group discussions. The team's it impossible for Bolivia to meet its interna- findings were presented at a stakeholder work- tional obligations. shop at the conclusion of the fieldwork. In 1985, under a new government headed by President Victor Paz Estenssoro, Bolivia initi- BoliviaandtheCDF:AGlass ated a series of economic reforms centered on market economy principles that put the private Half Full--Or Half Empty sector in the driver's seat. Specific measures included free-market pricing of both salaries and While the relationship between the Bolivian interest rates and a more open stance to foreign government and international aid agencies had trade, supported by a unified official exchange 8 rate. More stringent monitoring of public In addition, the Popular Participation Law finances controlled deficit spending. The result: (PPL) was enacted in 1994 to counter Bolivia's the public sector fiscal deficit dropped to 2.7 long-standing public expenditure bias toward percent of GDP in 1986, and by 1987 a negative urban areas--through a pattern set in the colo- economic growth rate had turned around to a nial era, most government funding had flowed positive 2.5 percent. Inflation had dropped by to the capital cities of the country's departments more than half. (as Bolivia's regional subdivisions are called). Along with the 1995 Administrative Decentral- Social reforms began in earnest in the early ization Law, the PPL established a more equi- 1990s, when the government proposed the table basis for the distribution and management Bolivian Social Strategy. The main thrust of of government revenues. the Strategy was a comprehensive approach to poverty reduction. It not only increased public Reform measures extended to Bolivian institu- spending for health, education, and sanitation, tions as well. These included giving the Central but also put in place economic measures Bank the power to monitor banks and other designed to increase productivity and revenues financial entities (including pension, securities, in the country's poorest sectors. and insurance firms), and reining in monopolies. Box 2.1. Bolivia at a Glance POVERTY and SOCIAL Latin Lower- Development diamond* America middle- Bolivia & Carib. income Life expectancy 2001 Population, mid-year (millions) 8.5 524 2,164 GNI per capita (Atlas method, US$) 950 3,560 1,240 GNI (Atlas method, US$ billions) 8.1 1,862 2,677 GNI Gross per primary capita enrollment Most recent estimate (latest year available, 1995-01) Access to improved water source Poverty (% of population below national poverty line) 60 ­ ­ Bolivia Urban population (% of total population) 64 76 46 Lower-middle-income group Life expectancy at birth (years) 63 70 69 Infant mortality (per 1,000 live births) 61 29 33 Child malnutrition (% of children under 5) 7 9 11 Access to an improved water source (% of population) 75 85 80 Illiteracy (% of population age 15+) 14 11 15 Gross primary enrollment (% of school-age population) 98 130 107 Male 99 131 107 Female 96 128 107 Role of International Aid On the health front, while services delivered 9 International aid has played an extremely under Bolivia's Basic Health Insurance system important role in Bolivia's structural reforms. have improved (for example, in maternal and Between 1990 and 2001, aid averaged slightly child health), staff shortages undermine the over 9 percent of GDP. quality of care in rural areas. In the education sector, while the illiteracy rate dropped from 20 percent in 1992 to 15 percent in 1999, the over- Stubborn Problems all quality of education is still quite poor, judged Despite Bolivia's significant progress, problems by a number of standards. For example, it takes remain on all three fronts: economic, social, and all students an average of seven years to com- institutional. Economic growth has been plete the fifth grade, and less than a third of stu- modest--an average of 3.1 percent between dents in rural areas complete the eighth grade. 1986 and 2001. In per capita terms, this has not been enough to make an appreciable dent in the poverty level: at the end of the 1990s, Bolivia Progress toward institutional capacity building is remained one of the poorest countries in the limited by the significant gap between how laws Americas, with a poverty incidence of 63 per- and regulations formally define operations, and cent, ranking behind only Honduras. In rural how public institutions actually operate. areas, poverty incidence reached 82 percent. Cronyism is rife; the fragmentation of the politi- cal system produces coalition governments under Bolivian informants interviewed for this evalua- which parties forge agreements to share state tion point to a crisis of political representation patronage. Concerns about corruption--the in the country, with a significant segment of the misuse of public power for private benefit--are population excluded politically and marginal- pervasive in Bolivia. Indeed, the country scored ized economically. This is particularly true of the 2.2 on a scale of 10 (highly clean) to 0 (highly indigenous peoples, who make up nearly 60 per- corrupt) in the 2002 Corruption Perceptions cent of the populace and who have traditionally Index by Transparency International. This score settled in rural areas, where the poverty level is reflects perceptions of business people, academ- 82 percent. The incidence of poverty combined ics, and risk analysts on the degree of corruption with political exclusion is a combustible mix- among Bolivia's public officials and politicians. ture: over the past several years the lack of a legitimate dialogue or common meeting ground It is against this backdrop of mixed indicators has resulted in demonstrations and violent that the CDF principles are being implemented street clashes. With the 1985 shift to market in Bolivia: on one hand, real economic, social, economy principles, the traditional "functional" and institutional progress; on the other, acute organizations such as the Bolivian Workers' tension and conflict. It is therefore a complex Center and affiliated organizations such as the context: one that both shapes the way CDF Trade Union Federation of Mining Workers principles are applied in the country and that lost much of their capacity to serve as a channel the CDF principles themselves help shape. for representation and dialogue with the politi- cal system. While there are other emerging sys- tems of representation--for example, regional Chapter Structure and indigenous organizations--they have not The following four sections discuss Bolivia's yet filled the vacuum. progress toward each of the CDF principles-- 10 Long-Term, Holistic Development Framework; Achievements Country Ownership; Country-led Partnership; Today there is a new government-donor initia- and Results Orientation (see Introduction to tive that provides a compatible structure for this volume for a description of the CDF princi- implementing the four CDF principles: the ples). The discussion of each CDF principle is Bolivian Poverty Reduction Strategy, or BPRS. divided into two segments: achievements already in place and remaining challenges. Preparation of a national plan to reduce poverty is a requirement for access to an IMF­World Bank debt relief initiative tapped by Bolivia. Long-Term, Holistic (Since Bolivia receives its substantial level of aid in the form of both grants and loans, the coun- Development Framework try's balance of payment obligations have remained a concern.) The program, the second To a large extent, Bolivia had already pioneered iteration of the Heavily Indebted Poor Countries CDF-like principles before they were articulated initiative, or HIPC II, offers a tradeoff to devel- by the World Bank in 1999. The ground was oping countries with a high debt burden: debt is prepared by the government that came to power restructured if the country commits to using the in 1997, which held a nationwide dialogue resulting budget savings for poverty-reduction designed to explain and enlist support for its efforts. HIPC II is delivering over $100 million domestic reform program focused on opportu- in debt relief to Bolivia each year. nity, institutionality, equity, and dignity. The government also reached out to the interna- Designed through a national dialogue process tional donor community, inviting agencies to with participants representing the country's align their aid with its political vision. municipalities (divisions of the country that include both rural and urban areas), the funda- In early 1999 the resulting donor-government mental goal of the BPRS is to reduce poverty partnership was spelled out in the New through both economic and social policies that Relationship Framework--a made-in-Bolivia fall under four interrelated categories: Comprehensive Development Framework. Expanded employment and income opportunities Given this head start on CDF principles, Bolivia through support of productive capacity was a natural choice as one of the initial CDF pilot countries.2 Increased capacity through improved pri- mary education, preventive health care, and Even though the New Relationship Framework better housing for international aid had prepared the ground Enhanced security for low-income groups, for the CDF, seeds sown in the late 1990s did including the elderly and children not flourish: the optimism and trust engendered by national dialogue and government-donor Expanded social integration and participation partnership withered with the waning influence for marginalized groups. of key government reformers and the persist- ence of a stubborn patronage system--and The BPRS includes a set of indicators to meas- even, according to some observers, under wide- ure results, starting with three-year targets. In spread corruption. addition, it shares long-term objectives with the Millennium Development Goals, a broad set of and long-term strategic plans and to convince 11 development goals with measurable outcomes international partners to endorse them. What supported by the international aid community. has sometimes been less developed is Bolivia's ability to carry them out. The link between The BPRS serves as the fund transfer mecha- planning and implementation is often weak for nism for HIPC II. Fund allocation has several a number of reasons; several relate to the central innovative features, which have been spelled budget process. For example, budgeted funds are out in the country's Dialogue Law: enacted to sometimes held up because of larger economic assure the continuity of the dialogue process, it concerns, or because a relevant agency's budget- requires updating of the BPRS every three years. ary discipline is lax. In addition, accounting is Provisions of the law include: both late and unreliable, with no penalties for poor performance. Significant increase in resources channeled directly to the local level. An additional stumbling block to current BPRS Municipalities determine resource use based implementation includes the disconnect on local demand. between the level of effort expected from local governments and actual municipal capacity. Portion of funds earmarked for local revenue While scattered local capacity-building efforts generation. are now under way, there is still considerable Municipalities discouraged from seeking ground to be made up before municipalities can additional funds from "godfathers"--whether efficiently use funds being transferred to them local or international--outside of BPRS under decentralization measures. transfers. In addition, opposition parties, civil society Bolivia's 2002 budget mirrors BPRS priorities, organizations, and international partners fault and expenditures budgeted within the BPRS the BPRS for failing to attack large structural framework rose from 21 percent of GDP in issues contributing to poverty. They cite, for 2001 to nearly 24 percent in 2002. Between example, the need to address land redistribu- 2001 and 2006 the BPRS will require $7.3 bil- tion, political inclusion of indigenous people, lion to implement; that amount is projected to the issue of coca growers, and increased genera- be available under agreements with interna- tion of tax revenues--a layer of issues they tional aid partners. believe should be added to the current BPRS priorities of improved services in education, There are other wide-ranging reform initiatives health, and sanitation. under way in Bolivia, including programs that address issues in education, the legal system, and Critics also cite what they see as the BPRS's land ownership. All touch on strong national unrealistic projection for economic growth, set and local interests--and therefore all generate at 5 percent per year, when the average rate was heated discussion, or even conflict. 3.1 percent between 1986 and 2001. Growth for 2002 was estimated at roughly 3 percent as well, Challenges but this is now seen as overly optimistic, given There is no question that Bolivia has a highly the ripple effects of neighboring Argentina's developed capacity to formulate comprehensive economic problems. 12 A final challenge to implementing Bolivia's of "shared ownership" between the two parties, comprehensive development vision as it is complete with mutual responsibilities and expressed in the BPRS is the fundamental ques- accountabilities. The lesson is that the owner- tion of whether a development strategy can be ship principle should not be understood as a rigid consistently applied over time in a democratic absolute that dictates exactly what roles the gov- society--given shifting priorities when newly ernment and international aid entities should elected leaders come to power. For example, play, but rather as a flexible relationship. even though President Jorge Quiroga played a key role in promoting both the BPRS and the Achievements CDF when he was vice president, major ele- The dialogue leading up to the BPRS drew from ments of these plans were missing from his a broader base of Bolivian civil society than pre- strategic plan for the country when he assumed vious national strategic planning efforts. In the presidency after the incumbent president addition, BPRS preparations encouraged greater stepped down because of illness. dialogue between government ministries. Some informants argued that the push for a As noted above, a further positive development national vision (as expressed in the BPRS) that was passage of the Dialogue Law, which laid a is broadly agreed between the parties and by foundation for continuing the process. The law society may actually be incompatible with a also provided a legal framework for a social basic characteristic of multiparty democracy: watch mechanism by civil society organizations, groups with different interests and values create an effort that was spearheaded by the Catholic different visions and plans that compete openly Church. The evaluation team raises a caution- with one another. ary question, however, in regard to establishing this supplementary process to the oversight pro- In a democracy, successive governments have vided by representative democracy: Will per- the right--and even the duty--to shape and ceived deficiencies in the political system lead reshape their own vision, strategies, and poli- to other such parallel mechanisms, rather than cies. Perhaps the answer to sustaining the CDF's changes that would allow the democratic system long-term comprehensive vision is to think of to function more effectively? the principle as being contained not in a single paper, but rather as a continuing process of con- A related issue is the technocratic bias that gov- struction and reconstruction. ernmental reform efforts tend to adopt (interna- tional agencies adopt it as well) in order to protect areas of public administration from the Country Ownership political system. There are conflicting forces at work here: unless a technocratic stance is taken, Successive Bolivian governments designed and it can be difficult to prevent reforms from being managed long-term reforms before the launching appropriated and abused by the informal public of the CDF. And many reforms for which there is administration system, with its widespread strong country ownership have been supported cronyism. However, a development dialogue by Bolivia's international partners. Rather than divorced from a significant part of Bolivia's real- reforms having "country ownership" or "donor ity risks becoming a technocratic discourse of ownership," what has often developed is a sense limited relevance. The government's decentralization efforts can sionalism and decreased politicization (box 13 be seen within a context that includes the own- 2.2)--and then typically as the result of aid- ership principle: as noted earlier, the decentral- related requirements. Not surprisingly, much of ization process was launched in 1994 through the explanation for lack of progress could be the Popular Participation Law (PPL). The PPL found in the resistance of political and eco- established "municipalities" that included both nomic entities with strong vested interests. rural and urban areas, and specified that 20 per- cent of internal revenue and customs funds go to Under the 12-month term of President Jorge the municipalities, based on their population. Quiroga (2001­02), however, an Institutional While donor agencies were not consulted in Reform Program (IRP) was implemented, designing the PPL, they have made significant which included: contributions to its practical application: donors Creation of a Civil Service Superintendency work with the government to monitor progress, and they finance both technical assistance and Declaration of net assets by civil servants ongoing policy dialogue. Here, too, say evalua- Reduction in number of people on the civil tion informants, there is a sense of shared gov- service payroll ernment-donor ownership. Increased wages There is also progress in the government taking Requiring employees to reapply for positions ownership of donors' long-desired goal of improving the professionalism of civil servants. Expanding efforts beyond those already made Previously, progress had been slow, with only in regulatory bodies--for example, now such regulatory bodies as the Central Bank and including the Ministry of Agriculture--to the Customs Service showing improved profes- implement reforms. Box 2.2. Customs Service Reform: Tough Government Decisions, with Help from International Agreements Until reform efforts in the late 1990s, the level of corruption in the Bolivian Customs Service had long been illustrated by a single surprising fact: even though many staff members received no salary, it was considered highly desirable to work there. While several governments had tried to reform and modernize the institution, political pressures from their own political parties stood in the way of making the tough decisions necessary for effective restructuring. Finally, the government's desire to access International Monetary Fund (IMF) resources forced the issue: Customs Service reform was among the required conditions. In 1998, the government asked the IMF for technical assistance in shaping a study to identify steps for implementing reform. After protracted political debate, enabling legislation was drafted and enacted. One of the law's key elements is shifting the authority to hire staff to a competitively selected firm, taking the Customs Service out of the process. These reform efforts have proven successful: Current problems in the institution now reflect concerns central to the Customs Service mission--for example, developing methods to reduce contraband. 14 The IRP initiative has the strong support of the tinuing strong base of support and effective international aid community, including the implementation. Since there is no fixed World Bank. roadmap to establishing the reform process, or for defining the exact role of international agen- Challenges cies, each reform initiative must have a careful While the majority of CDF evaluation inform- strategic design: a subtle and complex mixture ants stress that the dialogue leading up to the of government-donor leadership and broader BPRS was broader than that employed in shap- participation at appropriate times--and then ing previous national strategies, they neverthe- skillful "marketing" of the overarching vision. less agree that political and social support for the BPRS is less than what had been hoped for. Evaluation informants cited the need for inter- national aid agencies to develop a greater under- One explanation, briefly mentioned above, may standing of the country's political processes and be that there are actually two types of national institutions in order to be a truly effective agent dialogue that take place in Bolivia: The first is in this process--one that has earned a level of the "fight poverty" dialogue, which is typically trust developed over time through shared efforts the focus of international aid agencies and their among government, international agencies, and plans, procedures, and processes. The second is other stakeholders. a broader dialogue rooted in structural issues; it is mediated by the Catholic Church; by Bolivia's recently established Ombudsman's Office; and Country­led Partnership by the Permanent Assembly for Human Rights. It is this latter dialogue that expresses itself in The relationship between Bolivia and its inter- social conflicts, including the violent street national partners has improved significantly clashes of recent years. The two dialogues have over the past 10 years, notes the evaluation separate agendas and priorities; the great chal- team. Among the contributing factors are lenge is to build a bridge between them. increased government participation in donor- government Consultative Group meetings The BPRS participatory process did not build (facilitated by moving the meeting to La Paz the kind of stabilizing social agreement that from Paris in 2001); the New Relationship could serve as such a bridge. In fact, note Framework (a CDF precursor), which more informants, the process had several negative closely aligned government and donor efforts; repercussions. For example, by generating high and international agencies' gradual movement expectations on a widespread scale--expecta- away from traditional donor-driven and free- tions that were not quickly met--the National standing "island" projects toward broader sec- Dialogue led to deepened cynicism and a loss of torwide approaches that support national goals. government credibility. The dialogue's first phase had to be suspended in April of 2000 as A significant step toward government-led aid the government declared a state of siege to sup- partnership was taken at the 1999 Consultative press social protests. Group (CG) meeting, where an aid manage- ment group was established. The group assigned The Bolivian experience confirms that the ulti- a government and a donor representative to mate success of reform measures rests on a con- each CG working group (the working groups 15 Box 2.3. Basket Funding Lessons from the Ombudsman's Office: It Takes Time and Money to Reap the Benefits The Bolivian Ombudsman's Office was established by law in 1997 to implement a system for addressing citizens' government-related complaints and to provide such services as education on dispute resolution through mediation and consensus-building. Since many aid agencies wanted to support the Ombudsman's Office, extensive funding discussions were required. Initially each agency wanted to decide on its own how and where to provide assis- tance--the traditional project approach. But then a suggestion from one agency representative led donors to agree on pooling resources, which would allow for a more comprehensive approach, com- plete with performance indicators and a monitoring mechanism. Although it took a year to debate disbursement procedures and secure individual agency headquarters' approval for basket funding, in the end seven bilateral donors and the United Nations Development Programme reached agreement on the concept. Despite the length of time needed to sort out the basket funding mechanism and initially high transac- tion costs, both donor agencies and the Office itself state that the benefits gained through a long-term comprehensive approach and simpler disbursement procedures were worth it. being aligned with the four pillars of the gov- agile agencies with local decisionmaking power ernment's Plan of Action for 1997­02: opportu- to provide near-term technical assistance, and nity, equity, institutionality, and dignity). The employing the skills of agencies with more rigid continuing relationships established through this procedures (which require a lengthier headquar- process formed the basis for planning the BPRS. ters' approval process) at a later project stage. Also contributing to progress in partnership is donors' ongoing shift to a broader program focus These kinds of partnerships, among donor agen- and the related use of joint financing methods, cies and between government and agencies, including common pool or "basket" funding. require a comfortable level of trust. In Bolivia, Bolivia displays considerable flexibility in con- the trust level has followed something of a roller- structing these arrangements to accommodate coaster path: initially high in the mid-1990s, a individual donor agency financing require- 1998 power shift in the government--one per- ments. One example occurs when several ceived as accompanied by a significant increase donors finance a program through one common in corruption--undermined trust. Under the channel or donor, while donors with more rigid Quiroga Presidency, trust increased. Now, in requirements provide direct support, all within 2003, the government under President Gonzalo the same program and with shared monitoring Sanchez de Lozada has the task of building trust mechanisms (box 2.3). for its own administration, as every new govern- ment must do. The same kind of donor coordination also allows imaginative use of the competitive advantages of What is evident over the long term, however, is individual donor agencies--for example, using that by building on achievements already in 16 place, the CDF partnership principle has con- evaluation team saw obvious problems with siderable potential for making government- coordination and synchronization as they made donor cooperation much more effective. field visits. Beyond "island" projects that are still being imposed by agencies on the basis of pre- conceived ideas, there are also examples of road Challenges and municipal capacity building that simply Despite progress toward government-donor ignore Bolivia's decentralization policy. This partnerships, cooperation remains scattered and retrograde approach is being carried out by some fragmented. The government must still spend agencies that are on record as supporting decen- administrative time and resources on more than tralization reforms. 