Report No. 51880-BI Report No. 51880-BI Republic of Burundi Country Economic Memorandum (CEM)
Public Disclosure Authorized The Challenge of Achieving Stable and Shared Growth
March 2011
Poverty Reduction and Economic Management 3 Africa Region Republic of Burundi Public Disclosure Authorized Country Economic Memorandum (CEM) Public Disclosure Authorized
Document of the World Bank, Co-produced with the Government of Burundi, the African Development Bank and the Department for International Development (UK) Public Disclosure Authorized Currency Equivalents Exchange Rate Effective as of February 28, 2011
Currency Unit Burundi Franc USS 1.00 FBu 1,230
Government Fiscal Year January 1–December 31
Weights and Measures Metric System
Abbreviations and Acronyms
AfDB African Development Bank BNDE Banque Nationale de Développement Économique BRB Banque de la République du Burundi CEM Country Economic Memorandum CET Common External Tariff COMEBU Comptoirs Miniers des Exploitations du Burundi COMESA Common Market for Eastern and Southern Africa COMTRADE U.N. Commodity Trade Statistics Database CPAF Common Performance Assessment Framework DAC Development Assistance Committee DDR Disarmament, Demobilization and Reintegration DfID Department for International Development (UK) DPAF Donor Performance Assessment Framework EAC East African Community EDPRS Economic Development Poverty Reduction Strategy EITI Extractive Industries Transparency Initiative EPA Economic Partnership Agreement EU European Union FDI Foreign Direct Investment FSAP Financial Sector Assessment Program GDP Gross Domestic Product GWh Gigawatt Hours HI Herfindahl Index HIPC Heavily Indebted Poor Countries HIV/AIDS Human Immune Deficiency Virus/Acquired Immune Deficiency Syndrome JBSR Joint Budget Support Review MDG Millennium Development Goal
ii MEACA Ministry of East African Community Affairs MFI Microfinance Institution MTEF Medium-Term Expenditure Framework NTB Nontariff Barrier OCIBU Office du Café du Burundi OECD Organisation for Economic Co-operation and Development PAGE Project d’Appui à la Gestion Economique PEMFAR Public Expenditure Management and Financial Accountability Review PPP Public Private Partnership PRSP Poverty Reduction Strategy Paper RCA Revealed Comparative Advantage REGIDESO Régie de Production et de Distribution d'Eau et d'Electricité SITC Standard International Trade Classification SODECO Société de Déparchage et de Conditionnement SOGESTAL Société de Gestion des Stations de Lavage TFP Total Factor Productivity VAT Value-Added Tax
Vice President: Obiageli Katryn Ezekwesili Country Director: John Murray McIntire Country Manager: Mercy Miyang Tembon Sector Director: Marcelo Giugale Sector Manager: Jan Walliser Task Team Leader: Hannah Nielsen
iii
Table of Contents
ACKNOWLEDGMENTS ...... vii
EXECUTIVE SUMMARY ...... viii
INTRODUCTION ...... 1
CHAPTER 1 MACROECONOMIC AND FISCAL DEVELOPMENTS...... 2 A. Economic Growth: Historical Perspective ...... 2 B. Burundi’s Public Finances: An Overview ...... 11
CHAPTER 2 OBSTACLES TO ECONOMIC GROWTH AND EMERGING OPPORTUNITIES FOR BROADENING THE BASE FOR GROWTH ...... 15 A. Obstacles to Economic Growth ...... 15 B. Opportunities in Potential Growth Sectors ...... 32
CHAPTER 3 THE GOVERNMENT’S OPTION TO TACKLE OBSTACLES AND REAP OPPORTUNITIES FOR ECONOMIC GROWTH ...... 52 A. An Infrastructure Action Plan ...... 52 B. Maximizing the Benefits of Burundi’s EAC Membership ...... 59 C. Reforms Needed to Increase the Private Sector’s Contribution to Growth ...... 70 D. Encouraging Emerging Growth Sectors ...... 74 E. Summary of Priority Recommendations ...... 81
CHAPTER 4 IMPLICATIONS FOR THE MACRO-FISCAL SITUATION ...... 84 A. Economic Prospects, Resource Allocation, and Achieving the MDGs ...... 85 B. How Can Aid Contribute to Increased Growth? ...... 91 C. Conclusion ...... 96