850 separate programs or large projects funded by international agencies. Many of these pro- The evaluation team also saw evidence that grams will be "reallocated" under the BPRS-- international agencies continue to microman- that is, shifted to BPRS priority areas and away age projects--for example, by hiring consultants from lower-priority areas now considered over- to do tasks that could be carried out by the funded. Observers see reallocation as a true test country's public administrators. In addition, of donor agencies' willingness to adapt to the there are many parallel project and program country's national priorities. That reallocation management units. While their presence may be will take place under a new government that is warranted in a few cases, overall they under- defining its position in relation to the BPRS mine the long-term development of national could further complicate what is already a com- and local institutions, by denying them project plex process. management experience. An additional challenge for establishing coun- The local offices of international aid agencies try-led partnership is the tendency for donors to vary widely in terms of their authority to reallo- participate in one level of national dialogue-- cate already approved assistance and administer one focused on reducing poverty through programs at the local level: some have consider- improved services such as education and health able autonomy, while others are constrained by a care--while avoiding the second level of dia- highly centralized structure (box 2.4). Agencies logue, one that involves the more controversial in the latter category are often prevented from structural issues of land ownership and the "harmonizing" their procedures to make them exclusion of indigenous groups. Both govern- more compatible with those of the government. ment and donor stakeholders point out that (Donors are helping the Bolivian government sticking to the first, more limited, view threat- strengthen its own procedures so they offer a ens to divorce donors from Bolivia's political credible alternative to donor procedures.) reality. They suggest it may be necessary for donors to widen their network of contacts to There are other occasional tensions between include leaders of social movements, parliamen- bilateral and multilateral agencies; many can tarians, politicians from opposition parties, and be attributed to the dominant role played by the police in order to gain a more representative the World Bank and the IMF in the country. view of the issues facing the country. Sometimes the bilaterals band together in taking a common stand when they disagree Furthermore, while many donors say they are with the Bank or IMF, in order to gain suffi- prepared to adapt to national priorities, the cient counterweight. Other tensions arise because bilateral donors or a borrower-lender relationship. It is possible, 17 often have limited technical capacity to partici- however, to build asymmetrical relationships pate in policy dialogue or sectoral analytical that are both productive and reflective of CDF work. One solution is for bilateral donors to principles. One way is to make donor-govern- become more selective, concentrating their ment cooperation more efficient by reducing efforts on a few areas where they can make an burdensome transaction costs imposed by intellectual as well as a financial contribution. donor procedures--for example, by improving Doing so could also help "un-bunch" popular procedure harmonization and donor coordina- sectors where donors tend to concentrate. In tion. Another is to recognize that wise, bold, Bolivia, these include decentralization, institu- and respectful behavior is the best path to tional reform, and the Ombudsman's Office. sound policies. Finally, implementing the Country-led Partner- ship principle must balance the interests of all Results Orientation parties--not an easy task when international aid partnerships are inherently asymmetrical, The CDF evaluation informants agree that both based as they are on either a donor-beneficiary the government and international aid partners Box 2.4. The World Bank and "Planet Bolivia" The following example illustrates the difficulties international agencies can encounter in adapting to CDF principles. In 1999 the World Bank decided to conduct a pilot experiment and decentralize its operations to Bolivia, in order to better support country partners and priorities. A country director was posted in Bolivia (instead of a resident representative); the office in La Paz was reorganized to corre- spond to the four pillars; appropriately qualified professionals were appointed to boost the office's capabilities considerably; and extensive authority was delegated to the leaders of each pillar (instead of being centralized in sector units in Washington). This arrangement was so different from the Bank's usually centralized procedures that the Bolivia program was referred to internally as "Planet Bolivia." Bank staff engaged in an intensive participatory process with government counterparts over how to best serve country needs. This resulted in a proposal to, among other steps, decentralize authority so that the country office could approve individual operations within an agreed three-year program framework and budget. Although this proposal was enthusiastically supported by Bank staff and the Bolivian government, when it was discussed by the Bank's Board of Executive Directors (in 2001), they disagreed, arguing that such an arrangement would cede too much of the Board's stewardship and approval authority to the field. Operations were subsequently recentralized. A subregional office was opened in Lima, where the country director now resides. The staff was reduced in La Paz and the orga- nizational structure based on "pillars" was dismantled. This recentralization evoked a strong reaction from the largest bilateral agencies in Bolivia. They com- plained in writing in early 2002 that the Bank had "undermined the (CDF) pilot scheme, while a number of bilateral agencies have on the contrary decentralized their field (La Paz) operations to sup- port the CDF. It is ultimately ironical and unhelpful that the Bank itself is now no longer supporting this key requirement of the CDF." 18 are now more focused on measuring develop- would have to be added to the BPRS in order to ment results. On the government's side, results- gain this broader understanding. oriented management as well as disbursement based on achieving targets are now used at both In addition, there is a need to coordinate the the central and the local level. In addition, the BPRS social watch functions defined in the BPRS contains specific measurable goals. Dialogue Law with those that have been estab- According to Bolivia's Interagency Monitoring lished through other grassroots participation and and Evaluation Council, nearly 90 percent of decentralization efforts. Also needing coordina- the 81 short-term activities spelled out in the tion are the country's various methods for audits, government's 2001 BPRS Plan of Action are inventories, and financial accounting, which complete--although it is too soon to evaluate now overlap and result in duplication of effort. their impact. And, while Bolivia's central government has the Local stakeholders have taken an active role in skills needed to gather as well as process and defining intermediate indicators as well as end interpret data, these skills are not matched at results for the BPRS. And the Catholic Church the municipal level--where, in any event, it is is helping structure the social watch mechanism no easy task to devise performance indicators. authorized in the Dialogue Law as a way to At the same time, there are tendencies among involve civil society in monitoring the BPRS. In donors to make monitoring and evaluation sys- addition, the National Compensation Policy tems too complex, which results in less efficient establishes fiscal restructuring and efficiency data collection, processing, and feedback. A bal- goals at the municipal level, with incentives for ance needs to be struck between complex sys- municipal governments to meet them. tems and actual capacity. Some sectors, such as health, have been per- While progress toward implementing the CDF formance-oriented for years, either because of results-orientation principle has certainly been their scientific nature or through the influence made in Bolivia, weaknesses reveal that stake- of institutions such as the World Health holder incentives do not consistently support Organization. And, since 1982, the government the goal. For example, international agency has had access to the analytical skills of the respondents identified their own "disbursement Economic and Social Policy Analysis Unit, fever" as a barrier, not only to moving toward which, together with the country's National measurable results but to other CDF principles Statistics Institute, is widely respected for its as well. Disbursement is an integral part of monitoring and analytical skills. agency culture, where an official's success is fre- quently measured by their ability to spend budget allocations rather than by ability to Challenges achieve results. The strong focus of the BPRS on such "soft poverty" indicators as education and health Pressure to disburse funds is also a strong moti- limits understanding of the larger structural vator for Bolivia's government. The country's determinants of poverty. As noted above, these level of aid dependence means aid is tied not include lack of a political voice and restricted simply to accessing know-how, but to balanc- access to land and employment. Indicators for ing macroeconomic accounts: a low level of vulnerability and social and political exclusion execution lowers Bolivia's level of economic activity. Strengthening the results-orientation ditional micromanagement, but also the "macro 19 approach will therefore require a comprehen- management" that the broad focus of the CDF sive reexamination of the incentives--or and BPRS invite. rather the current disincentives--contained in the Bolivian environment. The challenges for both government and donors are enormous. For Bolivia, the main challenge is to extend national ownership of CDF principles Central Issues to all social stakeholders and important political parties in the country--and to do so in a climate Depending on an observer's perspective, of social unrest where large groups of people Bolivia's progress toward achieving CDF princi- have historically been excluded from the politi- ples can be seen as either a glass half empty or a cal process. glass half full. For example, a number of the recent institutional and macroeconomic reforms For international aid agencies, application of are well under way, including the Institutional CDF principles in Bolivia has reached a water- Reform Program and the 2001 Finance Law. shed: since they are now well internalized by a However, other reforms launched several years critical mass of local representatives, continued ago--such as those in education and decentral- progress requires changes at agency headquar- ization--still have a long way to go, perhaps ters--and, in some cases, reengineering of oper- indicating a lack of institutional and political ating methods. absorption capacity. But even if the challenges are daunting, the A further concern is that to some extent, the results so far show that progress is possible. CDF casts international agencies as stakehold- Given the vulnerability of the disadvantaged in ers in Bolivian arenas that are much broader Bolivia, and of the country itself, these chal- than they used to be. Local representatives and lenges need to be tackled--while the glass may agency headquarters staff must be on guard if indeed be half full, it nevertheless needs to offer they are to avoid lapsing into not only the tra- more to Bolivia's citizens. Curt Carnemark C H A P T E R T H R E E Burkina Faso 21 Methodology in analyzing the main themes and sectors dealt with in the case study. These were grouped in three categories: institutional issues; social sec- The Burkina Faso case study team was led by tors; and growth, employment, and income gen- Dominique Lallement, manager, Energy Sector eration. Altogether, roughly 200 people were Management Assistance Program (ESMAP). interviewed for the case study (see Annex 2 of Team members included Della McMillan, soci- the case study report, "Evaluation of the ologist, consultant; Kyran O'Sullivan, senior Comprehensive Development Framework: energy specialist, World Bank Energy and Water Burkina Faso Case Study," for a list). The team Department; Patrick Plane, macro-economist, also conducted an extensive literature review, Director of the Development Studies Center, including an analysis of government and donor University of Clermont-Ferrand, consultant assessments of Burkina Faso's experience in (France); and Kimseyinga Savadogo, macro- shaping its current poverty reduction strategy. economist, former dean of the Economics Depart-ment, University of Ouagadougou, con- In addition, the study team employed a random sultant. In addition, Jean-Pierre Ouedraogo, survey of a cross-section of stakeholders in economist, independent consultant, organized the four districts of Ouhigouya, Kaya, Boromo, focus group discussions with business owners and Tienkodogo. The districts were chosen to and managers. Begnadeyi Claude Bationo, con- represent a range of poverty levels and ecologi- sultant for CIDA, contributed a report on the cal zones. monitoring system for the country's poverty reduction strategy paper, the Cadre Strategique de Lutte Contre La Pauvrete, or CSLP. Eight Burkina Faso: A Proud University of Ouagadougou graduate students, supervised by Kimseyinga Savadogo, adminis- Development History-- tered a survey to a cross-section of development and Daunting Challenges stakeholders in four provinces. The team conducted fieldwork during two visits Unique Country Characteristics to the country (in April and May of 2002), using The implementation of CDF principles in a single reference framework in interviews and Burkina Faso takes place in a complex environ- in focus group meetings, the latter with six cat- ment shaped by a long history of participatory egories of identified stakeholders: central gov- traditions and long-range development strategies ernment, technical ministries, donors, local anchored in grassroots consultations, including institutions, civil society, and the general popu- through the colonial period. The governments lation. The reference framework was also used that followed independence (the first lasted 14 22 years) built on the reforms of predecessors; the Economic and Social Indicators administration of the second post-independence Supported by the international finance institu- government (1983­87), known as the revolu- tions, Burkina Faso launched a series of eco- tionary era, included a CDF precursor in the form nomic reforms in the 1990s. These included of a comprehensive development vision, spelled more prudent monetary and budget manage- out in Popular Development Plans for regional, ment policies as well as devaluation of the coun- provincial, and national levels. try's currency. The reforms worked; with an average 1994­99 growth rate of 5 percent, Nonetheless, the legacy of the revolutionary era Burkina Faso's economy grew faster than the 3 is mixed. The government had established local percent average for the Sub-Saharan region. citizen committees with the primary goal of rooting out corruption, and a secondary role of Despite this overarching success, however, not mobilizing popular support for selected national all economic indicators were positive in the strategies. In some instances, these committees 1990s. Among the negatives were a 10 percent became instruments of repression. To date, the balance of payments deficit; high production government has had difficulty engaging civil costs that undermined export competitiveness, society in national policy­related dialogue, administrative barriers that hampered domestic including establishing a broad understanding of and foreign economic investment, and the slow the country's development goals, such as those pace of privatization. reflected in the poverty reduction strategy put in place in 2000, the Cadre Strategique de Although the privatization effort was launched Lutte Contre la Pauvrete (CSLP). What has in 1991, more than a decade later roughly two- not been tamped down, however, is community thirds of public enterprises were still operating participation in local development, which con- under state ownership. The privatization tinues to be one of the country's strengths--as process, as in neighboring countries, has been illustrated by more than 200 NGOs and roughly undermined by a number of factors, including 14,000 associations. resistance from vested interests among both labor and management. Starting in the late 1980s, Burkina Faso devel- oped long-term poverty reduction plans in four In addition, Burkina Faso's big-picture economic sectors: education, health, water and sanitation, growth did not translate into progress in reduc- and rural development. In the 1990s, the plans ing poverty. The 1998 Poverty Survey showed a were supported by two poverty surveys--one slight increase since 1993, with 45.3 percent of done in 1994, the second in 1998. Both efforts the population below the poverty line, compared made important contributions to the country's with 1993's 44.5 percent. In addition, significant current poverty reduction strategy, and enabled rural-to-urban migration was associated with a it to articulate a comprehensive vision for long- sharp growth in urban poverty, which rose by 6 term development and poverty reduction well in percentage points over the same five-year time- advance of the 1999 launch of the CDF. For frame, from 10.4 percent to 16.5 percent. There example, the government shared its vision with are also continuing regional differences in aid donors at the 1995 government-donor poverty levels, with areas lacking natural Consultative Group meeting. resources typically showing higher rates. Burkina Faso ranks lower than most Sub- dued. At the same time, Burkina Faso was the 23 Saharan countries in social indicators. Infant first country in Africa to have a privately and early childhood mortality levels are 104 per operated radio outlet, and there are 10 news- 1,000 and 219 per 1,000, respectively, versus 91 papers, which are quite vocal. While the and 151 for the region. The country's literacy University of Ouagadougou's technical stan- rate is 25 percent (less than 6 percent for dards seem quite good, the government does women). While the school enrollment rate of not fully utilize its resources as a potential 43 percent is low for the region, it is neverthe- source of fresh policy approaches. less a marked improvement over the country's 5 percent level in 1983. (See box 3.1 for these and Burkina Faso's government is the major other social indicators compared to Sub- employer for nonagricultural sectors. Divestiture Saharan benchmarks.) and devolution of economic activities to the pri- vate sector should therefore remain government Need for Private Sector Growth priorities as a way to broaden the participation With the private sector contribution to GDP of the population in the country's economic and still nascent, the role of Burkina's central social development. government remains paramount. The state is the major source of new ideas for the country's Currently the government relies on the rural socioeconomic life, and public debate is sub- population for economic growth: farmers have Box 3.1. Burkina Faso at a Glance POVERTY and SOCIAL Sub- Development diamond* Burkina Saharan Low- Faso Africa income Life expectancy 2001 Population, mid-year (millions) 11.6 674 2,511 GNI per capita (Atlas method, US$) 210 470 430 GNI (Atlas method, US$ billions) 2.4 317 1,069 GNI Gross per primary capita enrollment Most recent estimate (latest year available, 1995-01) Poverty (% of population below national poverty line) ­ ­ ­ Access to improved water source Urban population (% of total population) 17 32 31 Burkina Faso Life expectancy at birth (years) 44 47 59 Lower-middle-income group Infant mortality (per 1,000 live births) 104 91 76 Child malnutrition (% of children under 5) 34 ­ ­ Access to an improved water source (% of population) ­ 55 76 Illiteracy (% of population age 15+) 75 37 37 Gross primary enrollment (% of school-age population) 43 78 96 Male 51 85 103 Female 35 72 88 24 invested heavily in land and water conserva- Burkina Faso's human resources are both its tion as well as in land development networks. greatest asset and a liability. When given the However, internal population migrations opportunity, entrepreneurial and technical com- undermine land-tenure security. The issue's petencies among the population are evident. At complexity is illustrated by the fact that even the same time, the country's 2 percent annual though achieving land tenure security has been population growth rate threatens development debated since the mid-1980s, little progress has (this contrasts with a population growth rate of been made. In addition, the significant rural-to- 2.5 percent yearly between 1997 and 2001 for urban migration (between 12 and 15 percent a Sub-Saharan Africa). There has been a regres- year) contributes to the lack of urban employ- sion in the reach of family planning programs ment opportunities for unskilled workers, turn- since the 1980s; only one of 10 women who ing many urban immigrants into the come to consult health clinics receive family unemployed homeless. Among the most vul- planning services. Birth control devices are nerable are adolescents and young adults, available but expensive; when they are used, making job-training programs a likely future wives often do so at some risk by making the priority for the government. decision without their husbands' consent. Unplanned teen pregnancy is also a concern. Aid Characteristics Aid levels as a proportion of GDP in Burkina Burkina Faso does, however, seem to be making Faso are nearly four times higher than the aver- progress in managing the AIDS epidemic. age for Sub-Saharan African countries--15.5 Evidence is found in the commitment of public percent of GDP in 1998, compared with the 4.1 resources to diagnosis and care as well as to percent regional average. Aid transfers to social support--the latter provided by the coun- Burkina Faso make up about half of the coun- try's large number of associations as well as local try's public expenditures. While project aid still community support. At the time of the evalua- predominates, its share of all aid has been tion team's work in the country, the rate of decreasing--between 1997 and 1999, from 87.7 reported HIV cases was estimated to be declin- ing, although AIDS education for adolescents percent to 79.6 percent. and young adults remains inadequate. Reducing Aid Dependency Through Economic Diversification Capacity Building Key There has been some progress toward increased The study team concluded that human capacity economic diversification, an important factor in building should therefore be a top priority of the achieving poverty reduction and reducing aid country's poverty reduction strategy, the CSLP. dependency. Agriculture has diversified to some (The CSLP is Burkina Faso's Poverty Reduction extent with the expansion of fresh vegetables Strategy Paper, prepared to access debt relief sold to coastal countries (cotton and cereals are under the World Bank­IMF HIPC initiative. As the traditional agricultural exports). However, such, it is also the embodiment of CDF princi- given scarce land resources, a better approach to ples in the country, in keeping with the HIPC economic diversification may be to export serv- guidelines.) Among the most pressing capacity- ices. By 1998, services already made up a greater building issues is to aggressively address the high share of the country's GDP than agriculture: 40 rate of adult illiteracy, which is roughly 75 per- percent compared with 32 percent. cent. At present, parents resist sending their children to primary school because the cost of Chapter Structure 25 schooling is seen as high relative to the oppor- The following four sections discuss Burkina tunity cost of losing their labor for cultivation. Faso's progress toward each of the CDF princi- Post­primary school professional training is vir- ples--Long-Term, Holistic Development Frame- tually non-existent in Burkina Faso, and univer- work; Country Ownership; Country-led Part- sity-level technical education is extremely nership; and Results Orientation (see Introduction limited but improving, even though employers to this volume for a description of the CDF prin- are looking for technical skills. ciples). The discussion of each CDF principle is divided into two segments: achievements already These, then, are the mixed social and economic in place and remaining challenges. indicators facing Burkina Faso as it works to implement the CDF principles embedded in its poverty reduction strategy, and to reach its par- Long-Term, Holistic allel target, achieving the Millennium Development Goals. On the plus side is that the Development Framework country has never had the total economic, polit- ical, and social breakdowns that led to entire Burkina Faso has a proud 40-year tradition of new national identities in several CDF pilot articulating an interconnected series of long- countries, such as Uganda and Vietnam. term development goals. There are two docu- (Burkina Faso was not a CDF pilot; it was ments that represent the tradition today: the chosen for the CDF evaluation because of its more recent--the Cadre Strategique de Lutte early success in accessing debt relief under HIPC contre La Pauvrete (CSLP), developed in by rather quickly drafting a poverty reduction 2000--builds on the first, the Lettre d'Intention strategy, and because of its above-benchmark de Politique de Developpement Humain level of aid.) And even though each Burkinabe Durable (LIPDHD), developed in 1995. The government since independence has formulated CSLP sets out the country's poverty reduction its own approach to economic and political strategy. The development visions set out in the development, under most administrations a two documents are fairly consistent. common thread has been citizen participation in identifying the country's long-term develop- Achievements ment vision. The strength of the CSLP is that it reflects the broad aspirations of the population for improved At the same time, Burkina Faso is faced with social services and income growth, and attaches daunting economic and social problems, from measurable goals to achieving those aspirations. the slow process of privatization and private In addition, it contains a workable process for sector growth to below-benchmark social indi- achieving the goals. The CSLP has gained legit- cators. However, if the country continues to imacy by building on both solid economic per- improve its economic management and sustains formance and statistical analyses in priority its commitment to poverty reduction, there is a social sectors--for example, on the two poverty good probability that it can implement the surveys done in 1994 and 1998. CSLP and, in the process, make solid progress toward achieving the Millennium Development Also critical to how well the CSLP will work in Goals, a parallel target of the government. transforming the country's long-term develop- 26 ment vision into on-the-ground programs is its Given these agricultural realities, the govern- recognition of the need for improved public ment should be solidly behind the CSLP's well- expenditure management. In 2001, the govern- articulated objectives for developing small- and ment adopted the Public Expenditure Improve- medium-size businesses. However, in case study ment Plan to address shortcomings in budget roundtable discussions, private sector represen- formulation, execution, monitoring, and audit- tatives complained about the government's dis- ing; a number of its recommendations are now tinct lack of urgency in improving the climate being implemented. for private sector development--citing in par- ticular the failure to address tax issues, credit These recommendations are also reflected in the availability, and administrative burdens. country's 2003­05 medium-term expenditure framework for the central government budget, Another CSLP weakness is its failure to identify which is the main instrument for transforming potentially synergistic linkages between sectors. CSLP goals into concrete reality. The increase For example, a linkage between the education in national expenditures for social sectors in the and economic development sectors could 2002 budget reflects this link. address the fact that while the education cur- riculum emphasizes non-technical subjects at Burkina Faso has also improved its ability to the secondary level, the economy has a growing control budget expenditures by creating several demand for technical skills. new oversight institutions; the challenge is to ensure that they function effectively. Here the And while the CSLP is gaining recognition and National Assembly can provide support by exer- legitimacy, it is still unclear whether it will pro- cising its new oversight authority, granted in the vide a strong enough framework to engage all 2002 budget execution laws. Taken together, development stakeholders in achieving the these measures for improving Burkina Faso's country's long-term development vision. Even ability to manage public expenditures enhance though the country has a growing number of the CSLP's credibility with development stake- political, intellectual, and entrepreneurial lead- holders. For donors planning to initiate or ers, many of the case study team's key inform- increase direct budgetary support, such fiduciary ants seemed hesitant to identify gaps in the safeguards are critical. government's long-term vision. By capitalizing on the country's university-related think tanks Challenges and the experience of the private sector, the The major weakness of the CSLP is that it government could strengthen public debate on depends too heavily on an agriculture sector Burkina Faso's development vision. characterized by uncertain rainfall for the coun- try's economic growth. The threat to the econ- omy was seen in the first two years of CSLP Country Ownership implementation, when a drought reduced cereal production (a traditional export) by 15 percent: While Burkina Faso has fashioned a number of GDP grew by only 2 percent, far short of the long-term development plans for itself over its country's targeted 7 percent. In addition, 40-plus years of post-colonial history, several exports of the most important crop, cotton, are stand out as having been shaped by a particu- adversely affected by protectionist measures larly broad participatory process. Of most rele- adopted by other cotton-producing countries. vance in the country today is the Lettre d'Intention de Politique de Developpment with the goal of broadening the plan's topical 27 Humain Durable (LIPDHD), developed in 1995 and geographical coverage. The workshops entirely by Burkinabe institutions; it is often focused on human resource development; rural referred to as the predecessor of the country's development, economic competitiveness, and poverty reduction strategy, the CSLP. good governance--all in the context of their link to poverty reduction. In addition, long-term sectoral plans for the rural sector (the Programme National de Gestion des There has also been progress toward implement- Terroirs, or PNGT) as well as 10-year plans for ing the decentralization of government adminis- health, education, and water development were trative services as laid out in the CSLP (which created with strong participation from civil soci- in turn is anchored in the democratic principles ety, government, and various donors that helped spelled out in the country's 1991 constitution). finance pilot studies (including the French Scheduled for full implementation in 2003, the government, GTZ [Deutsche Gesellschaft process began in 1993 with the creation of a fur Technische Zusammenarbeit], and IDA National Commission on Decentralization. The [International Development Association] for the first municipal elections took place in 33 munic- PNGT; the OMS/WHO [the World Health ipalities in 1995, and a second round was held in Organization], the Swiss, and IDA for the 49 municipalities in 2000. By the end of 2003, 2000­10 health strategy; and Danida and the Burkina Faso is scheduled to have 350 fully Netherlands for the water strategy). Overall, the established rural and urban municipalities, with development of goals and sector strategies has its 45 provinces grouped into 13 regions. traditionally been country-led, with regular and sustained stakeholder participation as well as sus- tained political commitment. Challenges Carrying out the decentralization strategy has The preparation of the country's poverty reduc- been the greatest challenge to CSLP implementa- tion strategy, the CSLP, was not entirely in this tion. Designed to distribute financial resources mold, however. Because Burkina Faso was under more equitably--and thereby give local popula- pressure to complete the CSLP in order to tions the opportunity to shape their own develop- access HIPC resources, the document was pre- ment initiatives--decentralization effectiveness pared in seven months, leaving little time to depends on municipal administrative and politi- work through a bottom-up process for consult- cal capacity. The case study team's assessment is ing local communities about its content. In that such capacity varies greatly, from the tradi- addition, the chairman of the National tional mayors and municipal counselors to newer, Assembly complained that the Assembly was better-trained municipal secretaries. Ensuring that marginalized from the process because discus- local administrative structures are able to design, sion was limited and rushed. budget for, and implement development programs will require careful government monitoring. Nevertheless, several consultative sessions did take place, including two regional focus groups In general, there needs to be more active that generated feedback on the draft document. engagement of Burkina Faso's political enti- In addition, CSLP consultations continued after ties--high-level government officials, the par- the document was finalized: four participatory liament, and political parties--in revising and workshops were conducted at the regional level implementing the CSLP. The government also 28 needs to take the lead in promoting broader- Finally, the government needs to de-link the based discussion of the overall CSLP strategy, as CSLP from sector ministries that receive HIPC well as potential synergies among specific sec- resources (primarily health and education). tors. Those involved in broader discussion Until that happens, the CSLP will not be should include the private sector and the NGOs broadly owned, and will not be able to serve as (which can also be seen as representing the pop- Burkina Faso's development framework over the ulation at large). Despite the CSLP's strategy for long term--that is, beyond the duration of the reducing the state's role in the economy, there HIPC initiative. was little private sector involvement in its ini- tial design, nor is there now during its imple- mentation. Reaching out to involve the private Country-led Partnership sector in the CSLP is critical, since it plays a key role in raising incomes and living standards The CSLP makes clear that it is the govern- by expanding nonagricultural employment. ment's responsibility to take the lead in coordi- Although some government entities have pur- nating donor activities and ensuring that aid fits sued more active dialogue with private enter- within the country's strategic framework. To prise, business owners' often perceive that rather make the process work, it is also the govern- than being helpful, the government has promul- ment's responsibility to define implementation gated inefficient regulations that create barriers instruments and to create a consultation process to success. for evaluating poverty reduction programs. In rural areas, despite the lack of awareness of Burkina Faso's participatory tradition has the CSLP on the part of village leaders and already provided a model for more effective cooperatives interviewed by the study team, the donor coordination: through five meetings team found a strong commitment in these pop- held during the EU-coordinated Condition- ulations to their own development, along with ality Test Exercise under the 1997­00 Special optimism that they can move away from aid Program for Assistance to Africa, the govern- dependency through their own cooperative ment provided donors with the financial infor- efforts. The anticipated implementation of mation they needed in order to understand the administrative decentralization reinforced this implementation process and to work on stan- optimism. To be successful, however, decentral- dardizing ("harmonizing") their disbursement ization will require: criteria. In response, donors agreed to support the government's economic management pro- Strengthening local capacity in ways that gram and to work on harmonizing their dis- meet local needs bursement procedures. Improving flow of funds tracking and moni- toring Achievements More timely arrival of funds that support In general, donors working in Burkina Faso rec- local schools and other government-sup- ognize the preeminent role of the CSLP in ported entities defining the country's development framework, and therefore as a process through which they Improving employment conditions for local can and should work together. By linking the staff, especially in remote hardship areas. CSLP to the country's medium-term expendi- ture framework, the government has provided a agement Program and chaired by provincial 29 level of fiscal security that allows donors to governors) have in some provinces proved to be increase their share of direct budget support. In effective vehicles for coordinating projects such addition, donors that are still providing project as water and small business development at the aid are increasingly aligning that support with local level. the CSLP. While some tensions remain regarding the gov- Challenges ernment's ability to implement sound economic While aid to Burkina Faso is generally aligned management policies, the donor-government with the CSLP, there is still room for improve- relationship is increasingly marked by greater ment in both donor-government partnerships understanding and coordination. This is partic- and donor coordination. One divisive factor is ularly evident at the sector level. An example is the lack of donor agreement on the country's agreement on rural potable water development, ability to manage direct budget support--those achieved through a series of donor-supported that do not think the country has adequate workshops that the government then built on to capacity prefer to continue to award aid through create standard water project approaches. the project approach. This disagreement has resulted in the development of parallel In addition, the CSLP has been notably success- processes--the European Union's on one hand, ful in directing HIPC resources toward social and the World Bank's on the other (box 3.2). So service sectors at the local level. Evidence is far, the government has been reluctant to dic- found in increased levels of health center sup- tate a single process. Although doing so would plies and the hiring of 800 primary school staff. reduce aid transaction costs, the government's In addition, local Development Committees decision may reflect a past perceived advantage, (created under a National Village Land Man- when different donor approaches resulted in Box 3.2. Donors to World Bank: "Know Thyself" A majority of donors interviewed by the evaluation team suggested that the World Bank should give more thought to the important role it plays in setting the tone for relations among donors in Burkina Faso. Particularly in the past, they noted, Bank officers would often deal directly with central govern- ment officials in shaping new policies, and then announce them after the fact to other donors work- ing in the country. While acknowledging the weight of the Bank's loans and direct technical assistance, those interviewed expressed the view that this behavior did not reflect the CDF country- led partnership principle. At the same time, however, most donors interviewed expressed respect for the hard work and compe- tence of World Bank staff based in the country--as well as their commitment to stepping back from leading donor-coordination entities they had helped create. In addition, those interviewed echoed a suggestion made by an earlier CSLP evaluation carried out by Danida: give local Bank staff more lati- tude in supervising Bank loans and financial monitoring systems. They also suggested that the Bank should consider expanding its in-country Burkina Faso staff, in order to give the Bank stronger repre- sentation in national and regional coordinating groups organized by sector theme. 30 greater financial resources for the country-- decade and acknowledging the country's talent even though it may no longer be the case. and potential. And for both government and donors, mutual performance assessment should The government could do a better job of coor- be strengthened within an environment of dinating supervision and evaluation missions to shared respect. the various governmental, NGO, multilateral, and bilateral CSLP partners. The current number of separate missions results in high Results Orientation transaction costs, noted several stakeholders interviewed for the evaluation. Frustrated by Monitoring progress toward CDF principles as these costs, some donors have decreased their embodied in the CSLP poses a significant chal- field presence and increased the proportion of lenge for Burkina Faso--as the first annual funds they administer through international progress report on the country's poverty reduc- NGOs or through decentralized projects tion strategy frankly acknowledged. After awarded to NGOs, village associations, or summarizing the ambitious goals (identify municipal governments. The problem is that inputs, measure progress toward implemen- these programs do not necessarily fit within the tation, capture immediate outcomes, assess CSLP framework. longer-term impact), the report observed: "This is a vast undertaking that calls for the mobiliza- One way for the government to reduce transac- tion of substantial resources and multidiscipli- tion costs would be to encourage donors to nary expertise." organize a greater number of joint government- multidonor missions. Although Burkina Faso had built a solid founda- tion for gathering data related to the country's In case study roundtable discussions, several rep- poverty reduction efforts before the CDF resentatives from the private sector suggested launch, the progress report goes on to note that, that private-sector­government consultations in general, monitoring and analysis of results on need to be transformed into true forums where the ground remain weak. private sector input is applied to economic policy decisions--as opposed to their current format, which is often narrowly focused on Achievements defending the interests of sectoral groups, or Starting nearly a decade ago, Burkina Faso even of an individual business. The result has crafted a number of poverty reduction initia- been short-term fixes rather than sustainable tives that included a monitoring element. economic policies. Among them: Surveys of household living conditions con- Finally, both government and donors need to ducted in 1994 and 1998, as well as a 1998 take steps to enhance trust. For the government, qualitative study that captured the percep- this should include making information more tions of the poor in identifying the causes of accessible to its donor partners and strengthen- poverty, in support of long-term sector devel- ing its accounting systems to discourage diver- opment plans sion of funds. For donors, it means giving Burkina Faso more credit for its considerable Monitoring and evaluation provisions in pro- development achievements over the past poor strategies for basic education and health A series of performance indicators for key The overarching goal of macroeconomic stabil- 31 public sector activities contained in the ity for 2000­02 targeted real GDP growth at Conditionality Test Exercise conducted an ambitious 7 percent annually, inflation at under the 1997 European Union­coordinated less than 3 percent, and a current account Special Program for Assistance to Africa. deficit of 13 percent of GDP. In addition, the budget process improvements The primary goal for cotton sector reform was contained in the CSLP provide for better inte- to end the SOFITEX monopoly. Although gration of government and donor efforts in the monopoly was ended in late 2001, the monitoring the effectiveness of assistance. The anticipated outcome, increased employment government and donors already cooperate in of the poor, has been slow to materialize. implementing the country's medium-term Increased private sector investment may be a expenditure framework, which plays an impor- better measure. tant role in grounding the broad range of CSLP funding initiatives in budget realities. Even though opening up isolated areas through rural road construction is seen as a CSLP pri- Finally, the government has established three ority, the program is not reflected in the committees to monitor CSLP progress: the CSLP monitoring framework. Ministerial Commission for Supervision of the CSLP; the Inter-Ministerial Technical Com- World Bank assistance is expected to increase the rate of privatization of telecom- mittee; and the Sectoral Monitoring Group. munications, water, and energy and of private The committees respectively focus on CSLP sector development. While progress in these decisionmaking, technical coordination, and areas is easily measured, the plan should also sectoral implementation. take account of poverty impacts. Challenges Although the CSLP does define a number of While Burkina Faso has created a number of indicators to monitor the impact of development documents that clearly identify improved budget formulation and execution which programs are of the highest priority, under public expenditure management, results taken as a whole the CSLP is unfortunately not have been ambiguous--raising questions one of them. By making the CSLP extremely about whether the monitoring plan for these comprehensive, the government diminished its policies is appropriate for measuring antici- usefulness as an indicator of which programs are pated impacts. at the top of the agenda. In addition, although a number of indicators are defined, the CSLP The weakness of the CSLP in monitoring and itself as well as the first CSLP Progress Report analyzing results on the ground undermines the rarely mention the methods to be used for mon- country's ability to understand the determinants itoring and evaluation. of poverty and to make causal linkages between inputs and outcomes. Burkina Faso should The 2000­02 development documents and those improve the culture and incentives in sector min- parts of the CSLP that do make clear judgments istries to generate and use results information at cite these priority programs along with their the operational level. In addition, the country accompanying monitoring challenges: could tap the potential of local communities to 32 Box 3.3. Achieving the Millennium Development Goals: A Probability Scorecard The government of Burkina Faso viewed the CDF evaluation as an opportunity to identify ways to strengthen local capacity in implementing the country's poverty reduction strategy--the CSLP--in order to enhance the likelihood of achieving the broad-based Millennium Development Goals (MDGs), adopted by much of the international development community. After reviewing the present level of internalization of CDF principles in the CSLP, the study team con- cluded that if Burkina Faso sustains its current commitment to poverty reduction and continues to improve its economic management, there is a good probability of successful CSLP implementation-- and thus of making solid progress toward achieving the MDGs. While reaching the MDGs may take longer than the 2015 target date, the foundation for Burkina Faso to do so in a somewhat longer timeframe is sound. The study team created a simple scorecard (using a rating scale of 1 to 5, with 1 representing low probability, and 5 high probability) to illustrate the team's judgment of the extent of CSLP internalization for each CDF principle. The table below presents the results of the team's assessment. Performance Rating of CDF Principles in Their Conception, Implementation to Date, and Likely Sustainability: CDF principle Conception Implementation to date Likely sustainability Long-Term, Holistic Vision 4 3 2 Country Ownership 5 4 5 Country-led Partnership 3 3 4 Results Orientation 2 3 3 monitor public expenditures, especially for the national monitoring system. Finally, a PMES improving services to the poor. could promote collaboration and coordination among civil society organizations, policymakers, Poverty Monitoring and and the private sector. Evaluation Strategy Implementing a national Poverty Monitoring and Evaluation Strategy (PMES) would provide The Way Forward a framework for providing results-measurement guidance to sectoral management information Burkina Faso is on a positive development path, systems and to decentralized government capitalizing on previous institutional and eco- administration. In addition, a PMES could rec- nomic management experience, strengthening ognize the important role of communities and its institutions, and building up its human capi- NGOs, both as users and as providers of devel- tal. Its steady efforts to improve its economic opment information, by making them part of management, adjust to the global market econ- omy, and move toward democracy have brought ing instruments is largely donor driven, it also 33 a measure of political stability and encouraged a reveals that several potential financing sources flow of foreign development assistance. are left out--for example, private sector investors, local financing institutions, and Overall, the country's poverty reduction strategy NGOs, as well as the financial and in-kind con- as contained in the CSLP provides a framework tributions of the general population. A system- for domestic development policy and for inter- atic analysis of the financing instruments nal and external resource mobilization. And needed to achieve CSLP goals could mobilize a although Burkina Faso was not a CDF pilot larger array of stakeholders. country, the evaluation shows that CDF princi- ples are embedded in the country's sociological Finally, human capacity building should be one of and cultural values--and are spreading to the the CSLP's top priorities, not only to improve political arena as well. Therefore, the CSLP has social indicators, but also to endow people with become an opportunity to advance CDF princi- productive capacity. The liability of the high rate ples in Burkina Faso. of adult illiteracy will continue to constrain eco- nomic and social transformation unless it is However, the case study does lead to the ques- aggressively addressed. And for future generations, tion of whether there is a missing CDF principle the challenge is to adapt the content of education (and thus a missing chapter in CSLP docu- to the needs of the changing economy. Only then ments). What seems to be missing is "Financing can the many competencies of the Burkinabe-- Instrumentation for Implementation." While hard work, entrepreneurship, administrative abil- the case study suggests that the choice of financ- ity, and the like--be fully realized. Curt Carnemark C H A P T E R F O U R Ghana 35 Methodology Fund (UNICEF), the United States, and the World Bank. The study team was led by consultant Nils Boesen, Process & Change Consultancy. Team A workshop with a wide range of stakeholder members included consultant Anthony Killick, participants was held at the end of the study Overseas Development Institute, London; Laura period, generating useful feedback. The team Kullenberg and Mirafe Marcos, World Bank especially appreciated the contribution of the Operations Evaluation Department; and con- Minister of Planning and Regional Develop- sultant Abena D. Oduro, Center for Policy ment, Dr. P. Kwesi Nduoum. Analysis, Accra. The study team worked in Ghana during April Background 2­9 and May 6­24, 2002. The team used the first visit to select issues and themes for review For the last decade and a half, Ghana's economy based on advice from stakeholders. Selected has produced steady, if moderate, growth. The areas included the Ghana Poverty Reduction 2000 presidential election resulted in a peaceful Strategy and health, agriculture, decentraliza- transfer of power between political parties after tion, and economic governance. In May the a run-off--one of the few such instances in team added the broader issue of public sector African history following the defeat of an capacity constraints. Team members worked incumbent party candidate. Most of the coun- individually on report chapters, which were try's social indicators compare favorably with compiled and edited by the team leader. averages for Sub-Saharan Africa. The team made in-country district visits and conducted document reviews. In addition, the These positives, however, mask underlying team administered a survey to 13 major donors negatives. For example, while overall poverty in Ghana, with follow-up interviews. Comments levels have fallen, two-thirds of the poor live on donor experiences and perceptions are prima- in extreme poverty, with diets that provide rily drawn from the latter source. Donors less than their minimum nutritional needs. responding to the survey included Denmark, Furthermore, in some areas of the country Canada, the European Union (EU), the Food poverty has increased. This is the case in three and Agricultural Organization (FAO), Germany, northern regions, where poverty levels range France, Japan, the Netherlands, the United from nearly 70 percent to almost 90 percent. Nations Development Programme (UNDP), the (See box 4.1 for additional social and eco- United Kingdom, the United Nations Children's nomic indicators.) 36 Unique Country Characteristics structural and institutional reforms, although The country's recent economic and political sta- progress in civil service reform was not signifi- bility is in marked contrast to its earlier post- cant--limited public sector capacity continues colonial turmoil. For 15 years after the 1966 to undermine Ghana's ability to implement ouster of the country's independence leader, comprehensive development strategies. Kwame Nkrumah, Ghana suffered under a series of 6 military and civilian governments--gener- Political parties were not allowed in the 1980s. ally recognized as either corrupt, inept, or both. As the decade closed, pressure for a democratic By the early 1980s the country's physical infra- government was building. In response, the gov- structure was in serious disrepair; social services ernment established District Assemblies, fol- had deteriorated; and the economy was in des- lowed by a series of regional consultations on perate shape. GNP had shrunk by nearly 40 per- how to shape a new political system. In 1992, cent since 1970, and economic growth rates had Ghana's citizens approved a new constitution been negative for three years. Import and export establishing a parliamentary democracy. values had shriveled, while inflation ballooned to triple digits. The 1992 democratic elections resulted in the PNDC party remaining in power (although the The Provisional National Defense Council party adjusted its name to National Democratic (PNDC) government that came to power in the Congress--NDC). The government launched a 1980s was successful in implementing a series of new series of economic and social reforms. The Box 4.1. Ghana at a Glance POVERTY and SOCIAL Sub- Development diamond* Saharan Low- Ghana Africa income Life expectancy 2001 Population, mid-year (millions) 19.7 674 2,511 GNI per capita (Atlas method, US$) 290 470 430 GNI (Atlas method, US$ billions) 5.7 317 1,069 GNI Gross per primary capita enrollment Most recent estimate (latest year available, 1995-01) Access to improved water source Poverty (% of population below national poverty line) ­ ­ ­ Ghana Urban population (% of total population) 36 32 31 Lower-middle-income group Life expectancy at birth (years) 57 47 59 Infant mortality (per 1,000 live births) 58 91 76 Child malnutrition (% of children under 5) 25 ­ ­ Access to an improved water source (% of population) 64 55 76 Illiteracy (% of population age 15+) 27 37 37 Gross primary enrollment (% of school-age population) 78 78 96 Male 82 85 103 Female 74 72 88 incidence of poverty was reduced, from 52 per- the financial stakes for finishing the country's 37 cent in 1991­92 to 40 percent in 1998­99. poverty reduction plan (already in progress Economic growth through most of the decade under a broad-based consultative process initi- was stable, although never exceeding 5 percent ated by the previous government) were signifi- annually. Investment rates were also fairly cantly increased. Starting in July 2002, a special stable, and growth in exports healthy. Further National Development Planning Commission economic growth was hampered, however, by (NDPC) task force created five Core Teams to fiscal imbalances resulting from declines in provide input into the Ghanaian Poverty exports and, in election years, significant expen- Reduction Strategy (GPRS). The government diture growth as the incumbent political party then hired mainly Ghanaian consultants to syn- sought to gain favor with voters. thesize the Core Teams' reports, ministry input, and other materials. Although shifting the administration of govern- ment programs to the local level--decentraliza- The resulting GPRS combined economic tion--was put on the policy agenda with the growth and social equity concerns, and included creation of the District Assemblies, progress was many of the country's long-standing reform limited since no coherent strategy for carrying goals. The wide-ranging GPRS initiatives had a out the change was developed. In addition, civil preliminary total cost of $8.3 billion over three service reform stalled, a victim of the country's years--roughly five times the available level of ingrained patronage system. $1.6 billion. This is because the broad-based consultative process used to develop the GPRS The economy lost ground in 1999 and 2000, did not include a process for prioritizing invest- depressed by falling prices for the major export ments within a hard budget constraint, and thus commodities of cocoa and gold, as well as a resulted in a list of investments far in excess of shortfall in international aid. Starting in late available resources. 