APPENDIX ...... 97 Appendix 1: Action Plan ...... 98 Appendix 2: Export Diversification ...... 101 Appendix 3: Regional Integration ...... 105 Appendix 4: Infrastructure Action Plan–Tables and Maps ...... 109 Appendix 5: Infrastructure Action Plan–Main Components ...... 113 Appendix 6: Business Environment ...... 118 Appendix 7: Map of Burundi ...... 119
REFERENCES ...... 120
iv List of Boxes Box 1.1: Rwanda’s Growth Experience and Possible Lessons for Burundi ...... 10 Box 2.1: The Transformation of Uganda’s Exports—A Success Story ...... 21 Box 2.2: Burundi’s Tax Regime ...... 26 Box 2.3: Land Tenure and Economic Growth ...... 35 Box 3.1: Proposed Measures to Remove Burundi NTBs within the EAC ...... 63 Box 4.1: The Common Performance Assessment Framework in Rwanda ...... 95 Box A.1: Potential Benefits, Costs, and Guiding Principles of Regional Integration ...... 106
List of Figures Figure ES.1: Real GDP Growth and GDP Per Capita, 1962–2008 ...... viii Figure ES.2: Composition of Public Spending, 2001–08...... xiv Figure 1.1: Per Capita GDP in Burundi and Rwanda, 1962–2008 ...... 2 Figure 1.2: Real GDP Growth and GDP Per Capita, 1962–2008 ...... 3 Figure 1.3: The Cost Of War: Potential and Actual GDP Per Capita ...... 4 Figure 1.4: Sectoral Contributions to Real GDP Growth, 1997–2008 ...... 5 Figure 1.5: Private and Public Investment as a Percent of GDP, 1997–2008 ...... 6 Figure 1.6: Annual Growth of GDP, Capital Stock, and Productivity Per Worker, 1990–2008 ...... 7 Figure 1.7: Real GDP Growth during Post-conflict Periods in Burundi, Rwanda, and Sierra Leone ...... 8 Figure 1.8: Composition of Public Spending, 2001–08...... 12 Figure 2.1: HI for EAC Member-Countries ...... 19 Figure 2.2: HI and GDP Per Capita ...... 19 Figure 2.3: Composition of Exports in EAC Countries, 1985–89 and 2005–06 ...... 20 Figure 2.4: EXPY of EAC Countries ...... 22 Figure 2.5: Governance Indicators, 1998, 2000, 2003, and 2008 ...... 28 Figure 2.6: Population Projections, 2005–50 ...... 31 Figure 2.7: Proportion of Individuals Living in Households with a Daily Caloric Intake per Adult Below 1,900 kcal ...... 33 Figure 2.8: Food Crop Production, 1990–2007 ...... 34 Figure 2.9: Farm-Gate Prices for EAC Members, 1977–2008 ...... 36 Figure 2.10: Burundi Coffee Production, 1977–2008 ...... 37 Figure 2.11: Medium-Contribution Scenario—Government Revenue ...... 51 Figure 2.12: High-Contribution Scenario—Government Revenue ...... 51 Figure 3.1: Funding Arrangements for the Core Infrastructure Program ...... 53 Figure 4.1: Selected MDG Trends under the Baseline Scenario, 2006–15 ...... 90 Figure 4.2: Selected MDG Trends under the Alternative Scenario, 2006–15 ...... 90 Figure 4.3: Allocation of Aid by Impact Category, 2001–07 ...... 92 Figure 4.4: Disbursement Ratio and Commitment-Disbursement Gap, 2001–07 ...... 93 Figure 4.5: Exchange Rate and Price Developments, 2001–08 ...... 94 Figure A.1: Product Space Map of Burundi ...... 101 Figure A.2: Transport Corridors for the East African Community ...... 111 Figure A.3: Railway Network of the East African Community ...... 112
List of Tables Table ES.1: Priority Recommendations ...... xv Table 1.1: Selected Economic Indicators, 1962–2008 ...... 3 Table 1.2: Supply-Side Components, 1997–2008 ...... 5 Table 1.3: Demand-Side Components, 1997–2008 ...... 6 Table 1.4: Comparison of Selected Indicators Six Years Post-conflict ...... 9