1998, the government began running budget deficits of such magnitude that the central When key aid donors made it clear to the gov- bank no longer automatically honored govern- ernment that the GPRS would serve as their ment checks. The resulting fiscal crisis illus- reference for shaping support, prioritization of trated that the country's reform measures had GPRS goals became essential. The new govern- not succeeded in diversifying the economic ment worked to ensure that their medium-term structure or markedly reducing dependence on economic priorities were included, particularly foreign assistance. their emphasis on the role of growth in reduc- ing poverty. The focus of the government elected in 2000-- the New Patriotic Party (NPP), with John By applying prudent fiscal management and a Kufuor as president--was therefore to stabilize tight monetary policy during its first year in the economy. After three months in office, the office, the NPP succeeded in cutting the infla- NPP government decided to apply for debt relief tion rate by half, from 42 percent to 21 percent under the enhanced IMF­World Bank Heavily (by 2002, the rate was 15 percent). Further- Indebted Poor Country (HIPC) initiative. Since more, in broad terms the government's 2002 a condition for accessing HIPC was develop- budget priorities did reflect an increased focus ment of a nationwide strategy to reduce poverty, on poverty reduction. Still missing, however, 38 was an overall plan to fit together medium-term built through political support, adequate imple- GPRS expenditures (even though a medium- mentation capacity, and continuous nurturing. term expenditure framework existed on paper, it In the Ghanaian context, a significant level of was not yet operational), other budget expendi- foreign assistance is accompanied by weak tures, and HIPC funds. public sector capacity. While the country and its donors have made progress in implementing Given Ghana's limited capacity to implement CDF principles, it is important to put this reforms, the GPRS's three-year time horizon progress in a realistic light--one that reflects needs to be extended. The three-to-five year the country's current capabilities. budget planning process required by the consti- tution can serve as a vehicle for doing so, and Chapter Structure the NPP government's intention to develop a The following four sections discuss Ghana's new long-term vision statement will also sup- progress toward each of the CDF principles-- port this process. Nevertheless, a realistic assess- Long-Term, Holistic Development Framework; ment of both central and local government Country Ownership; Country-led Partnership; capacity to carry out such long-term plans mod- and Results Orientation (see Introduction to erates expectations for success. this volume for a description of the CDF princi- ples). The discussion of each CDF principle is Although never effectively linked to the central divided into two segments: achievements already budget, several previous plans--specifically, the in place and remaining challenges. poverty reduction strategies prepared in 1999 and 2000--did reflect the four CDF principles. Ghana had become a CDF pilot country in 1999, Long-Term, Holistic through agreement by Ghanaian and World Development Framework Bank leadership. Ghana was subsequently chosen for the CDF evaluation because of its There are four questions to be asked in consider- pilot country status and its level of international ing whether the GPRS--the embodiment of aid--foreign assistance finances roughly 90 per- Ghana's approach to the CDF--is advancing the cent of public investment expenditures. CDF's long-term comprehensive (or "holistic") development principle. The four questions are: The current Ghana Poverty Reduction Strategy embodies the CDF principles. It remains the Does the GPRS contain a long-term vision? focus of donor agency efforts, as illustrated by its If so, does it strike a reasonable balance role as the key document at a recent govern- among economic, structural, and social ment-donor Consultative Group meeting, which dimensions? took place during the CDF evaluation period. Does it cover all sectors and establish appro- priate linkages among them? It is within this economic and social context that CDF principles are applied in Ghana. In Is it grounded in the central budget process? any context, implementing CDF principles is a large undertaking, often requiring major The first three of these questions are addressed changes and restructuring for both government here; the fourth is taken up under the Results- and donor agencies. These changes must be Orientation principle. As Yet, No Long-Term Vision devoted to reducing poverty versus those dis- 39 Does the GPRS contain a long-term vision? cussing long-term economic growth: 21 on eco- Under the version begun by the previous NDC nomic growth, and 5 on poverty reduction. government it did, but under the new NPP gov- Treatment of social sectors is also skimpy, with ernment this element has largely been aban- health issues (outside of HIV/AIDS) claiming a doned. The pre-election NDC spending spree mere 3 pages. The situation is underscored by meant that the new government had no choice President Kufuor's January 2002 State of the but to first address an immediate economic Nation address, which made no reference to crisis. Prodding by donors did lead the govern- poverty at all. ment to prioritize the wide-ranging list of GPRS initiatives so development work aligned The response to question number three-- to the government's goals could begin. whether the GPRS covers all sectors and estab- Initiatives that mirrored the new government's lishes appropriate poverty-fighting linkages platform and priorities were inserted and given between them--is not positive. The govern- precedence. Together, they primarily reflected ment's ministries jumped into the GPRS process an emphasis on measures to stimulate economic only toward the end. Early department and growth, including greater attention to private agency discussion that could have established sector development. such linkages simply did not take place. The result can be seen in the patchwork nature of A more fully articulated long-term vision may the final GPRS document. While some meas- materialize, based on the constitutionally ures are pro-poor, others have little connection required set of coordinated economic and devel- to poverty reduction, and there are no coherent opment policies the government presented to connections among them. Parliament in 2003 to cover the medium-term timeframe of 3-to-5 years. Thus the Ghanaian While it is desirable for a country's leadership experience has been to start not with a long- to have a clear sense of direction and to be able term vision, but instead from short-term crisis to carry the population with it (at the same management, then proceed to a medium-term time balancing macroeconomic and social con- poverty strategy, and to an eventual long-term cerns), the evaluation team has doubts about vision statement. The progression is under- requiring a more formal process in shaping a standable, given Ghana's political context. long-term development framework. Given the reality of a four-year election cycle, reaching Does the GPRS reflect balance between the agreement on a long-term vision statement government's goals for structural growth and would require a political culture and institu- social considerations? The GPRS represents the tions steeped in consensus-building. But these most substantial effort yet to put poverty reduc- conditions don't yet exist in Ghana, for a tion at the top of the policy agenda and demon- number of reasons: democratic traditions are strates the need for wide-ranging changes in still too new; the influence of clientelist sys- order to move toward the goal. The govern- tems of reward remains too strong; the overall ment's orientation to a more indirect "trickle- level of educational achievement is too lim- down" approach, as distinguished from direct ited; information flows are too weak; and local- poverty reduction measures, can perhaps be level participation is still a long way from being illustrated by the number of GPRS pages fully realized. 40 Country Ownership mentary committees and District Assemblies are not systematically included, which could poten- tially lead to a divide between the consultative Achievements processes and representative democracy. Although elections and democracy are not explicitly linked to the CDF ownership princi- There are other realities that affect Ghana's ple, in Ghana the most significant contribution ownership of the aid process. One is the high to broader country ownership of the aid process proportion of the central budget financed by may be improvement in the country's demo- aid--40 percent--and the other is the limited cratic processes. First, popular pressure led to the central and local government capacity for aid reintroduction of multiparty democracy in 1992. management. Still, if the government continues Then a democratically elected government to broaden the policy process, it could transform made some timid attempts at participatory poli- its role from aid validator to aid consultant and cymaking. And now the current government is then to aid participant. For now, however, the characterized by greater openness and tolerance, government is prevented from effectively taking and a more serious commitment to consulta- ownership of the aid process by lack of both tion--both civil society and the private sector capacity and a clear and convincing program are more fully engaged in policy consultations at around which coordination could coalesce. various levels, including those outside the GPRS process. Country-led Partnership Donors have generally been supportive of the shift toward country ownership at the policy level--illustrated, for example, not only by the Achievements GPRS process, but also in a sectorwide approach While both donors and government have taken (SWAp) in health and an agriculture policy steps toward building aid partnerships in Ghana, solidly anchored in the Ministry of Agriculture. progress so far has primarily been in laying a foundation through increased consultation and Challenges dialogue. Government-donor Consultative Nevertheless, by general agreement the quality Group (CG) meetings are held every two years of domestic policy dialogue leaves room for in Accra and co-chaired by the government. improvement. Civil society organizations And quarterly "mini-CGs," instituted even (CSOs) acknowledge that they often lack the before the official CDF launch, continue. Both capacity and resources to seriously engage in of these meetings are open to a wide range of policy discussions that require expertise and stakeholders, including civil society, and are research. And the documentation needed to well attended by donors. support policy analysis is often inaccessible or takes too long to obtain, leaving little time for Donors clearly accept the need for coordination. folding it into discussion. (The country's think Donor representatives have monthly meetings, tanks do have the potential to support CSOs in and experienced representatives surveyed for these areas, and to augment the policy analysis the CDF evaluation reported increased coordi- capacities of government and Parliament.) nation over the past five years. The World Bank Furthermore, the machinery for policy consulta- representative has played a key leadership role tion is not yet institutionally embedded: parlia- in shifting the image of the Bank to a more open Box 4.2. The First Health SWAp 41 In 1995, Ghana's health sector was in crisis. Many donor-funded programs were addressing the same issues; some areas were overfunded, while others were neglected. Government funding was declining. In response, the Ministry of Health developed a long-term vision for the sector. With support from six large donors (DFID, Danida, the World Bank, the Netherlands, the Nordic Development Fund, and the European Union), the government led an effort to build consensus on a five-year, medium-term reform strategy. The government convinced a handful of donors to fund the plan through a shared "basket" approach, and to adopt a common management arrangement: while previously Ministry staff had spent much of the entire year responding to accountants and auditors from 30 donors, the SWAp involves only one three-week review mission a year, liberating Ministry staff to do their core work. Donors receive the same financial statement and one annual report, reducing transaction costs. But according to Ministry of Health senior staff members interviewed for the evaluation, there has been some backtracking. Some donors are poised to reclaim ceded territory, the result of perceived poor government performance in areas formerly managed under donor-driven projects. And while basket funding from large donors has brought efficiencies, a large number of donors remain outside the basket. Finally, despite the SWAp's attention to expenditure and impact monitoring, improvement in health indicators is not well documented; a number of categories lack statistics that cover even four to five years. and inclusive institution that sees itself as a For the government's part, its commitment to partner of both government and donors. (For policy dialogue with donors appears strong, as example, the Bank was a strong advocate for illustrated by a vigorous endorsement of the donors taking a hands-off approach when the partnership concept in the CG meeting that government prepared the GPRS. The Bank also took place during the evaluation period. serves as the conduit for shared donor funding in Furthermore, the government has taken several the health SWAp.) Donors in general note that steps to reinvigorate discussion forums in the they have stronger relationships with the cur- areas of agriculture, public sector financial man- rent government than was the case in the latter agement, and decentralization. years of the previous administration. The degree of donor flexibility and delegation Challenges of authority to local representatives and While all parties are thus theoretically willing to national staff varies, though these are widely operate through partnership arrangements, seen as practices or even prerequisites that there are not many concrete, on-the-ground enable donors to pursue CDF principles. The successes--the health SWAp being one of the World Bank, among others, has moved the few. In the main, donors continue to operate in country director from Washington to Ghana, ways that have adverse effects on the public and has given national staff more control over sector. These include circumventing the gov- task management. Such decentralization is ernment budget; providing donor-driven assis- much appreciated by Ghanaians. tance directly to districts without coordinating through the central government; and paying the health SWAp has gone to extraordinary 42 higher salaries, which attract away the most tal- lengths to ensure sound financial management of ented government staff for aid project work. donor funds. The considerable challenge in working toward The government also helps create an environ- the CDF partnership principle in Ghana is ment favoring short-term efficiency over long- therefore to overcome the barriers to joint fund- term government capacity building and ing and uniform approaches that now prevent sustainability. Among the most prominent con- meeting-room agreements among donors from tributors is the four-year election cycle, with its becoming actual, coordinated programs. political demands for government accomplish- ments to cite at election time. In addition, con- As they frequently mentioned in interviews, sultants and special advisors connected to donor government employees are acutely aware of projects are often more efficient at delivering donor-supported project management units that the policy statements, reform plans, and justifi- operate outside of existing civil service mecha- cations for continued donor assistance than nisms. Donors, while less forthcoming on the underpaid civil service employees. subject, did acknowledge the mechanism's exis- tence. Most donors seem to consider these While these factors may explain why a parallel arrangements necessary evils, citing weak public civil service continues to exist in Ghana, it does sector capacity and the complexity of some proj- not explain why the government typically fails ects. The one initiative brought to the study to take a firmer stand when donors offer assis- team's attention in this area is aimed not at tance that does not reflect government priorities. abandoning the parallel practice, but at stan- Several civil service informants suggested that dardizing its employment conditions, salary the government should be firmer in rejecting levels, and allowances. projects that are not aligned with key priorities. In addition, employing "enclave" approaches Here, too, Ghanaian realities provide an explana- has led to duplication of efforts; limited commu- tion. National budget funds are far below the level nication of best practices; and--particularly in needed to keep ministry staff busy in meaningful capacity building--efforts focused on training work. Cars, gasoline, photocopiers, computers-- and workshops rather than on addressing such even paper--are in short supply. In districts vis- structural issues as the role of incentives in civil ited by the evaluation team, both health and service capability. agricultural workers were staying behind their desks because they have no way of getting to their Why do donors continue to use methods that clients. Under these conditions, even projects prevent central government and district-level that don't support government priorities are likely capacity building? Part of the answer lies in a to be accepted, because they come with comput- combination of demand for quick results; an ers, transportation, and operational funds that can institutional culture driven by disbursement be quickly released, thereby solving problems pressure; and, to some degree, the need to main- unrelated to formal project objectives. tain donor visibility. Also a factor is donors' ten- dency to avoid risk, a quality fostered by It all adds up to a kind of vicious circle: needed concerns about possible funding scandals in a reforms require more capacity for implementation country where corruption is high--which is why than is available. Also, the capacity development required for reform presupposes reform. One path insufficient government investment in moni- 43 out is broad-based civil service reform, which toring and evaluation. could restore sound incentives for government employees and phase out the distortions brought Achievements by parallel systems. Yet donors remain surprisingly Although the country's basic machinery for passive in this area--an area that appears to be measuring results is weak, Ghana's health sector key for continued progress toward CDF principles nevertheless pioneered one of the most innova- in general and the partnership principle in partic- tive SWAps in Africa--the SWAp uses per- ular. Another part of the solution could be formance data to assess progress, define sector increased donor support of the central budget. policy, and allocate resources. Results so far However, while donors are planning a SWAp in include improved financial management and balance of payment support, weak public sector reporting; an increased number of district out- financial management and questionable imple- reach clinics; and improved health outcomes in mentation capacity are again proving to be barri- some areas, notably for infant mortality and ers. In addition, stalled national procurement immunization coverage. reform is undermining donors' confidence in the government's commitment to providing improved In addition, the country's poverty reduction processes for shared aid approaches. strategy contains a number of specific, poten- tially measurable goals, and sets out fairly spe- But donors, too, can drag their feet--for exam- cific plans for monitoring progress. Efforts are ple, hiding behind the CDF's goal of national under way to strengthen the country's National ownership to avoid pressing the government for Development Planning Commission in order to greater attention to such critical, but politically build its capacity to measure progress toward sensitive issues as lagging civil service reform. poverty reduction. A puzzling factor is that And as long as the complex and sensitive issues preparation of the poverty reduction strategy did of public sector downsizing and pay reform are not include the Ghana Statistical Service. not addressed (estimated to be a 5-to-10-year Nevertheless, the service has set up a poverty- process), it is difficult to envision sustainable monitoring desk and developed a large-scale progress in public sector financial management, survey to produce district-level indicators--indi- or indeed in any other comprehensive public cators that have not been previously available. sector reform area. The conditions for creating sufficient commitment--in CDF terms, country There has also been progress toward integrating ownership--are simply not likely to be there. medium-term expenditure framework goals into central budget processes, which has the benefit Results Orientation of shifting the traditional emphasis on inputs to a results-oriented focus on outputs. In addition, mechanisms to better track government expen- There has been less progress toward achieving ditures, put in place in 2001, have resulted in the results orientation principle in Ghana improved performance. than toward the other three principles (as is the case with all the CDF evaluation coun- tries). This is not surprising, since it is a diffi-Challenges cult principle to achieve quickly, particularly Although an Economic Policy Coordinating given Ghana's weak technical capacity and Committee has been established to monitor the 44 central budget forecasting and monthly expen- Conclusions ditures, the committee's technical skills have deficits. The Budget and Public Expenditure Implementation of CDF principles in Ghana Management System could provide additional has been limited; the principles have not yet capacity, but implementation has been delayed been institutionalized. Results from initiatives and the system is not expected to come on that do reflect the principles have been mixed. stream until 2003. The strongest examples of progress under the principles (or of CDF-like principles before the The pioneering health SWAp has also had mon- official launch) are the SWAp in health; parts of itoring deficiencies. For example, there have the country's poverty reduction strategy, in par- been surprisingly few health indicators that con- ticular, the participatory process on which it is sistently track results; a number of indicator based and the increased focus of the central tables contained more "nil returns" than statis- budget on poverty reduction; and increased dia- tical entries. Among the factors undermining logue between donors and government and the ability of the Ministry of Health to gather among donors themselves. data at the local level are a poorly handled reor- ganization with an associated loss of personnel Ghana's Poverty Reduction Strategy (GPRS) and weak monitoring and supervision in the document has become the roadmap for relations wake of decentralization of Ministry functions. between the government and donors: donors The government's overall fiscal constraints are have made commitments to aligning their aid also a factor: the annual public sector expendi- according to its priorities. What has yet to be ture on health in Ghana is $7.50 per capita. assimilated, however, is the consultative process on which the GPRS is based: neither Progress in public sector financial management Parliament nor District Assemblies are con- in Ghana is hampered by poor data generation sulted in GPRS implementation. In addition, and dissemination and general government GPRS initiatives (largely shaped by the previous weakness in monitoring and evaluation skills. government) have not been linked to a holistic, These fundamental problems must be addressed long-term vision or an operational MTEF--the before the results orientation principle can be current government has preferred to stake out effectively implemented. In agriculture, donors its own approach to poverty reduction, with an have been strongly supportive of government emphasis on economic growth spurred by pri- initiatives to improve monitoring capability-- vate sector development. insisting, however, that systems be kept simple to reflect current capacity. Nevertheless, the current government has shown serious intentions to address the difficult One key to improving results monitoring is the issue of civil service reform and for taking on an gradual strengthening of the government's polit- entrenched patronage system in the process. ical accountability to citizens at all levels. Here This resolve is important, because improving the country's civil society organizations can play the capabilities of government employees is the an important role--for example, by ensuring bottom-line issue that must be addressed in that accountability mechanisms are developed order to make greater progress toward CDF prin- countrywide, not just among Accra's well-edu- ciples. Building public sector capacity will cated elite. require a significant long-term effort--a level of 45 Box 4.3. A Country with Real Potential In terms of donor attention, it appears that Ghana's star may be rising. As well deserved as this may be from the geopolitical, democratic, and economic perspectives, it poses the risk of increasing or main- taining the country's already high level of aid dependency. And, rather than promoting the application of CDF principles, there are already indications that high aid levels may instead promote increased donor competition and crowding in favored areas. Yet it will be difficult for the government to refuse donor offers in these areas, given the benefits that any project brings to a resource-starved and under- paid public administration. A solution could include government and donor cooperation in developing a strategy for external assistance. Elements could be monitored by peers from other aid-recipient countries and from donors and NGOs not active in Ghana. Basic elements could include: Government commitment to gradually decreasing overall aid levels Clearly stated preferences in aid approaches that balance the government's preference for budget support with donor requirements for accountability, and that balance need for short-term results with long-term capacity development objectives Framework for both strategic and operative aid coordination Commitment from donors to phase out distorting incentives, and to streamline and harmonize reporting procedures1 Shared donor commitment to avoid inappropriate offers of assistance; based on CDF principles, the agreement would be specific in spelling out desired behavior. 