v Table 1.5: Composition of Government Revenue, 2001–08 ...... 11 Table 1.6: Central Government Fiscal Deficit, 2001–08 ...... 12 Table 2.1: Selected Indicators Comparing Burundi to Sub-Saharan Africa and the World Economy . 15 Table 2.2: Classification Summary of Burundi’s Exports ...... 23 Table 2.3: Sectoral Distribution of Bank Credit ...... 25 Table 2.4: Annual Projections with One Large-Scale Nickel or Gold Mine ...... 47 Table 2.5: Comparison of Annual Official Exports and Survey Production, 2008 ...... 49 Table 3.1: Selected Indicators for the Base Case Scenario ...... 55 Table 3.2: Key Outcomes for the Base Case and Alternative Scenarios ...... 56 Table 3.3: Selected Socioeconomic Indicators for EAC Members, 2008 ...... 60 Table 3.4: External Tariffs in the EAC, 2000–08 ...... 60 Table 3.5: Similarity and Complementarity Indexes for Burundi, 1995–2007 Averages ...... 65 Table 3.6: Specific Measures to Improve the Performance of the Food Crops Sector ...... 75 Table 4.1: Baseline Scenario—Projections for the Main Economic Variables, 2009–15 ...... 85 Table 4.2: Alternative Scenario—Projections for the Main Economic Variables, 2009–15 ...... 86 Table 4.3: Baseline Scenario—Actual, Estimated, and Projected Government Finance, 2002–15 ...... 87 Table 4.4: Baseline Scenario—Medium-Term Expenditure Framework, 2007–15 ...... 88 Table 4.5: Alternative Scenario—Actual, Estimated, and Projected Government Finance, 2002–15 .. 89 Table 4.6: Alternative Scenario—Medium-Term Expenditure Framework, 2007–15...... 89 Table A.1: Action Plan Priority Recommendations for the Time Period 2010–15 ...... 98 Table A.2: Detailed Classification of Burundi’s Exports...... 102 Table A.3: Horticulture: Markets, Constraints, and Opportunities for Burundi ...... 103 Table A.4: Recommendations for the Artisanal and Small-Scale Mining Sector ...... 104 Table A.5: Burundi’s Post-Independence Experience with Regional Integration Agreements ...... 105 Table A.6: Matrix of Strategic Objectives and Indicative Outputs for the MEACA ...... 107 Table A.7: Policy Matrix for Advancing Regional Integration in Burundi ...... 108 Table A.8: Basic Infrastructure Coverage for the East African Community, 2006 ...... 109 Table A.9: Service Costs and Difficulties in the East African Community ...... 109 Table A.10: Description of Various Scenarios Considered in the Report ...... 110 Table A.11: Development Expenditures on the Core Program ...... 116 Table A.12: Routine Maintenance Expenditures on the Core Program ...... 116 Table A.13: Recommendations for Improving Financial Sector Performance ...... 118
vi
ACKNOWLEDGMENTS This Burundi Country Economic Memorandum (CEM) is the first since 1984. The CEM reviews developments in Burundi over the past 10 years and identifies the tightest constraints to economic growth. It draws on studies prepared by the government of Burundi, the World Bank, other partners, and academics. After synthesizing recommendations from those studies and from original analysis, the CEM presents a strategy to promote growth, reduce poverty, and improve peoples’ lives. The Burundi Country Economic Memorandum (CEM) is a joint report from the World Bank, the government of Burundi, the African Development Bank, and the U.K. Department for International Development. The team would like to thank Her Excellency Madame Clotilde Nizigama (Minister of Economy, Finance, and Development Cooperation, Burundi) and the government steering committee for the CEM for their excellent collaboration throughout the preparation of the CEM. The report was prepared under the overall leadership and guidance of Jan Walliser (Sector Manager, AFTP3) and Eric Bell (Lead Economist, AFTP3), who provided invaluable support, input, and advice during the preparation of the memorandum. The team also greatly benefited from the support and guidance of John Murray McIntire (Country Director, AFCE1), Mercy Miyang Tembon (Country Manager, Burundi), Kathryn Funk (Country Program Coordinator, AFCTZ), and Steffi Stallmeister (Senior Country Officer, AFCTZ). The former task team leader Dorsati Madani (Senior Economist, CICSA) conceptualized the CEM, established