1. Such as the Framework for Multi-donor Budget Support (MDBS). effort not yet seen in Ghana, despite the gov- dialogue with the government and with the ernment's stated willingness to tackle the job. wider Ghanaian society. For their part, donors have largely failed to embrace this challenge. Instead, they have been Finally, continued progress toward achieving CDF undermining progress by setting up independent principles is premised on a government willing to project units that operate outside of the existing articulate a clear program and to assert leader- civil service. ship--and on donors who judge the program to be adequately oriented toward poverty reduction. For government, then, the challenge is to restore an efficient civil service and continue to While the analysis of CDF implementation in deepen democratic principles in the country. Ghana suggests that progress is likely to be grad- For donors, the challenge is to work together ual, with both forward and backward steps, the to align their processes; to continue to move example of the SWAp in health--in many ways from project to program aid; and to increase a model of CDF planning and implementa- agency decentralization, bringing agency staff tion--illustrates that the approach can be exe- from headquarters to local offices to improve cuted in the country. Chara Tsitoura C H A P T E R F I V E Romania 47 Methodology The Romania case study methodology also included document analysis. Taken together, the methodology and results are believed to be The Romania CDF case study team was led by sufficiently representative and robust to permit John Eriksson of the World Bank's Operations valid inferences about the strengths and weak- Evaluation Department. Other team members nesses of how CDF-related development pro- included B. Lynn Salinger, economist-consult- cesses are working in Romania. ant, Associates for International Resources and Development, Cambridge, Massachusetts; and Interviewees were selected from 64 institutions. Professor Dumitru Sandu, Faculty of Sociology, University of Bucharest. A Bucharest consult- ant, Manuela Sofia Stanculescu, of the Institute for the Study of the Quality of Life, Bucharest, Romania: A Long and conducted a Government-Donor Relations Rocky Transition Survey in support of the case study in June 2002. The three team members did fieldwork in Unique Country Characteristics Romania in two stages, during February 4­12 In 1989, with the Romanian economy close to and March 11­29, 2002. Through more than total collapse, more than 50 years of Communist 90 interview meetings in Romania and Wash- Party rule came to an end with the overthrow of ington, the team gathered the views of roughly Nicolae Ceausescu. While it was generally 150 central and local government officials and understood that the country's transition to a parliamentarians, as well as representatives of modernized economy would entail social civil society, the private sector, and donors and costs--as the state-provided social safety net international development agencies. and command economy were replaced with democracy and a market economy--the transi- The semi-structured qualitative survey con- tion period was expected to be short-lived. ducted through these interviews was supple- mented by two structured surveys: a questionnaire Successive Romanian governments, however, survey of government and donor representatives have made only stop-and-go progress in imple- that focused on government-donor relations, menting modernization reforms. Privatization of and a largely quantitative survey employing a state-owned enterprises has been slow, and cor- standardized questionnaire. The latter was ruption and weak regulatory institutions have administered to a random sample of more than hampered private sector business development. 700 respondents countrywide. Thus, the late 1990s saw three years' of negative 48 economic growth, with annual consumer price for in institutions--Romania has limited inflation averaging more than 80 percent. Since capacity to deal with the psychosocial and 2000, however, the situation has improved. educational problems of permanently institu- Economic growth averaged over 5 percent tionalized children. The World Bank and during 2001­02, and annual inflation was down other donors have worked with the Romanian to 23 percent by 2002. government to gradually reduce the number of institutionalized children through placement Despite recent positive economic trends, in foster homes. Romania's long and rocky transition period has had significant social costs. Countrywide clo- Despite social indicators that lag behind other sures of factories, farms, and mines have led to countries in the region, Romania did manage to increased unemployment. Average unemploy- reduce its infant mortality rate by roughly 30 ment rose from 6 percent in 1997 to 8.1 percent percent between 1989 and 2000, signaling by the end of 2002. Just over half of those with- important progress toward a better quality of out jobs are considered long-term unemployed. life. In addition, there is growth in ownership of durable goods and consumption in the form of With the deterioration of the country's econ- private cars and housing. omy in the last half of the 1990s, Romania's poverty rate increased. (It is too soon to know In 2000, international aid to Romania was whether the apparent turnaround of the econ- equivalent to 1.2 percent of the country's gross omy will mitigate this trend.) Romania has used national income, with the European Union two methods to measure poverty. The more (EU) as the largest donor. The other three recently developed method shows that poverty main sources of aid are the World Bank, the increased by about 5 percentage points between International Monetary Fund (IMF), and the 1995 to 2001, from just over 25 percent to European Bank for Reconstruction and nearly 30 percent, with no sustained improve- Development (EBRD). Pre-accession EU assis- ment even in extreme poverty levels. While tance provided an average of $620 million per extreme poverty levels fluctuated during the year over 1999 and 2000. The averages for the period, the trend was generally flat. (See box 5.1 international finance institutions were World for social and economic indicators.) Bank, $415 million; IMF, $375 million; and the EBRD, $280 million. There are long-standing pockets of poverty in the eastern and southern regions of the coun- Overall, however, the Romanian people are try. The most vulnerable groups are the young, becoming increasingly frustrated by the slow single-parent families and families of the pace of economic reform and high unemploy- unemployed, rural households, and members ment, and the resulting impact on many social of the Roma ethnic group. Romania's popula- indicators. There are growing concerns about tion also suffers from severe health problems, corruption and poverty; journalists cite the as reflected in infant mortality and life limited effect of media criticism on political expectancy rates. As a result of rising poverty decisionmaking and weak utilization of the during 1995­01, social dysfunction increased. media by the public to express popular criti- Of particular concern to the government and cism. Voting incumbents out after only one term the international community was the in- in office may be the only option for Romanian creased number of abandoned children cared citizens to express their dissatisfaction. Chapter Structure force behind Romania's long-term vision.3 For 49 The following four sections discuss Romania's Romanians, entry into the EU symbolizes the progress toward each of the CDF principles-- opportunity to join a peaceful, politically Long-Term, Holistic Development Framework; mature, and economically thriving regional Country Ownership; Country-led Partnership; system. It also serves as a kind of shorthand for and Results Orientation (see Introduction to eventual reduction in poverty and improvement this volume for a description of the CDF princi- in the quality of life for most Romanians. ples). The discussion of each CDF principle is divided into two segments: achievements already The current government, elected in 2000, has in place and remaining challenges. developed a Governing Program for 2001­04 and an accompanying Action Plan, prepared by the Ministry of Development and Prognosis. Long-Term, Holistic These documents constitute a medium-term Development Framework strategy, with primary emphasis on economic objectives and constraints that reflect heavily the goals and requirements of EU and NATO Achievements accession. The Action Plan is a detailed, multi- Accession to the EU and the North Atlantic chapter framework of specific actions to which Treaty Organization (NATO) is the primary the government is committed. It is not an Box 5.1. Romania at a Glance POVERTY and SOCIAL Four comparator Romania countriesa 2001 Population, mid-year (millions) 22.4 67.3 GNI per capita (Atlas method, US$) 1,720 4,161 GNI (Atlas method, US$ billions) 38.6 279.9 Data below are for most recent year available, 1995-01 Poverty (% of population below national poverty line) 30 18b Urban population (% of total population) 55 68 Life expectancy at birth (years) 70 73 Infant mortality (per 1,000 live births) 19 9 Access to an improved water source (% of population) 58 98c Illiteracy (% of population age 15+) 2 1c a. Comparator countries: Bulgaria, Czech Republic, Hungary, Poland. b. Data available only for Poland. c. Based on data for Bulgaria, Hungary, and Poland. 50 analytic document, a fact recognized by the tions of doing so do not seem to be well under- Ministry, which is also charged with monitoring stood by most Romanians. Unanswered ques- its implementation. tions include: "What do the accession goals imply for Romania's economic and social There has been a surprising level of agreement in organization? What do they imply for its rela- the country that accession is the right path, given tionships with the international community? Romania's splintered, multiparty approach to What are the economic, social, and political democracy. In 2000, eight political parties and costs?" And most fundamental of all, "What alliances were represented in the presidential kind of society does Romania hope to be in 10 elections. With so many parties, planning beyond or 20 years?" the next election cycle is difficult, let alone com- mitting to a long-term vision for the country. At the same time, Romanian development stakeholders are suffering from "strategy fatigue." Romania is slated to become a NATO member in As discussed in more detail under the country 2004 and a member of the EU in 2007. While the ownership principle below, between 1997 and adoption of the EU's Acquis Communautaire, 2000 Romania went through three broadly based with its 31 "chapters" of laws and regulations, is processes designed to create a long-term devel- designed to bring newcomer countries' practices opment vision for the country. Stakeholders are into alignment with those of EU members, it does now looking for greater government and donor not provide a coherent, long-term strategy for focus on achieving results, not more long-term several policy areas. These include most social policy formation. The new National Anti- policy issues (education, health, and poverty Poverty and Promotion of Social Inclusion reduction), as areas to be shaped by individual Action Plan does have the potential to provide countries themselves. the bridge from strategy to action in the poverty reduction arena. In 2000 the Council of Europe issued a Memorandum for Social Inclusion that requires But because the country's Governing Program each member state and country in accession to as well as the Anti-Poverty Action Plan are the EU to issue a Country Joint Inclusion weakly linked to the central budget process, Memorandum. Although this does not constitute tightening the relationships between these a requirement for an anti-poverty strategy as strategic frameworks is critical. Overall, there is such, Romanian leaders decided that promulga- a weak link between what is published in the tion of a Romanian anti-poverty and social inclu- budget and what is actually spent; budget prior- sion plan would both serve an important internal ities are not clearly stated. In 2002, the World need and meet this requirement for EU accession. Bank assisted in the preparation of a Public The prime minister issued the National Anti- Expenditure Review to address these issues. Poverty and Social Inclusion Action Plan in June 2002. The Plan contains recommendations for 2002­04 and for a longer, 10-year period. Country Ownership Challenges When the country ownership principle works While there is widespread support in Romania well, the recipient country, not development for joining NATO and the EU, the implica- assistance agencies, identifies development Box 5.2. Romania's Shared Vision--The 1999 CDF Consultations 51 As a first step in launching Romania as a pilot country working to build CDF principles, the World Bank planned a broadly based consultative process to create a shared vision for the country: 20 focus groups meeting in different parts of the country, with roughly 500 leaders from government, the private sector, civil society, and the donor community taking part. "It was the first real experience [in the coun- try] of soliciting opinions from all stakeholders," said a representative from a Romanian-based NGO. "From then on, many other donors and even the government began to open up their doors to NGOs. The CDF gave NGOs a framework for understanding what role was possible for them; how they could actually have influence." One result was a plan to institutionalize the process through a CDF Secretariat in the Prime Minister's Office. Although political instability and a change in government brought by the 2000 elections derailed the plan, participants interviewed by the case study team nevertheless cite two legacies of the 1999 effort: the identification of poverty reduction and entrance to the EU as key goals of Romanian stakeholders. Most participants found the consultation experience to have been very worthwhile. Although there has been progress toward implementation of CDF principles since 2000, neither the government nor the World Bank appears to have given enough attention to communicating that progress to those who participated in the 1999 CDF consultations. The majority of development stake- holders interviewed by the case study team see a need to revitalize the CDF approach as an avenue to increased government accountability. goals and formulates strategy. The process as the first CDF pilot country. The third focused involves regular and broad-based participation on developing a Medium-Term Economic from stakeholders, including sustained public Strategy (MTES) for the country, a pre-acces- support from top political leadership and intel- sion requirement of the European Union. Many lectual conviction on the part of key policy- of the MTES results were used by the govern- makers. The process builds consensus for a ment elected in 2000 to shape its 2001­04 long-term development vision. Action Plan. A certain synergy existed among these exercises. For instance, some political Achievements leaders elected in 2000 had participated in the Between 1997 and 2000 no fewer than three 1999 CDF consultations, with spillover contri- strategic visioning exercises took place in butions to the MTES. Romania, championed by different develop- ment organizations. Each involved broad-based While EU accession has become Romania's de consultation; many of the same stakeholders facto long-range strategy, there are nevertheless attended two or even three of the exercises. several recent examples--"nuclei"--of CDF- like consultative processes in the country. The The first, the National Strategy for Sustainable most significant include a plan for improving Development, was supported by the UNDP. The Romania's business environment; passage of the second, Romania's Shared Vision, was organized Law Regarding the Free Access to the by the World Bank in support of Romania's role Information of Public Interest, modeled on the 52 U.S. Freedom of Information Act; and the work slower private sector growth and less direct for- of the Anti-Poverty Commission in the Office eign investment. Several analyses in the late of the Prime Minister. A direct descendent of 1990s identified concerns with Romania's legal, the 1999 CDF consultations is the Jiu Valley regulatory, and judicial environments as barri- development strategy (see box 5.3). ers to increased private sector and foreign investment. In response, the World Bank's Business Environment Action Plan (BEAP). Not 2001 Private Sector Adjustment Loan provided surprisingly, Romania began its post­commu- resources for a business environment advisor. nist-era transition with a relatively small pri- The respected Romanian businesswoman vate sector. And compared with most other named to the position established a public and countries in the region, Romania has had private sector collaborative effort to draw up an Box 5.3. Jiu Valley Development: A Regional CDF Model The economically distressed Jiu Valley has been the focus of a CDF-inspired collaborative effort to improve the lives of its residents. Primarily a mining region, the Jiu Valley has been hard hit by several rounds of contract buyouts and mass layoffs of miners, which reduced mining-industry employment by more than half--from roughly 42,000 in 1997 to 18,200 in 2002. Because of the absence of jobs else- where and lack of retraining opportunities, relatively few families relocated. This among other factors led to social unrest that included violent protest marches. It was in this environment that a team made up of World Bank and government CDF Provisional Secretariat staff contacted local valley representatives to explore using the CDF's participatory approach to shape a development plan. A broadly based two-day seminar held in the valley laid out the approach: a 29-member steering committee would identify regional problems and potential solutions, and a larger representative body would provide additional insights. The process resulted in a draft Jiu Valley development strategy that represented local views on needs, priorities, and potential actions: the region's first example of a range of stakeholders coming together to work on a shared future--but would it be implemented? A delay in moving forward did cause concern, but in the summer of 2002 the government announced a plan to invest $360 million in the Jiu Valley. (Of an initial $80 million, $7.6 million would be supplied by the World Bank and the EU.) While there are some significant differences, the government's plan largely incorporates major elements of the locally produced strategy. The following results of utilizing CDF consultative processes for the Jiu Valley can be identified: Key inputs were mobilized for the Jiu Valley Development Strategy. The Jiu Valley Association, a local forum of local and regional stakeholders, was established. An Inter-Ministerial Commission for the Jiu Valley was formed. Much greater public awareness of regional development needs and strategic possibilities was created. The results are relevant as a model for the application of the CDF approach at the regional level. action plan, the Business Environment Action Challenges 53 Plan, or BEAP, to reduce administrative barriers While it is understandable that CDF principles to business. The plan is being implemented have faded as EU and NATO entry have under a Ministry division created for this pur- emerged as Romania's development lodestars, pose. Policy and participation issues in 13 sepa- many stakeholders who took part in the 1999­00 rate areas have been identified and several CDF consultations are disappointed by the targeted actions undertaken. CDF's eclipse. One Romanian government offi- cial interviewed by the evaluation team noted Freedom of Information Act. Romania's entry that his minister had asked, "What happened to into NATO is, in part, contingent on the the CDF?" In this official's view, "We still need a country firmly establishing the rule of law, forum to debate broad national and interna- democracy, and human rights--including citi- tional issues, [and] for civil society to give the zen access to government information. Given government feedback on how it's doing." Other the country's communist past, making free participants told the evaluation team that they access to information the norm has been a were disappointed when expectations generated complex undertaking. However, by late 2001 a by the groundbreaking CDF consultative process coalition of key lawmakers, the Ministry of were not matched by concrete results. Public Information, and Romanian and inter- national democracy and human rights advo- Although not as broadly based as the 1999 CDF cacy groups had led an effort that resulted in process, the Romanian government does pro- passage of the Law Regarding the Free Access vide opportunities for policy consultation to Information of Public Interest. Commonly among various stakeholders. These include the known as the Freedom of Information Act social dialogues held by the National Economic (FOIA), it was passed with both government and Social Council and the consultations with and opposition party support. interest groups organized by the Ministry of Development and Prognosis to monitor the National Anti-Poverty and Social Inclusion country's economic progress. Promotion Plan. The plan was drafted in 2001 with input from national ministries through The challenge, as a number of respondents to the their Social Dialogue Commissions (made up of evaluation's questionnaire survey noted, is to government, business, and labor representa- improve the effectiveness of these consultations. tives). In addition, the National Economic and Suggested avenues for improvement include: Social Council--with the same three categories Casting a wide net to include a broad cross- of representatives--as well as associations section of society made up of local authorities and parliamentary Social Protection committees provided input. Seeking input and participation from local Although the terms of Anti-Poverty Commis- experts sion members end with the 2004 elections, Employing a variety of tools, including members have nevertheless crafted a phased media, Internet, and focus groups to commu- action plan: in addition to recommendations for nicate conclusions and results. 2002­04, they have identified strategic direc- tions that cover a 10-year period and are in line Also needed is more effective and efficient with the Millennium Development Goals. communication among stakeholders. A signifi- 54 cant share of the country's development experts The National Authority for Protection of the and decisionmakers agree that better circula- Child was formed in 1999 in a response to tion of information is critical for CDF imple- increasing international concern about the mentation, especially in the face of such country's growing level of child abandonment challenges as weak institutional capacity, a and institutionalization. plethora of new regulations that still lack implementation guidelines, and centralizing An aptly named "High Level Group"--made up and sectoral tendencies that can run counter to of ambassadors, senior government officials, and the holistic and inter-sectoral approach of the heads of local offices of donor agencies--forged CDF. Networking the already-existing CDF consensus on a policy approach: gradually "nuclei" initiatives (for example, the Business reduce support for the traditional child care Environment Action Plan and the Freedom of institutions and place abandoned children in Information activities) through communica- foster families. On a parallel track, the approach tion and related programs or projects could fur- supports measures to prevent abandonment ther promote CDF principles. from occurring in the first place. The Child Protection Authority's 20-member Country-led Partnership Executive Group continues to meet. It includes 15 government representatives from central and local government; one from the Federation of As the only transition economy among the NGOs; and 4 from donor organizations--the case study countries, Romania does not have a EU, UNICEF, WHO, and the World Bank. Poverty Reduction Strategy Paper around Accomplishments include structural reforms in which donors can organize, as do low-income the child protection system and efficient sharing countries. While some donors do fit their assis- of program information and policy develop- tance into Romania's Medium-Term Economic ments among participants. Strategy and the Governing Program and Action Plan, most aid projects remain donor The other example of a national-level multi-stake- driven. Coordination of overall economic holder group is Romania's National Intersectoral assistance is limited and ineffective. One donor HIV/AIDS Commission. Established by the gov- representative offered an explanation: "Yes, ernment in response to international and domestic the government should do it, but the govern- concern, the Commission works through a ment itself is poorly coordinated." In addition, Permanent Secretariat housed in the Ministry of multi-stakeholder coordination is weak; with Health and the Family, supported by six subject- few exceptions, it is neither country-led nor area working groups. Among other topics, the donor-led. groups focus on social protection, treatment, and prevention of transmission from mother to child, Achievements and operate at both the central and local govern- Most aid coordination continues to take place at ment levels. The Commission's president is the project level in Romania. Coordination housed in the Prime Minister's Office, and vice rarely takes place at the sectoral level, and presidents are drawn from relevant ministries as almost never at the national level. There are, well as from the National Union of HIV Persons. however, two exceptions: national efforts in the Additional members represent other ministries areas of child protection and HIV/AIDS. and agencies, major donors, and NGOs. Challenges given its lead role in the budget process, the 55 There is interest among senior officials of the Ministry of Finance would be a logical choice. Ministry of Public Finance in a country-led part- nership role, particularly to ensure that pro- posed projects support the country's Governing Results Orientation Program Action Plan. As a practical matter, however, there are several constraints: lack of While there is progress toward implementing data on existing grant assistance, no agreement the CDF results orientation principle in within the government or among donors on Romania, efforts are generally at an early stage. which government ministry should lead aid In this fledgling democracy, Romanian citizens coordination, and weak government capacity need to build their knowledge and skill in for leading aid partnerships. demanding results from their government. Partly owing to the low number of jointly Achievements financed projects and agreements, donor proce- dures vary widely, leading to high transaction Romanians have, however, begun the journey costs for the Romanian government. In addi- toward demanding greater governmental ac- tion, lack of aid coordination means that some countability. With support from local NGOs and sectors have too many donors to handle com- at least one bilateral donor, at the end of 2001 the fortably, and other sectors have aid shortfalls. country passed a law guaranteeing freedom of public access to government information. The Unanswered partnership questions include, Ministry of Public Information has published a "What should be the aid coordination role for guide that explains what information is available the EU, Romania's largest donor, as well as for and how to apply for access, as well as what may the next three largest, the EBRD, IMF, and the still be restricted. In addition, the Sunshine Law World Bank?" And, "Should more members of is intended to provide citizen access to the legal the bilateral aid community be encouraged to and regulatory reform process by requiring public co-finance programs in collaboration with hearings on pending initiatives. other donors, or should they move to sector- wide assistance through direct support to the The Romanian government's capacity for mon- central government budget?" itoring and reporting development results gen- erally remains weak. The Ministry of Health At the time of the case study, Romania's Ministry and the Family took a step toward results ori- of Public Finance, with promised support from entation in 1997 when it gave increased the UNDP and the World Bank, was planning a emphasis to preventive care, and all ministries comprehensive aid database. A further positive are now in the process of adopting results indi- step could include periodic meetings between cators as part of a government-wide program- government and donors in order to air policy, based budgeting system. But there is a long way strategic, and procedural issues. (The last to go before the central government's goal of Consultative Group meeting for Romania took using results information for budget resource place in 1999.) Finally, the government should allocation can be realized. target both resources and reform efforts on strengthening capacity for aid management, and Romania's national development framework, settle on a lead ministry for aid management-- outlined in the Governing Program for 2001­04 