the core team, launched parts of the analytical work and led early discussions with the counterparts. The main author and task team leader of the report is Hannah Nielsen (Economist, AFTP3). The core CEM team includes Eric Mabushi (Economist, AFTP3), Jean-Pascal Nganou (Economist, AFTP3), Edgardo Favaro (Lead Economist, PRMED), Henry Mooney (Economist, PRMED), Christian Lim (Economist, African Development Bank), Chiara Selvetti (Economist, DfID), Tania Rajadel (Consultant, AFTP3), Nicaise Ehoue (Senior Agriculture Economist, AFTAR), Remi Pelon (Mining Specialist, COCPO), Gilbert Midende (Consultant, AFTP3), John May (Lead Population Specialist, AFTHE), Caroline Freund (Lead Economist, DECRG), Andre Ryba (Consultant, AFTFW), Aurelien Beko (Consultant, AFTP3), and Daniel Benitez (Senior Economist, FEUSE). Valuable contributions were also provided by Kene Ezemenari (Senior Economist, AFTP3), Vandana Chandra (Senior Economist, DECOS), Israel Osorio-Rodarte (Consultant, DECPG), Susana Carrillo (Senior Governance Specialist, WBIGV), Craig Andrews (Lead Mining Specialist, COCPO), Charles N'cho-Oguie (Consultant, AFTP3) and Kalamogo Coulibaly (Consultant, AFTP3). Additionally, the team appreciates very useful comments from Lev Freinkman (Lead Economist, AFTP3). The core team worked closely with a steering committee consisting of representatives of various ministries. The counterpart team included the following members: Domitien Ndihokubwayo, Juvénal Bumviye, Alexis Bizimungu, Emile Sinzumunsi, Melchior Barantandikiye, Mireille Bizimana, Rose Kamariza, Adèle Mbonankira, Cyriaque Miburo, Willy Ntamagara, Réverien Nivyayo, Léopold Bizindavyi, Terence Ntabangana, Hilaire Ntakiyica, Fidèle Gahungu, Charles Rugema, Bonaventure Sota and Pierre-Claver Rurakamvye. Additional workshop participants included Léopold Nkunzimana, Gratien Ninteretse, Pierre Bayihishako, Anaclet Birushabagabo, Balthazar Nakumana, Florence Nshimirimana, Jean-Bosco Batungwanayo, Jean Mugishawimana, Consolate Musamisomi, Nestor Niyungeko, Cyprien Gicemzi, and M. Matsebahene. The CEM benefited immensely from very helpful comments by the peer reviewers for the overall report, Hinh Dinh (Lead Economist, DECOS) and Peter Moll (Senior Economist, OPCCE); the reviewers for the regional integration component, Diep Nguyen-Van Houtte (Senior Operations Officer, AFCRI) and Paul Brenton (Lead Economist, AFTPM). Comments and suggestions from outside the World Bank are also gratefully acknowledged, in particular, from the Burundi International Monetary Fund team. The team would also like to extend its gratitude to Mariama Daifour Ba (Program Assistant, AFTP3) and Aurore Simbananiye (Team Assistant, AFTBI) for logistics support. Resources received from the Belgian Poverty Reduction Partnership Trust Fund to finance a background study on artisanal and small-scale mining in Burundi are also greatly appreciated.
vii
EXECUTIVE SUMMARY
1. Burundi’s growth policy should focus on: (1) closing the infrastructure gap; (2) gaining from regional integration; (3) improving the business environment; (4) promoting new growth sources; and (5) strengthening its fiscal position through improved revenue mobilization, more efficient spending, better public financial management, and more effective aid.
BURUNDI’S ECONOMY YESTERDAY AND TODAY
2. Burundi is very poor and has suffered from years of conflict. The 1984 World Bank CEM said: “Burundi is among the poorest countries in the world; [for a population of 4.2 million] per capita GNP is estimated at about US$240 (1983); three-fourths of the adult population is illiterate; infant mortality is still high, despite considerable progress registered in the last twenty years; and access to potable water and electricity is limited to the urban centers -- 5% of the population.”
3. The situation has worsened since 1984. The population is now around 8.4 million. GDP per capita is about US$110 (constant 2000 prices); a significant fall from the level estimated 25 years ago, making Burundi the second-poorest country in the world. Starting from an already low developmental level, Burundi was ravaged by a civil war from 1993 to 2005. War killed some 300,000 Burundians, displaced many others, destroyed capital, repressed investment, and damaged the State’s capacity to provide basic health, education, water and electricity. A simple estimate of the cost of the war indicates that, without the conflict, Burundi’s GDP per capita would be about double its current level.
4. A modest recovery has started. Figure ES.1: Real GDP Growth and GDP per Capita, 1962–2008 Real annual GDP growth has stabilized at 3.5 percent since 2005, though this rate is 25% 180 160 below the average of the East African 20% Community (EAC). Strengthened 140 macroeconomic policies have added 15% 120 modestly to growth since 2005, although the global crisis lowered growth in 2009. 10% 100 Monetary policy is geared toward 5% 80 stabilizing prices and succeeded in 60 0%
keeping inflation on average in the single Constant 2000 US$ digits between 2005 and 2007. Inflation, 40 5% however, increased substantially in 2008 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 20 because of food and fuel price shocks 10% Real GDP growth rate (left hand side) 0 before returning to estimated 10.5 percent GDP per capita (constant 2000 US$, right hand side) in 2009. Tax revenue has stabilized Source: World Development Indicators. between 17 and 18 percent of GDP, despite the narrow tax base and a reduction of the petroleum tax. Another indication of political stabilization and recovery of the country is the return of more than half a million refugees since 2002.
5. Burundi is not likely to meet many of the Millennium Development Goals by 2015. Burundi lags behind the world and East Africa on all measures of health and human capital. An estimated 67 percent of Burundians live below the poverty line. The country ranked toward the bottom of the Human Development Index in 2008. Lack of education among adults and children limits productivity and growth.
viii Although primary-school enrollment has improved significantly in recent years, completion rates are low, and secondary and tertiary enrollment rates have not grown apace with primary completion.
6. Compared to other post-conflict countries, however, Burundi has not experienced the typical growth spurt. As section A3 in chapter 1 shows, Burundi’s growth in the post-conflict period does not compare well with other post-conflict countries, such as Rwanda and Sierra Leone. This is primarily because of the long conflict that continued even after the official peace agreement of 2000. Investment, output and productivity growth have been less than in comparable post-conflict countries because Burundians lack confidence in their country’s economic and political stability.
7. The government has begun reforms. The Government has begun some reforms addressing, for example, coffee liberalization, the business environment, and public financial management. Key reforms are the ongoing liberalization of coffee, a new investment code, revision of the mining code, the adoption of a new privatization law, the establishment of the new Investment Promotion Agency (API) and better public financial management based on a new organic budget law. Critical next steps are adopting the revised mining code, adopting the revised commercial code, and finalizing the revision of the tax code. Regarding public finance reforms, the size of the wage bill remains a concern.
WHAT BLOCKS ECONOMIC GROWTH IN BURUNDI?
8. Burundi is mainly rural and subsistence agriculture dominates. Agriculture contributes about half of GDP and employs approximately 90 percent of the labor force, as detailed in section B1, chapter 2. Poverty is widespread in rural areas. Agricultural productivity is low and has barely improved since independence. Given high population growth, low productivity gains have led to a decline in production per capita. Malnutrition is common and Burundi often requires food aid.
9. Burundi’s exports are undiversified, consisting mainly of low-value, unprocessed primary products. Export concentration on a few agricultural exports makes Burundi vulnerable to external price or quantity shocks and to bad weather. Section A4 of chapter 2 demonstrates that, compared to most of the other EAC members as well as some countries that previously exported mostly coffee, Burundi has not diversified its exports and this fact has made growth more volatile.
10. The expansion of Burundi’s main export, coffee, faces severe challenges. Some 600,000– 800,000 households (perhaps one-third of the population) grow coffee. Though Burundi has the agricultural conditions to produce high-quality, high-value coffee, the sector underperforms with declining production and quality partly because of official prices that do not stimulate production (see section B2, chapter 2). Unlike Rwanda, Burundi has been slow to match the latest developments, such as the increased importance of specialty coffees or the marketing of high-quality coffee through direct sales instead of auctions. Government policy has not always been beneficial to Burundian coffee growers. The ongoing reform, especially the privatization of the coffee washing stations, is a positive step.