56 and Action Plan, contains an extensive list of The case study team heard differing assessments objectives, sorted into 38 "domains" (examples of how effectively the Freedom of Information are financial policy, the labor market, and social Act and the Sunshine Law can be implemented. welfare). However, while the Action Plan does A stumbling block for FOIA implementation is specify steps to support each domain objective, the level of government resources available for only a few of these goals are measured through the effort: a broad public education campaign quantifiable results. Two exceptions include and additional government staff in all agencies limiting the annual budget deficit to a specified will be necessary to make it effective. In the case percentage of GNP and setting the percentage of the Sunshine Law, while the government is of state-owned industries to be privatized. optimistic about its implementation ability, NGOs wonder if the enormous task of bringing The Ministry of Development and Prognosis has Romania's laws and regulatory structures into responsibility for preparing a number of reports line with EU entry requirements won't sap its that track progress toward the implementation energy for doing so. of government programs. Beyond the Govern- ing Program Action Plan, these include the EU- International development targets can help related Pre-Accession Economic Plan and the stimulate demand for measurable results, and Business Environment Action Plan (PEP and benchmark targets set by the UN Millennium BEAP, respectively). Declaration of 2000 could serve such a role in Romania. However, while the country's Anti- While the Governing Program Action Plan's Poverty Action Plan does echo some of the lan- Social Welfare poverty reduction objective is guage used to describe the MDGs, Romania's supported by only one input-oriented action plan contains far fewer quantitative targets. step--clearly failing to provide a fully articulated poverty reduction strategy--the country's National Anti-Poverty and Social Promotion Steps for Improvement Inclusion Action Plan (prepared as groundwork What would be needed to accelerate Romania's for EU entry) does present a more comprehen- progress toward measuring development success sive plan. One of the plan's priority goals for by results rather than inputs? According to a 2001­04 is to establish a monitoring system for 2001 World Bank study, integrating results- measuring progress toward plan implementation. based management into Romania's governmen- tal structures will require six significant steps: Challenges Assign clear primary responsibility at the The Ministry of Public Finance has the respon- ministerial level in order to ensure the suc- sibility for analyzing the central budget by cess of a results orientation approach. sector--with input from relevant ministries and Strengthen the program budgeting effort agencies as well as local governments--and already under way, including a Court of then integrating individual ministry budgets Accounts audit to assess implementation into a consolidated budget. The Ministry efforts for all line ministries. admits, however, that making such judgments is limited because the sector expertise of its staff is Make results-based management a top-level still in its infancy. Therefore, monies still tend central government priority and build the to be budgeted based on prior spending levels. capacity to carry it out. Create a phased strategy to bring perform- have evolved into a long-term development 57 ance management to the local level. framework was, however, overshadowed by EU and NATO accession requirements as organiz- Enhance national statistical capacity. ing structures for the country's development. Encourage coordinated donor involvement in supporting performance management. Therefore, Romania's EU accession strategy, as expressed through several instruments--the A closely related factor is the need for public Medium-Term Economic Strategy; the Gov- service reform, including both salary reform erning Program and Action Plan, 2001­04; and (many government employees are so poorly paid now the National Anti-Poverty and Promotion they must take second jobs) and the gradual of Social Inclusion Plan--can be seen as a long- introduction of a link between pay and perform- term, holistic development framework. ance. The UK's Department for International Development (DFID) has funded an advisor in The extent of ownership of this de facto strategy the Prime Minister's Office to work on public by civil society, the private sector, and the pop- service reform, including moving senior public ulation as a whole is an open question, however. servants to a performance-based system. In addition, the strategy's linkage to the central budget and expenditure process remains tenu- Currently, however, Romania's development ous; without stronger linkage, it cannot be seen approach remains focused on inputs rather than as a fully operative strategy. It is easier to see results; programs and projects are rarely linked CDF linkages at the sectoral and local levels, at to quality-of-life goals. While the shaping of least for three CDF principles: strategy, owner- laws, regulations, and new strategies does have ship, and partnership. All three have been an important role in Romania's accession- applied, for example, in the Business Environ- related modernization strategy, putting all this ment Action Plan, the Freedom of Information activity in the larger context of the MDGs could Act, and the Jiu Valley Strategy. link the country's development to the larger quality-of-life improvements that global leaders have committed themselves to. It could also move Romania beyond aid largely delivered A Proposal through disconnected, consultant-led programs Despite the eclipse of the CDF framework by and projects to a more cohesive development Romania's focus on EU and NATO accession, framework--one that pays greater attention to many observers suggested to the case study team measuring results. that there was still a need for a CDF structure: "There needs to be a forum for civil society to give the government feedback as to how it's CDF Interrelationships doing," is how one Romanian government offi- cial put it. And indeed, a number of Romanian The countrywide consultative process that leaders interviewed for the case study expressed launched the CDF in Romania in 1999 both an interest in revitalizing the CDF initiative. outlined the elements of a comprehensive development vision and sowed the seeds for As a first step, laying out the record on CDF- country ownership of that vision. What could like accomplishments and comparing them to 58 initial goals articulated in 1999 could be Focusing on the resources and mechanisms for empowering. Perhaps donors--in particular, the development of such nuclei is of critical the World Bank--could provide significant importance for the development of a decentral- intellectual and financial support to Romanian ized CDF in Romania. The right mechanisms institutions and experts to nurture the current could help build a broad consensus in support CDF "nuclei" (for example, the Business of the country's development agenda--as well Environment Action Plan and the Freedom of as hold its public officials and the country's Information activities), with the goal of build- international development partners account- ing networking and convening mechanisms. able for results. Yosef Hadar C H A P T E R S I X Uganda 61 Methodology categories: those with key informants, identified by knowledge of subject and seniority within The Uganda case study was conducted during a key agencies or sectors and by referrals; group mission undertaken between October 25 and interviews with different stakeholders, including November 23, 2001. The findings are structured selected members of Parliament and govern- around the four CDF principles: Long-Term, ment officials; and group meetings with District Holistic Development Framework; Country Councils and local civil society representatives. Ownership; Country-led Partnership; and Results The education, health, and water and sanita- Orientation. The study team employed several tion sectors were selected for in-depth methods to gather information and insights: reviews on the basis of their apparent Document review. progress in applying CDF-like principles. Agriculture was selected because it is the Administration of a questionnaire survey to a mainstay of the Ugandan economy; 80 per- group of 33 persons from central govern- cent of the population lives in rural areas. ment, NGO umbrella groups, and several pri- vate business firms. (Among groups not Selection of districts was based on diversity represented were churches, trade unions, of socioeconomic status: Pallisa, as relatively business associations, and local government.) poor; Jinja, as relatively wealthy; and Kamuli, falling in between. The selection of Interviews with key informants and other districts does, however, have a regional bias, representatives from central and local gov- since all three are in Uganda's eastern region. ernment, parliamentarians, civil society and private sector organizations, and local donor representatives--150 individuals in all. A methodological limitation also results from the sampling approach used for selecting repre- Document review consisted of desk review of sentatives of the main stakeholder groups: economic, social, and political trends. These respondents were mainly based in Kampala; included development indicators in education, thus, the selection has an urban bias. The situa- population, and health as well as changes in the tion is mitigated to some extent by the inclusion poverty profile and the rural-urban divide. of civil society organizations that are umbrella organizations representing their grassroots The questionnaire survey was administered to NGOs (for example, the Uganda NGO Forum all stakeholders in a group meeting setting. The and DENIVA). results were later shared with a group of parlia- mentarians and other stakeholders who had also Despite these limitations, the evaluation team, completed the survey. Interviews fell into three led by John Eriksson, Operations Evaluation 62 Department, World Bank, believes that the Plan, or PEAP, was--and remains--grounded in case study findings yield a valid picture of the sectorwide approaches that are linked to the CDF implementation experience in Uganda. country's medium-term expenditure framework. The consulted representatives--150 in all-- A further CDF precursor was donor acceptance drawn from civil society, government, donors, of Uganda's Poverty Reduction Action Plan as parliamentarians, and the private sector, are the country's aid framework, as illustrated by among those most knowledgeable about devel- increased donor coordination and greater donor opment issues in Uganda. In addition, their support for the Ugandan central budget. voices were complemented and corroborated by document reviews. Uganda developed its precursor CDF principles as a way to put a fractured country back together The study team is confident that, taken after nearly 15 years of civil war and fiscal mis- together, an informed picture of progress toward management. Driven by the need to sustain implementation of the CDF principles emerges. security and stability, the country's post-war In addition to John Eriksson, team members leadership reached out to diverse factions in included Mirafe Marcos, Operations Evaluation establishing a new government and shaping eco- Department; David Pedley, Department for nomic reform policies--for example, by solicit- International Development (UK); and Rosern ing input not only from relevant government Rwampororo, consultant. ministries, but also from academics and elected representatives. The result was broad country ownership of both process and results. Uganda: Rising from Uganda's success in achieving economic stabi- the Ashes lization has encouraged international donors to provide significant levels of aid, with the World Bank the leading provider by far. Unique Country Characteristics During 1998­00, international aid provided In all six of the case study countries, some ele- just over two-thirds of Uganda's government ments of the CDF approach were already in expenditures, and accounted for 11 percent of place by the time World Bank President James the country's gross national income (GNI) Wolfensohn officially launched the CDF in (compared to an average of 3 percent of GNI 1999. In five of these countries, however, pre- for low-income countries). The private sector cursor CDF processes were not very effective-- also responded to Uganda's improved eco- the biggest stumbling block being lack of an nomic climate: its share of GDP increased action plan to translate theoretical principle from about 8 percent in 1990­91 to a peak of into budget-based reality. nearly 12 percent in 1996­97. Over 1999­00, private investment averaged about 11 percent The exception was Uganda. Not only were of GDP. development principles that anticipated the CDF in place well before the official CDF The combination of economic stabilization launch, but articulation of the principles was and allied donor support had impressive results. accompanied by a comprehensive action plan The country's inflation rate dropped from a focused on poverty reduction. Launched in high of more than 40 percent in 1992 to single- 1997, Uganda's Poverty Eradication Action digit levels from 1994 to 2000. GDP growth averaged 6.4 percent a year in the 1990s; in housing, as well as increased citizen access to 63 1999­00, it was 5.1 percent, and in 2001­02, services and information and to income-earning 5.6 percent. opportunities. (See box 6.1 for basic social and economic indicators.) Poverty reduction. By 1995 economic stabiliza- tion allowed the Ugandan government to turn Presaging the overarching goals of the CDF its attention to a countrywide assault on approach, the PEAP provided a comprehensive, poverty. The impetus was President Yoweri long-term development framework focused on Museveni's reaction to the World Bank's draft poverty reduction. The result: When the CDF Country Economic Memorandum (CEM). concept was officially launched by the World President Museveni commissioned a National Bank in 1999, the PEAP, with its already estab- Task Force on Poverty Eradication to take up lished sense of country ownership, became the the CEM's challenge for the country to do much recognized vehicle for implementing CDF prin- more to tackle poverty. Eighteen months' of ciples--which Uganda's pioneering comprehen- widespread stakeholder consultation produced sive approach to poverty-reduction had greatly the Poverty Eradication Action Plan. contributed to. Launched in 1997 and revised in 2000, the Remaining Challenges PEAP recognizes that while economic growth is While Uganda has achieved economic stability necessary for poverty reduction, other factors and a significant reduction in poverty, there are must also be in place. These include good gov- several challenges to continued progress. These ernance and improved health, education, and include the drain on the country's resources Box 6.1. Uganda at a Glance POVERTY and SOCIAL Saharan Low- Development diamond* Uganda Africa income 2001 Life expectancy Population, mid-year (millions) 22.8 674 2,511 GNI per capita (Atlas method, US$) 260 470 430 GNI (Atlas method, US$ billions) 6.0 317 1,069 GNI Gross per primary Most recent estimate (latest year available, 1995-01) capita enrollment Poverty (% of population below national poverty line) 35 .. .. Urban population (% of total population) 13 32 31 Access to improved water source Life expectancy at birth (years) 42 47 59 Uganda Infant mortality (per 1,000 live births) 88 91 76 Lower-middle-income group Child malnutrition (% of children under 5) 38 .. .. Access to an improved water source (% of population) 52 55 76 Illiteracy (% of population age 15+) 38 37 37 Gross primary enrollment (% of school-age population) 128 78 96 Male 132 85 103 Female 124 72 88 64 caused by insurgencies in two border areas and as by major civil society organizations and pri- by the cost of battling HIV/AIDS. The conflict- vate-sector representatives. Serving as Uganda's afflicted North is the only region of Uganda Comprehensive Development Framework, the where poverty has increased, rather than PEAP has several key characteristics: decreased. In addition, although Uganda has begun to shift the planning, delivery, and man- Comprehensive articulation of goals aimed not agement of basic services to local governments, just at poverty reduction, but at poverty erad- much remains to be done to build local capacity ication. Many of the PEAP's sub-goals are and expertise to make districts and communities reflected in the more recent Millennium truly effective delivery channels. Development Goals (MDGs), also shared by the international aid community. Several of And while Uganda's high level of international the PEAP's key targets are more ambitious aid in the form of both grants and loans is, of than MDG targets. course, a blessing, it also presents challenges: a growing share is structured as support provided Practical relevance, since the PEAP has spawned Sector Action Plans that cover directly to the general budget or to specific sec- 5­10 years. That stakeholders from govern- tors--aid seen by donors as more vulnerable ment, civil society, and donors have worked than project aid to accountability concerns. together in a sector over a number of years Furthermore, exchange rate appreciation and has resulted in realistic strategies that link debt unsustainability have also emerged as issues. plans to available resources. Chapter Structure Hard budget constraint, due to the PEAP's The following four sections discuss Uganda's linkage to the medium-term expenditure progress toward each of the CDF principles-- framework. Strong leadership, supported by a Long-Term, Holistic Development Framework; stable economic environment, encourages Country Ownership; Country-led Partnership; budget compliance. and Results Orientation (see Introduction to this volume for a description of the CDF princi- A living framework, because the PEAP is ples). The discussion of each CDF principle is revised periodically to take account of divided into two segments: achievements already changes in the needs of the poor as identified in place and remaining challenges. by reliable data arising from the monitoring of poverty trends. Long-Term, Holistic Given its widely recognized status as an author- itative and operationally relevant guide to Development Framework poverty reduction in Uganda, the PEAP fills another aid-related role. A summary of the Achievements PEAP serves as the country's Poverty Reduction The overarching role of the Poverty Eradication Strategy Paper (PRSP), a requirement for coun- Action Plan. The PEAP is now well-established as tries seeking enhanced debt relief under the the strategic framework for Ugandan develop- joint World Bank­International Monetary Fund ment. It is known throughout central govern- HIPC initiative. The Ugandan government suc- ment, at district levels, and in Parliament, as well cessfully argued that since the PEAP already constitutes a poverty-reduction strategy docu- reforms and strategic development focused on 65 ment, it would not be a good use of limited reducing poverty. Senior-level officials have also human resources to create a separate, very empowered lower-level leaders to design and similar, document. The World Bank is support- carry out strategies for poverty eradication. ing the implementation and monitoring of Uganda's PEAP/PRSP through Poverty Reduc- A lead institution dedicated to building on previous tion Support Credits--credits that previously reforms. Guided by a visionary permanent secre- went to specific government sectors. tary and supported by a highly capable eco- nomic team, the Ministry of Finance, Planning Savings from debt relief (as of 2000­01, US$98 and Economic Development (MFPED) built on million) have created space in the central budget its successful implementation of macroeco- to finance poverty reduction activities and for nomic reforms to become the lead architect for monitoring results. While this is, of course, the PEAP. MFPED has encouraged line min- highly beneficial, the release of HIPC money istries to focus on poverty reduction and has also also raises expectations for results--expectations encouraged other stakeholders--civil society, that must be carefully managed. Uganda has set donors, and NGOs--to become more involved up a Poverty Action Fund (PAF) to identify in formulating poverty-reduction policies and in these monies within the budget, allowing easy monitoring results. The MFPED has been accountability for how they are spent. instrumental in the move toward sectorwide planning, linking budgeting with the strategic investment plans of sector ministries; both line Challenges ministries and sector working groups now have Financing the PEAP and managing expectations. significant input into the budget process. Since not all PEAP programs can be financed in the medium term, difficult decisions must be Movement toward donor integration in the budget made in order to reprioritize existing spending, process. The increasing openness of the budget mobilize additional resources, cut costs, and/or process has encouraged some donors to integrate scale down and stretch out selected PEAP tar- their funding with the government's budget gets--all of which will require firm budget disci- system, and to make commitments earlier in the pline. Government is committed to preparing a budget cycle. The result has been increased gov- long-term expenditure framework that will fully ernment control over the budget process, elaborate the financing needs of the PEAP and enhancing the efficiency and effectiveness of ensure that expenditure allocations are in line expenditure allocation. And, since more finan- with PEAP costs. cial resources are directly available to sector ministries, it has encouraged them to develop sectorwide plans. Country Ownership MFPED as the main coordinator of donor activity and sector working groups. The Ministry of Achievements Finance, Planning and Economic Development Top leadership with a strong commitment to reform. has forged agreements between donor partners Since inflation was brought under control in and line ministries to work within government 1992­93, senior government officials have systems and frameworks. Contributing to the shown a firm commitment to major economic success of this shift is the willingness of some 66 donors--in particular, the World Bank and decentralization process has had mixed results DFID--to make the change from program across districts, in general it has brought citizens at implementer to program financier. In addition, the grassroots level much closer to policy plan- the government has consolidated aid coordina- ning and decisionmaking. tion units that were previously scattered over several ministries into one Aid Liaison Broadened ownership through civil society participa- Department under MFPED's wing. The move tion. Stakeholder interviews, questionnaire survey keeps donor-funded projects within the limits of results, and pertinent documents all show that the sector investment plans, and has largely pre- government has broadened country ownership of vented donors from tempting line ministries the PEAP by involving civil society in: into undertaking projects centered on donor, Shaping PEAP revisions rather than country-defined, priorities. All donor-funded project proposals must be cleared Monitoring PAF performance by the Aid Liaison Department and the govern- Contributing to the annual budget process ment's Development Committee, which is through their participation in the Sector chaired by MFPED. Working Groups that produce the Budget Framework Papers. Innovative use of increased fiscal space from debt relief and donor support to target poverty. In addi- However, civil society and private sector organ- tion to fiscal discipline, MFPED took the lead in izations and local governments in outlying areas establishing borrowing discipline and prudent tend to be excluded from the process. debt management. The Ministry negotiated sub- stantial debt reductions and rescheduling from Active government and CSO involvement in most bilateral creditors, and developed strate- Consultative Group meetings. For the last three gies that wiped out virtually all commercial years, CG meetings, designed to review govern- loans and cleared the bulk of arrears and penal- ment-donor relations, have been held in-coun- ties. In addition, it established a "grants before try and co-chaired by the World Bank and the loans" policy. In cases where loan financing is Ugandan government. The meetings reflect the only option, borrowing can only be done on broad participation by diverse parties. concessional terms that are comparable to those of IDA. Challenges Strengthening local government capacity in order to Power and decisionmaking is shifting from central gov- further advance "letting go" by the center. The cen- ernment to local governments. The sweeping decen- tral government provides financing to districts in tralization mandated by the Local Governments the form of both conditional and unconditional Act of 1997 is reinforcing country ownership grants. While conditional grants allow the cen- by giving local governments, and the citizens tral government to ensure that expenditures are who elect them, control over a wider range of consistent with the PEAP, the strings attached to policies and decisions. A number of donors are such grants also reduce local government's abil- providing support to districts to strengthen local ity to respond to changed circumstances. capacity. These include the World Bank's Local Government Development Program and USAID's Giving local governments greater freedom in project to strengthen decentralization. While the resource allocation will, however, first require: Strengthening local government capacity in has not yet returned to the Committee. 