11. Lack of infrastructure raises the costs of Burundi’s isolation. The poor coverage and state of infrastructure create costs in time and money that lower the return to work, discourage domestic and foreign investment, and constrain economic growth, as outlined in section A2, chapter 2. High transport costs, caused by absence of infrastructure, hinder internal trade and reduce Burundi’s trade opportunities with East Africa and the world beyond. The internal obstacles are compounded by the high costs in the ports of Dar es Salaam (Tanzania) and Mombasa (Kenya), through which Burundi trades. Modern electricity, which is available only to 2 percent of households, is expensive.
12. The private contribution to growth has been too small. The public sector’s share in investment has been too high, averaging 74 percent of total investment between 2001 and 2004. The main reason is
ix that Burundi has an unfavorable business environment on top of little infrastructure, one that discourages investment, activity diversification, and job creation. Domestic private investment has only recently started to pick up and remains low, averaging about 8 percent of GDP between 2005 and 2008 and almost half of total investment. Foreign direct investment is less than 1 percent of GDP.
13. Finance is a major obstacle for formal and informal business. Only 2 percent of Burundians have bank accounts and only 4 percent are members of microfinance institutions (MFIs). Financial products are undiversified and do not cater adequately to the rural population or small and medium-size enterprises (see section A5, chapter 2). Given the lack of access to commercial banking, microfinance has an important role to play; however, nearly 80 percent of members/customers of MFIs are wage earners. The industry, therefore, excludes a majority of the population, namely non-wage earners, while serving a market that is uncertain for it in the medium term as it can also easily be served by the banking industry.
14. Burundi’s population is expected to continue to grow rapidly in the next 40 years, as illustrated in section A8, chapter 2. Its high population growth rate (2.6 percent in 2005) exacerbates many of the problems of a country that already has one of the highest population densities in Africa. The agriculture sector alone cannot support the growing needs of the population, underscoring the need to create urban and rural non-farm employment.
15. Poor governance and limited institutional capacity imply lower productivity for all sectors of the economy and inhibit new investment in physical and human capital. Given the importance of government spending as a share of GDP in the post-conflict economy, increasing the efficiency of that spending is vital. Section A6 in chapter 2 describes how poor governance is currently an obstacle to more efficient public spending. In addition, the government’s capacity to adopt quick and effective policy decisions is weak. This fosters a perception that recovery is weak and causes delays in private sector involvement.
HOW CAN PUBLIC ACTIONS INCREASE ECONOMIC GROWTH?
16. The CEM has identified priorities for public actions to promote growth and to make it more equitable. Table ES.1 presents the key recommendations which are linked to the relevant text of the chapters of the main report. Table A.1 in Appendix 1 includes the agency responsible for the implementation of the recommendation.
17. Burundi’s growth policy should focus on: (1) closing the infrastructure gap; (2) gaining from regional integration; (3) improving the business environment; (4) promoting new growth sources; and (5) strengthening the fiscal position through improved revenue mobilization, more efficient spending, better public financial management, and more effective aid.
18. Peace and stability are indispensable for growth but will not produce a dividend without proper management by the Government. The Burundian Government has a larger than usual role in economic development because it must make up the investment lost during the conflicts and because the private sector is still weak. The government’s main job is to keep the peace and maintain political stability, without which public and private investment cannot grow. However, the Government should also realize the peace dividend and benefit from reduced and redirected defense spending. The Government therefore faces the political and fiscal choice between spending on security and the need to invest more in infrastructure and in human capital. To the extent possible, and keeping in mind the need for security and political stability, it is essential to reduce the share of defense and security expenditures as soon as possible to create fiscal space for increased spending in the priority economic and social sectors.
x 19. As part of a policy for stability for economic growth, Burundi must lower its population growth rate. Lower fertility rates would affect most development areas positively; and would particularly facilitate the formation of human capital, help reduce poverty levels and land shortages, relieve pressure on the environment, and improve the health of women and children.
Closing the Infrastructure Gap
20. An infrastructure action plan lays out a detailed strategy for improving the infrastructure in Burundi, taking into account the regional context1. It focuses on the power, transport, and communications sectors; and proposes actions that need to be taken, their timing, and a possible funding structure. The action plan could have a significant positive impact on the whole economy, leading to sustained growth, business, and employment opportunities; more and cheaper infrastructure; and increased tax revenues. The main features, objectives, and the economic impact of the action plan are described in chapter 3, section A.
21. Complementary policies to investment in infrastructure are critical to getting the full return on private investment. Priority interventions for the short and medium term are presented in section A5, chapter 3. Specific key recommendations are:
• Electricity: Electricity supply must be improved by using all possible energy resources, with more provinces connected to the network and REGIDESO financially restructured. • Transport: Given the geographic isolation of Burundi, the country could greatly benefit from rehabilitating the national highways and the provincial and community roads. Strengthening the role of civil aviation could attract international airlines and increase the availability of cargo space. Rail is also important to Burundi but it can do nothing about it for now.
Gaining from Regional Integration
22. Burundi should benefit from membership in the EAC. As described in section B, chapter 3, joining the EAC in 2007 opened opportunities for trade in goods and services, for developing physical and regulatory infrastructure linking Burundi to a larger regional market and to the rest of the world, and for development of services provided by the regional organization by pooling resources from all member- countries. Although there is a risk of trade diversion, the larger economic zone would open opportunities for the production of new goods and services currently not visible in the trade accounts. The ongoing regional integration process can also create a push for economic and structural reforms and can strengthen political and regional stability.
23. Regional infrastructure is vital yet Burundi can do little about it. Burundi depends on Dar es Salaam and Mombasa for most of its external trade. Efficient port management is critical to improving competitiveness of products coming from the region as is the network of roads, railroads, and bridges that connects these final destinations to Burundi. EAC membership does not guarantee progress, but it gives Burundi a forum to advocate for better infrastructure management in regional ports, rails, and roads that it does not operate directly.
24. A comprehensive approach to regional integration, prioritizing and sequencing actions needs to be implemented, most importantly addressing institutional constraints and economic reforms. Key recommendations are the following (see section B4, chapter 3):
1 Developed by the African Development Bank.
xi • Institutional reforms: It is essential that the Ministry of East African Community Affairs (MEACA) is staffed with competent personnel, and the necessary institutional arrangements need to be in place, requiring extensive capacity building. It is important to implement a coordination, monitoring, and evaluation mechanism for all EAC-related issues. Furthermore, a communication campaign aimed at sensitizing the population to support the integration initiative should be implemented. • Economic reforms: Eliminating all remaining nontariff barriers is critical. Furthermore, a strong framework for public-private dialogue is advisable.
Improving the Business Environment
25. The private economy must grow more. The private sector faces constraints in law, infrastructure, finance, and regulation, plus the tax of bad governance. One critical supply constraint is finance. A Financial Sector Assessment Program (2009) and an Investment Climate Assessment (World Bank 2008a) and the Doing Business Indicators show that Burundi’s business climate discourages investment. Priority recommendations, as given in chapter 3, section C, are the following:
• Access to finance in rural areas must improve. The capacity of the central bank (Banque de la République du Burundi – BRB) to supervise microfinance institutions needs to be strengthened. New MFI regulations should allow them to develop new products. Similarly, technical assistance should be provided to banks to develop new products and lending mechanisms to better meet the needs of small and medium-size enterprises and of rural areas. • The regulatory environment should be improved to attract domestic and foreign investment. Critical components are the finalization of the tax code revision, and the finalization and adoption of the revised mining code. The new investment code includes some of the best practices in investment legislation, among them investor protection and the freedom to transfer capital and dividends. The amended tax code includes tax incentives mentioned under the revised investment code. The revised mining code is meant to prepare the country for potential large- scale investments and clarify the role of artisanal and small-scale mining. • Increase competition and improved governance. Competition could be increased by finalizing and adopting the revised Privatization Law. Adopting and implementing the national policy of good governance and its action plan would demonstrate the recognized importance of improved governance.