67 ways that are community-led and gender Depending on final language, if enacted the leg- responsive islation could either reaffirm the participation of NGOs in the development process or curtail Achieving more timely disbursement of their role--many NGOs see the bill as an funds to local governments and facilities attempt to limit the role of civil society. For such as schools their part, some members of Parliament inter- Improving flow-of-funds accounting and viewed for the case study were quite explicit in tracking their criticism of what they regard as the exces- sive role of the many development NGOs. MPs Achieving better employment conditions saw these organizations as insufficiently repre- for local staff, especially in remote hard- sentative, lacking accountability and adequate ship areas. capacity. They felt that these organizations are deflecting both government and donors from Increasing domestic revenue to reduce aid depend- more direct engagement with Parliament--an ency. Country ownership will be further engagement the MPs would welcome. enhanced when the budget becomes less dependent on external financing. (Fifty-seven Evolving role of Parliament. The role of percent of the FY2000­01 budget was financed Parliament in the budget process was expanded by donors through grants and loans.) Shifting by the Budget Act of 2001. Stakeholders con- the balance will require sustained economic sulted for the CDF evaluation view the act as growth and/or increasing the current low share having both pros and cons. On the positive side, of domestic revenues in GDP. More efficient tax the creation of a new standing Budget Com- collection would be one avenue toward increas- mittee and supporting Budget Office has the ing the domestic revenue share. potential for contributing to constructive engagement with the Ministry of Finance, Plan- Creating a positive environment for the private ning and Economic Development--thereby sector. Uganda's economic policies call for broadening country ownership without weaken- growth led by the private sector. All parties ing the long-term, comprehensive development agree that the emphasis on removing remaining framework. On the downside, the insertion obstacles to private investment and exports of Parliament into the resource allocation under the Medium-Term Competitive Strategy process at an earlier stage has generated concern makes good sense. Beyond that, though, the that hard budget constraints may be more diffi- Government will have to be careful to ensure cult to maintain. that the private sector's desire for more proac- tive support does not lead the government to implement measures that would risk costly mis- Country-led Partnership allocation of resources into investments that do not produce positive economic benefits. Achievements Reaffirming the role of civil society. While a new Donor assistance aligned with the Poverty NGO Registration Bill was tabled in Parliament Eradication Action Plan. Several factors have in 2001, the responsible committee referred it resulted in better alignment of external aid with back to the Minister of Internal Affairs; the bill the PEAP: mechanism. SWAps have led to a relatively 68 Rigor of the budget process linking the PEAP to the medium-term expenditure framework high degree of budget support and procedural standardization, or "harmonization," on the part Increased effectiveness of the Aid Liaison of donors in Uganda. For example, in the edu- Department and the Development Com- cation and health sector SWAps, 50 percent of mittee in screening and reviewing projects aid is now in the form of budget support. For the Increased budget support by donors most part, donors accept the report of the Joint Sector Review as fulfilling their reporting Efforts by those donors still providing project requirements (see box 6.2). support to ensure alignment with the PEAP. (The World Bank is among donors in the In a further advance, while the Uganda PRSC forefront of aligning assistance strategies provides budget support under a policy that had with government development strategy; an initial focus on three SWAp sectors (educa- others include DFID; the EU; Irish Aid; the tion, health, and water), it now also includes Netherlands; the Norwegian Agency for cross-cutting public sector reforms. Development Cooperation, or Norad; and the Swedish International Development Broader mandate for budget sector working groups Authority, or Sida.) (SWGs). The government has asked the budget SWGs, which have traditionally met only during Progress of SWAps and the PRSC. By requiring all the budget process in order to formulate sector partners to think and act in sectorwide terms, Budget Framework Papers, to expand their role SWAps promote alignment with the PEAP. A and develop output and outcome indicators for SWAp typically includes a sector policy; a their respective sectors. The government has mutually agreed long-term Strategic Investment also asked the groups to consider assuming an Plan; and an established joint sector review ongoing monitoring and evaluation role. Box 6.2. An Essential Element: The Joint Sector Review The purpose of the Joint Sector Review (JSR) is to allow government, donors, and other stakeholders such as NGO umbrella groups to review strategy and performance, and to propose changes in sector policy and resource allocation for the coming year. Participants therefore see the JSR as an essential element of an effective sectorwide approach. Benefits include greater strategic alignment; a higher degree of standardization in procurement, budg- eting, and disbursement; and improved donor specialization in areas of expertise. (The World Bank, for example, is reducing the spread of its involvement across sectors, with its future portfolio split between support for four sectors through the PRSC--education, health, rural development, and water--and for major infrastructure projects such as power.) An example of the benefits that can result from the JSR is the experience of the Ugandan education and health sectors, where the number of mutually negotiated undertakings by donors declined sharply over the last two years: from 59 to 7 in education, and from 32 to 10 in health. Transaction costs also dropped significantly. Challenges Further harmonization of procedures and practices. 69 Vulnerability of budget support. Budget support aid While sectorwide approaches--especially in that is not earmarked for specific purposes can be education and health--have resulted in used to cover expenditures for any government progress toward standardizing donor procedures purpose. Some donors providing unrestricted for budgeting, disbursement, procurement, budget support expressed concern in case study reporting, and evaluation, line ministries still interviews that their governments would react complain that more progress needs to be made, unfavorably to an escalation in Ugandan defense especially for project aid. (Ministries cite World spending, border conflicts, or increased evidence Bank project procurement regulations as being of corruption, since donor budget support could particularly complex.) Dealing with many dif- then be viewed as contributing to such undesir- ferent donor procedures is an administrative able expenditures as increased military costs. burden for government and drives up transac- (That in Uganda the dialogue with donors and tion costs. other stakeholders covers the entire donor/ government expenditure program, including Partnership in capacity building for improved serv- ceilings on defense expenditures, reduces the ice delivery. The challenge for government and sting of this argument.) its development partners is to find more effec- tive ways of supporting capacity building at The Ugandan government has included two both the central and local levels, so that aid can assurances in the PEAP to meet such concerns. address needs identified by the country itself The first is a continuing commitment to holding and its district subdivisions. A corollary chal- defense spending at 2 percent of GDP; the lenge is to let go of practices associated with second is to increase the share of public expen- supply-driven technical assistance--aid that ditures allocated to the PAF. meets donor needs--which tends to undermine country and local capacity building. For exam- Expanding the SWAp frontier. The PEAP states ple, the persistence of Project Implementation that over time the plan should be implemented Units (PIUs) is a continuing irritant for gov- through sectorwide approaches in all major sec- ernment, which sees their technical assistance tors. However, developing a coherent sectoral choices as reflecting donor rather than govern- approach will be particularly challenging in sec- ment preferences. Another perceived drawback tors that include many different ministries with is PIU pay scales: typically higher than govern- differing interests and competing budgetary ment scales, they are seen as causing resent- claims--Public Administration is one example. ment and as distorting incentives. Devising a shared approach is also difficult for Managing the transaction costs of partnership. sectors such as Social Development that achieve While the benefits of aid coordination across their most effective results through close collab- sectors in Uganda continue to exceed costs, oration with several ministries. In addition, the communication costs of implementing stakeholder differences over strategic priorities these donor-government partnerships are not may be greater in some sectors (for example, trivial. In preparation for PRSC 2, the govern- agriculture) than in others. The second PRSC ment reviewed cross-sectoral coordination provides a framework and processes designed to mechanisms with a view to streamlining them strengthen coordination and meet these collab- and further reducing both government and orative challenges. donor costs. 70 Mutual performance assessment. Mutual assess- involvement of civil society in monitoring per- ment by government and development partners formance and results. The information require- of each other's performance currently takes ments of donors have also stimulated capacity place in varying degrees through sectoral and for the collection, but not necessarily the in- cross-cutting coordination mechanisms. A country analysis and use, of results information. Workshop on Partnership Principles, held in September 2001, constituted a cross-sectoral A milestone poverty monitoring and evaluation review of the performance of all partners in strategy. The strategy, announced by the minis- development cooperation. Holding such a ter of finance in June 2002, has both a medium- review periodically would provide a means for term and a long-term focus, and is compre- continued refinement of objectives, establish- hensive in its coverage of government systems. ment of baselines, and mutual assessment of Major elements include a systematic assessment progress toward country-led partnership. of the national budget process and strengthened tracking of funds to priority areas, as identified Surmounting the macro challenge. The combined in the PAF. The strategy is being built into the effect of implementing the PEAP and shifting budget process in stages: data are collected from responsibility for service delivery to local gov- line ministries on inputs and outputs, and infor- ernment has been to greatly expand the demand mation on impact on development programs for public services at the local level. However, if will be obtained from periodic surveys and stud- a donor desires to provide additional budget sup- ies (including the Demographic and Health port to meet this growing demand during the Survey, conducted every five years, and the budget year, the offer will be turned down by the annual National Service Delivery Survey). MFPED if it would break the annual medium- term expenditure framework ceiling. Deter- The role of civil society in monitoring results infor- mining the actual macro-effects of growing aid mation. The involvement of both government flows requires a better understanding of the rela- and CSOs in monitoring the use of PAF funds tionships among aid inflows, the real exchange has become a significant force for increased rate, and productivity growth. The government transparency in how funds are used. Represen- and the World Bank, along with one or two tatives of civil society view monitoring of public other donors, are undertaking an empirical study expenditure by communities as one of the most to shed more light on these relationships. important initiatives for improving the quality of services to the poor. Results Orientation Poverty eradication goals echo--and sometimes exceed--Millennium Development Goals. Uganda's Achievements MDGs are set within the framework and time Monitoring and analysis of results on the ground are horizon of the PEAP and related documents, a key PEAP feature. Soon after the 1995 start of such as the Education Strategic Investment the broad-based process used to formulate the Plan. In several cases--poverty, education, PEAP, demand to analyze and use results infor- HIV/AIDS, and safe water--Uganda's targets mation emerged as a desirable PEAP element. are more ambitious than MDG targets. Additional demand has been stimulated by the Furthermore, progress so far suggests that they PAF, by decentralization, and by the increased are achievable, as long as the necessary financ- 71 Box 6.3. Grassroots Monitoring of Poverty-Reduction Progress The Uganda Debt Network, a CSO that has assumed a proactive role in monitoring progress in poverty reduction, carries out its mission through work with grassroots organizations. For example, the network has helped the Kamuli District's Poverty Monitoring Committee track the effectiveness of local PAF dis- bursements through: Organized dialogue with elected representatives Collection of grassroots information through primary schools, health centers, and local administrative units. Findings were presented at a workshop to provide feedback to all stakeholder groups. While local elected representatives chose to view the results as personal criticism, district technical staff has used the findings to shape a comprehensive plan to remedy identified shortcomings. The Kamuli District Poverty Monitoring Committee has followed up with relevant government departments: one result is that timeliness of drug deliveries to local health units has improved. ing is forthcoming and quality issues are audits in order to improve delivery of services to addressed (a concern, for example, in the areas the poor. Weak culture and incentives to gener- of education and water and sanitation). ate and use results information at the operational level were also cited as barriers to monitoring and evaluation. If CSOs are to play a larger role Challenges in monitoring service delivery, they will need Achieving fiduciary assurance. Widespread cor- support for strengthening their capacity to do so. ruption at a significant level, reported in surveys and anecdotal accounts, is a major concern for Using results information: Outcome-Oriented the government and among development part- Budgeting and Results-Oriented Management. ners. It constitutes an impediment to greater Although Uganda is a pioneer among low- foreign investment and to increased aid in the income countries in introducing two major form of budget support. While pending pay instruments for using results information to reform will reduce one incentive for corruption, improve performance, neither is fully opera- a continuing challenge will be the vigorous pur- tional. The two systems are Outcome-Oriented suit of complementary measures. These include Budgeting (under MFPED), which feeds results the existing anti-corruption campaign as well as information into the budget process; and efforts to reform the procurement system, Results-Oriented Management (under the strengthen financial tracking, and increase Ministry of Public Service), which feeds results transparency of accounts at all levels. In addi- information into institutional and individual tion, PRSC 2 contains further support for anti- performance appraisal processes. The two sys- corruption efforts. tems developed independently in different ministries; to maximize their impact, they Effective service delivery monitoring yet to be institu- should either be integrated or aligned. PRSC 2 tionalized. Key government informants reported contains steps designed to bring about better the need for conducting more "value for money" alignment of the two initiatives. 72 Sequences and Budget Support, Aid Reinforcing Relationships Quality, and Fiduciary among CDF Principles Assurance The importance of sequencing and reinforcing From the perspective of the Ugandan govern- relationships among CDF principles emerges ment, budget support is higher quality aid than from a review of Uganda's experience. First project aid. Benefits include flexibility in allo- came ownership, exemplified by a country deci- cating aid between recurrent and development sion that basic macroeconomic and market lib- costs, as well as lower management and report- eralization reforms were essential. This was ing transaction costs. followed by a holistic development strategy, the Poverty Eradication Action Plan. Implemented Donors, however, have greater fiduciary concerns through the medium-term expenditure frame- about providing aid in the form of budget support. work, the PEAP reflects tough budget choices While recognizing the risks, several donors have that in turn reflect the strategic priorities of sec- nevertheless concluded that enough progress on torwide approaches. These achievements fiduciary assurance has been made (including encouraged development partners to align their account transparency and financial tracking) to approaches with those of the government, justify providing general budget support aid. strengthening country-led partnership. Donor Furthermore, PRSC 2 provides a framework for movement toward budget support and sector- monitoring continued progress in the transition wide approaches reinforced country ownership. to budget support aid. The challenge for donors Finally, the combined impact of applying the still providing project support is whether they are first three CDF principles has required that prepared to follow the budget support approach. more attention be paid to the fourth principle, The challenge for the Ugandan government is to results orientation. continue to allay donors' fiduciary concerns attached to budget support aid. Francis Dobbs C H A P T E R S E V E N Vietnam 75 Methodology The Vietnam country case study was launched in September 2001 with a preparatory mission The Vietnam CDF evaluation was carried out by to Vietnam. A subsequent mission from a multidisciplinary team of national and inter- November 20 to December 12, 2001, collected national experts drawn from the development the majority of the evaluation data. community. Led by Alf Morten Jerve, assistant The study's conclusions rest on a number of director, Chr. Michelsen Institute, Bergen, information sources: interviews with leaders Norway, team members included Ray Mallon, and selected staff of 70 agencies and organiza- economic advisor and private consultant, tions; relevant reports and other documents; Hanoi; Keiko Nishino, executive director, and survey results from slightly more than 100 Global Link Management, Tokyo; Han Manh respondents engaged in development work in Tien, director, Concetti, Hanoi; and Laura the country (three-quarters of whom were Kullenberg, Operations Evaluation Department, Vietnamese serving in government agencies or World Bank. The services of Mr. Mallon and NGOs, and one-quarter expatriate represen- Ms. Nishino were funded by the Asian tatives of aid organizations). In addition, team Development Bank (ADB) and Japan's Ministry members attended two Consultative Group meet- of Foreign Affairs, respectively, as in-kind con- ings, several key workshops, and made field tributions to the evaluation. Mr. Hajime trips to Ho Chi Minh City and Da Nang Province. Takeuchi of the Japan Bank for International Cooperation (JBIC) and Mr. Graham Walter of ADB participated in the planning mission for Vietnam and the CDF: the evaluation. Progress, Yes, but The team also benefited from the contributions Challenges Remain of a local advisory group consisting of senior advisors of key research institutions and donor Unique Country Characteristics agencies, including the Central Institute for Over the past decade, the socialist government Economic Management; the Development of Vietnam has launched a remarkable series of Strategy Institute; the Vietnam Union of market-based economic reforms focused on pri- Science & Technology Association's Institute of vate sector development and the global integra- Management; DFID; the Asian Development tion of Vietnam's economy. The success of these Bank (Hanoi); the Ministry of Foreign Affairs initiatives transformed the country's formerly of Japan, Tokyo; and the Japan Bank for closed economy into one with a high ratio of International Cooperation, Tokyo. both foreign investment and exports to GDP. 76 Vietnam's policies to combat poverty have also CDF principles--especially when combined been markedly successful. The country's poverty with other strengths. rate dropped from 58 percent in 1993 to 32 per- cent in 2001--a reduction of nearly half, virtu- Even though many Western donors and multi- ally unparalleled among low-income countries. lateral institutions only resumed aid programs in (See box 7.1 for regional and low-income coun- the early 1990s, Vietnam has established a good try comparisons of Vietnam's poverty level and track record of aid management, with a success- other social indicators.) ful liberalization program and no donor condi- tionality. The country's long tradition of central The international aid community has provided planning has built a strong sense of government significant support for these reform efforts, par- ownership over policies. And the country's ticularly poverty reduction. Today Vietnam is series of 10-year socioeconomic plans--dating the second-largest current IDA borrower, with back to 1945--gives Vietnam experience in set- Japan as its largest single donor. Without a ting development targets and monitoring doubt, Vietnam is a development success story. progress toward achieving them. All these factors, as well as the country's long A Solid CDF Foundation and unique experience as an aid recipient (first, Compared with many developing countries, with most assistance from Russia and other com- Vietnam's economic and social reforms and its munist countries; more recently, from the West), commitment to poverty reduction provide a contributed to Vietnam becoming a CDF pilot particularly solid foundation on which to build country in 1999. Early in 1998, Vietnam's then Box 7.1. Vietnam at a Glance POVERTY and SOCIAL East Asia Low- Development diamond* Vietnam & Pacific income 2001 Life expectancy Population, mid-year (millions) 78.7 1,826 2,511 GNI per capita (Atlas method, US$) 420 900 430 GNI (Atlas method, US$ billions) 33.4 1,649 1,069 GNI Gross per primary Most recent estimate (latest year available, 1995-01) capita enrollment Poverty (% of population below national poverty line) 32 ­ ­ Urban population (% of total population) 25 37 31 Access to improved water source Life expectancy at birth (years) 69 69 59 Vietnam Infant mortality (per 1,000 live births) 37 36 76 Lower-middle-income group Child malnutrition (% of children under 5) 34 12 ­ Access to an improved water source (% of population) 56 74 76 Illiteracy (% of population age 15+) 6 14 37 Gross primary enrollment (% of school-age population) 110 107 96 Male 113 106 103 Female 107 108 88 Prime Minister, Nguyen Manh Cam, had antici- and Results Orientation (see Introduction to 77 pated the focus of the CDF by urging donors to this volume for a description of the CDF princi- cooperate with each other and with his govern- ples). The discussion of each CDF principle is ment--citing the number of single-donor aid divided into two segments: achievements already projects that were overtaxing the country's aid- in place and remaining challenges. management capabilities. Upon the 1999 launch of CDF principles by World Bank President James Wolfensohn, the Bank's country director Long-Term, Holistic strongly supported Vietnam as a CDF pilot, also Development Framework citing the need for a new type of collaboration among donors that would allow more focused, sectorwide approaches. Achievements The CDF's long-term approach to development The Bank's country office was strengthened is, for the most part, a logical fit with Vietnam's with staff seconded from other agencies--some historical commitment to centralized planning, working exclusively on promoting partnership. which has long included 10-year national strate- This proved to be an excellent way to establish gies and 5-year economic development plans. trust and exchange skills and perspectives, and One sticking point, though, has been that shap- secondees interviewed felt they had an influ- ing long-term national strategy (including ence on Bank thinking as well. The World Bank development strategy) was, until recently, also provided technical expertise to other aid strictly a party-led activity. agencies and to Vietnam's government to help shape sectoral approaches--that is, approaches Vietnam has, however, made major changes in that attack issues from several related fronts (for how it sets long-term strategies, for the first time example, by recognizing the impact of environ- inviting broad nongovernmental and donor par- mental issues on health indicators). ticipation in reviewing drafts of the most recent 10-year plan. According to evaluation survey Vietnam's impressive reforms are still in a tran- responses, there is strong agreement, among both sition stage. For several years the World Bank Vietnamese and expatriate respondents, that the and other aid agencies have cited the need for a planning process has undergone significant second wave of reforms to sustain the country's change, becoming much more open and consul- economic and social progress. Donors also have tative. Furthermore, respondents agree that significant concerns about the efficiency and strategies are now more focused on poverty effectiveness of aid in Vietnam, noting in par- reduction. (It is true, however, that Vietnamese ticular the need for stronger government respondents have a considerably rosier perspec- involvement in project management and tive when it comes to the question of whether greater efforts to reduce corruption and waste. long-term objectives have become more realistic: 86 percent of local representatives say yes, while Chapter Structure only 16 percent of expatriate respondents agree.) The following four sections discuss Vietnam's progress toward each of the CDF principles-- Challenges Long-Term, Holistic Development Framework; The process of reconciling a commitment to Country Ownership; Country-led Partnership; socialism with market-oriented management has 78 not been entirely smooth, however. The pace of mechanism to bridge the theory-to-practice gap. reform has been slow, as officials with differing While the PRSP process was launched globally views wrestle with consensus. And the same in late 1999 to ensure that countries granted need to compromise among divergent views too debt relief by the World Bank and the IMF used often results in only a hazy statement of the budget savings to reduce poverty, these plans country's long-term vision. In other words, the have also provided the action steps for imple- mere existence of long-range strategies does not menting the CDF's aid-effectiveness principles. automatically translate into a fully defined long- For example, Vietnam's CPRGS also plays an term and comprehensive development frame- important role in supporting the results-orienta- work; the quality of the plan also counts. tion principle. The quality of Vietnam's long-range planning could be improved by: Country Ownership Expressing political choices and policy in unambiguous ways Achievements Strong government ownership of policymaking Setting targets that are realistic and formu- has been a hallmark of the Vietnamese political lated to give clear policy direction process since independence, and the tradition has been extended to ownership of the overall Explicitly linking sector reforms to the coun- development process. Strengthening the gov- try's overall development objectives ernment's policy commitment has been the suc- Defining the relationship between long-term cess of Vietnam's economic reform process, strategies, medium-term planning, and which was well under way (yielding increased annual budget allocations. economic output, employment, trade, and investment) when the multilateral financial The most critical gap in the long-range plan- institutions resumed work in the country in ning process is that plans are not yet linked to a 1993. Even if foreign aid had little direct impact medium-term expenditure framework, which on successful economic reform (although aid- results in a disconnect between the expressed related training and advice were factors in long-term vision and strategies, and how the developing reform policies), the country had annual budget is actually spent. This gap is illus- nevertheless come to appreciate the value of trated by the government's reports to the learning from international experience. Consultative Group (CG) meetings, which are designed to explore donor-government issues: As previously noted, Vietnam's national plan- funding requests remain in the "wish list" cate- ning process was traditionally restricted to the gory, rather than reflecting clear choices among party structure. During the preparation of the investment sectors. last two 10-year strategies, however, domestic and foreign experts were asked to comment on The development community holds out hope drafts (with more consultation and debate that the country's Comprehensive Poverty during preparation of the most recent strategy). Reduction and Growth Strategy (CPRGS)-- Drafts of the most recent five-year plan were Vietnam's version of a Poverty Reduction also more widely circulated, to all provinces as Strategy Paper, or PRSP--may provide the well as to NGOs for the first time. In addition, nonstate organizations and private sector repre- technical assistance projects, which remain 79 sentatives say they now have better access to largely donor-driven. government information, which includes partic- ipation in state-led meetings and workshops This is related to a major complaint from most (NGOs and private sector representatives are Vietnamese interviewed--the donors' nearly now, for example, invited to CG meetings). exclusive use of foreign consultants for interna- And donors applauded the government's 1998 tional development projects, when, they argue, Grassroots Democracy Decree, which estab- using more national experts would both strength- lished a legal framework for citizen participation en local project ownership and improve project at the local government level, thus increasing design (which has often been flawed because local agency accountability. most international consultants have little knowl- edge of the Vietnamese context or language). Overall, the government's strategic relationship with donors has been handled well; aid has served Change is also needed on the government side. practical needs and has opened the country to Ministries need to improve inter-ministry com- new international alliances. And while Vietnam munication and reduce the level of so-called has encouraged better aid coordination at the "silo thinking,"which undermines opportunities operational level, it has not done so by reducing for donor cooperation and a more holistic the competition of ideas or the diversity of its approach to development problems. As in most bilateral relations, as might have been the case countries, individual ministries often prefer to when donors collaborated on larger projects. deal bilaterally with a trusted, flexible, long- time donor rather than be part of a broader sec- toral or national program, where they fear losing Challenges control over funds, program content, and Even though government ownership of policy- bureaucratic power. making and the overall development process is widely recognized, both donors and government And some ministry officials are concerned that representatives note that ownership doesn't multi-donor sectorwide approaches, or SWAps, necessarily extend to individual projects. They could increase rather than decrease their admin- cite in particular weak national involvement in istrative burden if a number of donors descend Box 7.2. The MDGs: Proof of Ownership The capacity of Vietnam's central government to engage in fruitful dialogue with donor agencies has improved significantly in recent years. An example is the shared effort to "localize" the MDGs to reflect local conditions and the country's long-range plans. After committing to the MDG goals, the government--supported by the government-donor-NGO Task Force on Poverty--modified these international human development targets to reflect the government's own strategic goals. In addition, Vietnam moved up the target date for achieving the goals--from 2015 to 2010--in order to bring them into alignment with the country's strategic planning timeframe. The localized targets have been incorporated in the country's CPRGS. The CPRGS provides the country's action plan for poverty reduction and economic growth, approved as such by the prime minister. 80 on a sector without standardizing ("harmoniz- While it is too early to assess the long-term ing") their procedures. impact of the working groups on aid effective- ness, a number of changes are already evident. In addition, Vietnam needs to reform and These include donor input into government strengthen public administration, including the sector strategies, which increases coherence; development of more professional project man- increased likelihood that donors consult with agers: strong country ownership depends on each other before planning projects; and gov- civil servants that have the time, resources, and ernment recognition of the benefits of informal skills to do their jobs well. contact with donors. Taken together, these changes can be seen as a "new way of thinking Finally, respondents note that government and doing business" in the planning and man- could improve both its perception of and man- agement of international aid in the country. agement of prospective development aid. To some government officials, aid is still seen as a Donors have increased joint analytic work as "free good" that augments underfunded budg- well as the number of jointly funded aid proj- ets, rather than funds for specific development ects. While parallel funding is the most outcomes that will have to be accounted for common approach, examples of the newer and repaid. In the few cases in which the gov- pooled, or "basket funding," concept are emerg- ernment has declined a project (an important ing. In addition, several bilateral donors indicator of government ownership), political (Denmark, Germany, Sweden, and the United sensitivities were the cause, rather than devel- Kingdom) have increased their involvement in opment considerations. projects that are jointly funded with the World Bank and the ADB. Country-led Partnership It has become clear that in order for partnership efforts to work, countries and aid agencies need Achievements to grant greater autonomy to their field offices-- There is tangible progress toward government the Netherlands, Sweden, and the World Bank and donor cooperation in Vietnam. For exam- are leading examples of this approach in ple, there are now more than 20 active partner- Vietnam. Without the authority to make deci- ship groups. Covering most broad development sions on the spot and negotiate partnerships with themes, some groups include only donors; some flexibility, movement toward greater col- others also include government representatives. laboration is painfully slow. Also helpful, note In 2000, the World Bank, ADB, and UNDP observers, is careful selection of key agency per- jointly compiled a report on the partnership sonnel: in particular, they note the importance groups, summarizing their policy focus, activi- of selecting senior staff with the leadership and ties, and progress. This is updated every six interpersonal skills suited to the relationship- months and the effort underscores the nearly building process--which implies an ongoing universal acknowledgment encountered by need for staff with negotiation and communica- the evaluation team that Bank leadership has tion skills as well as specific technical expertise. had a significant impact on creating both the The evaluation team noted that with the World environment and the mechanisms for partner- Bank's new focus on partnership, as well as ship in Vietnam. increased staff resources in the country, the Bank 81 Box 7.3. Lessons from the Poverty Working Group Founded in 1999, the Poverty Working Group (PWG) has met every month to facilitate dialogue between government ministries whose work has a bearing on poverty reduction (for example, health and education) and aid agencies with similar concerns. In doing so, the PWG: Promotes greater understanding of donors' priorities among Vietnamese counterparts Supports ministry-donor dialogue with analytic work Reduces the language barrier by highlighting the need for simultaneous translation in government­aid agency meetings Through analytic input, provides the framework for the PRSP process (called the Comprehensive Poverty Reduction and Growth Strategy, or CPRGS, in Vietnam). The PWG also helped reduce skepticism about the benefits of partner efforts by publishing a well- received joint report, Vietnam: Attacking Poverty. DFID and the World Bank coordinate the informally structured group. has assumed the important role of convener in issue and agreed in 2001 to take concrete steps recent years--making the continuity of staff to harmonize their aid procedures. This group with communication and partnership-building has now grown to nine donors and has expanded skills even more important. its mandate. On the government's side, Decree 17-CP, issued In addition, the government has taken an in May 2001, outlined the basic legal framework important step by promulgating a new law on for aid management; donors are now working to competitive bidding, which opens up the possi- align their procedures within this framework. In bility of private companies competing for public addition, several ministries have signaled their contracts. The CDF survey suggested, however, intent to establish procedures for donor rela- that the government could do more to enhance tionships. And ministries are slowly embracing the environment for partnerships by clearly multi-donor sector programs--SWAps--which communicating its strategic direction to donors bring greater resources and a clearer strategic as a group. focus to development projects. As noted earlier, both the government and Challenges donors are working to reduce aid administration While the Vietnamese government's aid man- and management burdens by "harmonizing" agement abilities show improvement, manage- procedures; that is, working toward standard ment capacity at the provincial level and below practices that can reduce the higher transaction remains a weak link. One observer from the costs associated with fragmented aid efforts. donor community estimated that only about Six like-minded bilateral donors (the "Like- one-sixth of provinces now have the ability to Minded" group) funded an assessment of this operate as full partners with donors. 82 Box 7.4. Consultative Group Meetings: The CDF in Microcosm Country Consultative Group (CG) meetings are held annually--and more informal sessions, semi-annu- ally--to explore donor-government issues, particularly projected donor financing. In Vietnam, the tenor of these meetings has changed dramatically; they have become a microcosm of CDF principles: Formerly held outside the country, meetings are now held in Vietnam and co-chaired by a govern- ment representative. The government acceded to donor requests to include representatives from the private sector and NGOs. Donors were invited to comment on the government's most recent 10-year strategy plan. There is general consensus among government and aid agencies that the partnership approach is the best way to do business in Vietnam. The Grassroots Democracy Decree, approved in Implications for Donors 1998, was designed to foster local community Building partner relationships can overtax the participation in development initiatives by pro- abilities of some donors, particularly small viding for information sharing, consultation, donors that have limited staff and administra- and local involvement in decisionmaking and tive resources. A possible solution is for smaller project monitoring. Direct involvement of local donors to build focused expertise that would government in shaping aid programs and proj- allow them to make significant contributions in ects remains limited, however, the result of sev- specific (but fewer) sectors. Some strategic eral factors--including weak capacity and little selectivity is beginning to emerge: The Nether- leeway for spending local revenues on develop- lands, for example, has focused on developing ment activities. Planning capacity is particularly expertise in the governance area; its technical weak in many of the rural provinces that have input is recognized for its high quality. Sweden the highest proportions of people living in focuses on the health sector; the Asian poverty. For their part, many donors have a poor Development Bank on forestry; and Denmark understanding of local language, circumstances, on fisheries. and institutions, which also hinders effective planning at the provincial and local levels. While at face value harmonizing aid procedures would seem to be a logical and straightforward Nevertheless, a number of donors are working to way to increase aid efficiency, in practice it is increase project disbursements at the provincial extremely difficult and time-consuming. At the level, and some provincial authorities have been very least, changing procedures requires difficult involved in partnership groups--for example, in decisions at the headquarters level. It often the areas of forestry and disaster mitigation. The requires internal restructuring and significant central government has decided to adjust its expense as well, and, in some countries, new approach to land clearance for infrastructure national legislation. The interim solution has projects, recognizing the need for a participatory been to seek a middle ground where like-minded process in order to solve resettlement problems donors can harmonize as far as possible, or har- without major popular discontent. Aid-financed monize in clusters, with lenders and grantors projects helped trigger the shift. working separately. For example, the multilat- eral banks (World Bank, Asian Development time, the National Assembly is taking more 83 Bank, and Japan Bank for International interest in aid issues; it has commissioned its own Cooperation) are now conducting joint program assessments of aid efficiency and effectiveness. reviews and have started a joint harmonization The general tenor is illustrated by the results of initiative with government. the survey taken for this evaluation. The great majority of respondents believe that aid effective- While most donors endorse the CDF principles, ness has improved at least "somewhat"; one out of that does not necessarily mean they practice what 10 respondents cited "major improvements." they preach. CDF-like steps for donors would include developing longer operational and finan- Challenges cial horizons; working to reduce staff turnover; Some CPRGS targets appear to be too ambi- and maintaining policy continuity--observers tious: they are based on a 7.5 percent annual note that donors change policies and shift their growth rate in GDP, which may not be realistic, focus to new aid approaches (or even to develop- given the slowdown in global and Regional ment "fads") more quickly than governments. economies. In addition, there is a sharp division in survey respondents over the basic question of whether Vietnam's development targets are Results Orientation measurable. While 86 percent of local respon- dents said they are, more than two-thirds of Achievements expatriate respondents said they are not. Even though the results-measurement principle has proven to be the most elusive to implement, Other barriers to results-oriented measurement initial steps have been taken. Among the most include the continuing reluctance of some gov- significant is the fact that the country's CPRGS ernment agencies to share planning and project requires measuring performance by results. documents; weak institutional capacity; and the need for more energetic reform of the civil service. The most basic need for improving results meas- urement is, of course, gathering the data that While the Vietnamese government did launch a allows analysis. In supporting the CPRGS, the Public Administration Reform (PAR) program government has committed to improving its in 1995, initial progress was slow. In an effort to statistical basis for measuring income-level improve the process, a working group on public changes through a multi-purpose survey of Viet- administrative reform (led by the UNDP) was namese households; the survey is now being formed in 1998, with the goal of adding donor developed by the General Statistics Office. support to the government's efforts. In addition, the country's liberalized stance that The group met frequently, with substantial gov- now allows NGOs and private sector representa- ernment participation, and by 2000 had com- tives to have input into strategic planning has pleted a public administration review, which in enhanced national budget and aid accountability. turn provided major input into a government And the general availability of public informa- Master Plan for PAR. The plan sets out action tion is improving, as illustrated by government steps for achieving a number of reforms, includ- publication of the 1999 National Budget, the first ing improving personnel development and man- time a budget has been published. At the same agement and modernizing the administrative 84 system. While the Master Plan significantly reinforcing and intended to be implemented increases the focus on intended outcomes, it is together. Not as well recognized, however, are too early to judge results. the tensions among the elements. An exam- ple occurs when a donor's partnership agenda Donors need to give more thought to their own and the government's ownership interests do procedures in relation to the results-orientation not coincide. principle; in particular, to the reward mecha- nisms at work when development indicators are Just such a conflict developed in Vietnam over aggregated and viewed through a long-term per- health sector reform. The health system had a spective. Targets need to be linked to geograph- serious quality problem, and private health care ical areas, not simply to national averages. is now a preferred alternative for those who can Targets also need to be linked to concrete afford it. The result is that even though the reform outputs (for example, covering pre-pay- country has an impressive record in preventive ment schemes in the health sector, not just care, its extensive network of health centers is measuring improvement against human devel- grossly underutilized. opment indicators). Choosing the future direction of the Vietnamese Finally, while there is increased access to gov- health care system is, however, as much an ide- ernment information, including the national ological as a technical decision. At issue is how budget, there could be wider sharing of develop- much of the system should be privatized and ment information--by having it translated into market-based versus the size of the segment Vietnamese, for example, in order to make it financed by public revenues. The case illustrates available to affected constituencies. And the the dilemmas in the aid partnership when recip- media could be more involved in monitoring aid ient institutions (in this case, the Ministry of effectiveness. To date, they have mainly focused Health, which had proposed publicly funded on aid flows, not results. One solution proposed pre-payment plans) and a donor (here the by Vietnamese academics is to create a third- World Bank) disagree on project direction. party independent technical body that would review major aid projects, focusing in particular And, while the government clearly recognizes on quality control and the bidding process. the need for health care reform, the pace has been slow. A positive interpretation is that the Finally, nongovernment stakeholders warn of the government is "making haste slowly"; a less pos- risk to project quality control if donors step too itive one, that vested interests in the system far back in the name of partnership, citing weak oppose reform. monitoring capacity at the provincial level. When the Bank suggested that the Ministry of Health was sometimes more of a bottleneck Tensions than a clearinghouse for consensus and deci- sionmaking, the Ministry noted that "a main prerequisite for good collaboration is that the Ownership­Partnership Tradeoff donors share the basic value premise of the There are important synergies among all the health sector, as expressed in plans and deci- CDF elements: the principles are mutually sions by the party and the government." Clearly, the donors' partnership agenda and world means at least several decades, while 85 government's ownership concern do not always political pressures in aid-giving countries often coincide. Nevertheless, the sectorwide approach demand results indicators within two years. A allows such disagreements to be transparent and final tension involves applying CDF principles publicly debated--which in itself is an improve- to countries emerging from conflict, or countries ment over past practices. under severe political stress. In these situations, the government is often focused on day-to-day Other tensions at work in Vietnam as well as in survival or reconstruction; CDF principles may other CDF case study countries include the be out of their reach. asymmetry of power between donors and recipi- ents. The key to any aid management paradigm In all cases, the responsibility of both donors is how much, and under what conditions, and recipient countries is to find a workable bal- donors cede decisionmaking to recipients-- ance, often between two worthwhile goals. keeping in mind their accountability to either When that balance is found, CDF principles add their legislatures or their boards. significant value to the aid relationship. In Vietnam, the evaluation team found wide recog- Another fundamental conflict can be found nition of the relevance of CDF principles to the between the CDF's long-term and results-ori- aid process, with particularly impressive per- ented focuses. Long term in the development formance in the area of partnership. A N N E X Annex: Evaluation Methodology 87 Selection of Countries for the primary purpose of the evaluation was to look at implementation issues, the Management Case Studies Group and Steering Committee decided that selecting the longest-running and best-perform- The following six countries were selected for in- ing CDF pilots, together with one non-CDF depth case studies: Bolivia, Burkina Faso, pilot, was the optimal approach for generating Ghana, Romania, Uganda, and Vietnam. the type and quantity of data required. (A dis- Selection was determined in the first instance cussion of sampling strategy that guided this by a country's desire to participate in the evalu- choice is found in the GAO Guidelines for Case ation. Priority was given to former CDF pilot Study Evaluations.)4 countries with the most implementation experi- ence and the highest performance rankings (according to the CDF Secretariat) on grounds How Country Studies that these cases would offer the greatest poten- tial for learning. One non-CDF pilot country Were Conducted (Burkina Faso) was chosen as a control. Because the PRSP is defined as an instrument for imple- Country studies were carried out in two phases: a menting the CDF principles, all countries preparatory mission followed by intensive field selected (except Romania) were also PRSP work. During the preparatory mission the evalu- countries. Consideration was also given to ation was planned and designed in consultation Regional balance and avoiding overlap with with country authorities and donor representa- similar evaluative efforts. tives in order to reflect individual country inter- ests and circumstances. Small focus group The budget for the case studies was predeter- meetings were convened to propose and test key mined, so the inclusion of more than six cases, evaluation questions, taking as a point of depar- while desirable, would have been at the expense ture the questions in the Design Paper.5 In each of depth. Other configurations were considered, country, priority themes as well as sectors were such as three CDF pilot and three non-CDF identified (for example, public sector reform, pilot, or three PRSP and three non-PRSP coun- health, decentralization). Interviews were con- tries. However, the first alternative would have ducted in country capitals and selected districts, reduced the weight given to learning from the municipalities, and project sites. The evaluation CDF pilot experiences, and the second would teams typically included a representative from have shifted the balance toward countries where OED, and a combination of national and inter- aid plays a smaller role relative to other sources national consultants. In some cases, such as of development finance. In the end, given that Vietnam, additional consultants and agency staff 88 were provided by interested donors (for example, sample of local government officials, civil soci- Japan MOFA/JBIC, ADB). DFID provided an ety representatives, and the private sector in agency staff member for the Uganda team. four districts of varying socioeconomic levels, and got a response rate of roughly 75 percent. In In sum, the evaluation was carried out in "CDF Romania the response rate was nearly 100 per- fashion" and involved a wide a range of stake- cent for a survey administered to 722 experts holders engaged in development work (from the and decisionmakers throughout the country policy level down to implementation of proj- who deal with development issues. In Bolivia ects). These included representatives from detailed questions were sent to international national and local governments, donor agen- agencies before the start of the evaluation. In cies, mass organizations, the private sector, Uganda, a survey was administered in a group national NGOs, legislatures, academics, civil meeting of stakeholders from government, civil society organizations, and international NGOs. society, and the private sector. An analysis of The evaluation teams were to have met period- the 33 completed questionnaires formed the ically with the country reference group and basis for subsequent workshops with parliamen- share preliminary observations with donors and tarians and questionnaire respondents. national stakeholders in a closing workshop. Structured Interviews were conducted with representatives from national and local govern- Types of Activities ments, donor agencies, mass organizations, pri- vate sector leaders, national NGOs, legislatures, A variety of activities and evaluation tools, academia, civil society organizations, and inter- listed below, were employed during the course of national NGOs. Between 70 and 84 interviews the evaluation. (group and individual) were conducted for each country, resulting in an average of roughly 145 Literature reviews. Teams assembled and ana- people interviewed per country. Lists of people lyzed pertinent literature, documents, and data interviewed are included as annexes in each sources, and key country reports were posted on case study. the CDF evaluation Web site. Focus Groups were organized based on sectors, Questionnaire Surveys of key stakeholders. themes (e.g., health, institutional reform), and When possible these were designed in-country professional affiliation (e.g., ministry staff, pri- and administered by local institutions. In vate sector, church). Vietnam, a local consulting firm distributed sur- veys to 290 Vietnamese and expatriate develop- Field Trips to selected districts, municipalities, ment practitioners about how they perceived and project sites were made in all countries and changes in ODA management with reference to included interviews with local government offi- the CDF principles. One-hundred-and-seven cials and politicians, project managers, opera- people responded (of whom three-quarters were tional NGOs and donor agencies, and municipal Vietnamese from government and nonstate or district council members. agencies, and one-quarter were expatriates from donor agencies and international NGOs). In Closing Workshops were held with key stake- Burkina Faso, a survey was administered to a holders and decisionmakers to discuss the team's preliminary findings. These workshops between May and September 2002, and a total 89 also served to draw out the policy and program- of 26 donors (17 bilateral and 9 multilateral) matic implications of the findings. across the 5 country cases responded. This survey consisted of a total of 29 questions A Study of Transaction Costs was carried out that sought information on composition of in Bolivia, Ghana, Romania, Uganda, and donor portfolio, donor coordination, donor Vietnam as a specific parallel exercise to track administrative and procedural requirements, changes and trends in government-donor part- and donor implementation practices. Local nerships. The final report identifies, and to the consultants administered the surveys in Bolivia, extent possible quantifies, the impact of donor Romania, Uganda, and Vietnam. In Ghana practices and procedural requirements on gov- and Romania, the surveys were followed up by ernments and the transaction costs they imply. targeted interviews with donors and key gov- A Transaction Costs Survey was administered ernment officials. Endnotes 91 Introduction 1. CDF Evaluation Secretariat, OED. 2001. "Design Paper for a Multi-Partner Evaluation of the Comprehensive Development Framework," pp. 10-11. Chapter 2 2. There are several reasons Bolivia subsequently became a case study country contributing to the current CDF eval- uation. These include the interest of Bolivian authorities, the country's development of a Poverty Reduction Strategy Paper under the Heavily Indebted Poor Countries (HIPC) initiative, and the significant role international aid plays in the country. Chapter 5 3. Romania was formally invited at the NATO Prague Summit in November 2002 to participate in accession talks, expected to result in Romania joining NATO in 2004. Annex 4. GAO, Program Evaluation and Methodology Division. 1990. Case Study Evaluation, Chapter 3, "Design Decisions in Case Study Methods." Transfer Paper 10.1.9. Washington, D.C.: GPO. 5. CDF Secretariat, OED. 2001. "Design Paper for a Multi Partner Evaluation of the Comprehensive Development Framework." < www.worldbank.org/evaluation/cdf>