Promoting New Sources of Growth
26. The strategy for new sources of growth has two parts. The first is to address the weak performance of agriculture, including both food and export crops. It should be the priority to eliminate widespread food insecurity and malnutrition. Second, in the medium term, diversifying the economy and strengthening exports could lessen the vulnerability to external shocks and create new opportunities for economic development and employment (especially off-farm employment). Revenues from current exports - mainly coffee - can be stabilized and increased through better supply policies, as described in chapter 3, section D.
27. Given the share of agriculture in employment and land use, addressing the causes of low agricultural productivity is of utmost importance. As the Sources of Rural Growth Study (Baghdadli, Harborne, and Rajadel 2008), the National Agricultural Strategy of Burundi (Republic of Burundi 2008b), and an analysis of the coffee sector show, low agricultural productivity of food and cash crops is the main obstacles to growth after low domestic investment. As a consequence, Burundi still relies on food aid, has almost no private external resources (besides aid), and is vulnerable to external shocks. Regarding food
xii crops, the main constraints are the poor economic incentives for farmers to introduce modern inputs, the poor physical infrastructure and extension services supporting agriculture, the weak market organization that is heavily distorted by government intervention, and the low development of farmer associations and cooperatives. With regard to cash crops, the liberalization of agricultural prices and the definition of a subsidiary role for the public sector in the production and commercialization of coffee and tea (in particular, the ongoing privatization of washing stations) are steps in the right direction. However, the speed of implementation of these programs could be accelerated.
28. Key recommendations include the following:
• Food crops: Critical for increasing production and productivity is strengthening input distribution systems to supply more inputs to small farmers; rehabilitating or modernizing production tools and infrastructure (including rural roads); and organizing, structuring, and professionalizing producers. • Coffee: Most important is the complete implementation of the ongoing privatization agenda. Next steps include strengthening the capacity of the regulatory agency (ARFIC) and the interprofessional association (INTERCAFE), facilitating access to finance, and improving processing practices.
29. Diversification out of subsistence agriculture is particularly important. Given demographic trends and the scarcity of land, Burundi must move away from subsistence agriculture because its income per unit of land is too low. Although the transformation away from subsistence agriculture will not be achieved for many years, the move should begin now based on a detailed analysis of potential products and markets. The analysis presented in chapter 2, section A4, shows that Burundi’s products are not yet internationally competitive. Nonetheless, the CEM identifies sectors and products that have great export potential and are in line with Burundi’s Poverty Reduction Strategy Paper (PRSP) and National Agricultural Strategy. Key interventions to facilitate diversification and the increased contribution of other sectors, such as horticulture and fishing, are the availability of cargo space to export fresh produce and the compliance with sanitary and phytosanitary standards.
30. Mining can contribute more to Burundi’s economy. Mining is done by artisanal and small- scale miners, employing about 20,000 people in low productivity jobs and generating little tax income. Industrial mining could boost value added, tax receipts, employment, and local supply of intermediate goods, especially food and construction materials. There is currently no commercial mining because the war stopped it.
31. To enable the mining sector to play an important role in Burundi’s economy, several reforms are necessary. Given the potential contribution of the mining sector, a sectoral strategy accounting for the characteristics and constraints of the available mineral resources is essential, as outlined in section D4, chapter 3. The government has only recently started to reform the legal framework for mining and a few international companies have shown interest. Burundi could greatly benefit from joining the Extractive Industries Transparency Initiative and from using available facilities for assistance in negotiations with international companies. To account for the large share of artisanal and small-scale mining, a unit could be created within the Ministry of Energy and Mines to deal specifically with this subsector. Support for the formalization of artisanal mining is also advisable because it would raise productivity and allow small miners to capture more value-added.
xiii Expanding Fiscal Space
32. The government has to expand fiscal space. As outlined in section A of chapter 4, Burundi’s fiscal balance is expected to improve moderately in the next 5 years, but domestic resources will remain limited and external resources are needed to finance the fiscal deficit. The fiscal position needs to be strengthened by improving revenue mobilization, implementing prudent spending policies, and reforming public financial management. The Public Expenditure Management and Financial Accountability Review (PEMFAR) and an analysis of the medium-term framework demonstrate the need for improved allocation of government resources. A key recommendation is that defense and security spending be reduced to the extent possible while ensuring sufficient resources for the security of the country, as noted above. At the same time, the public sector wage bill needs to be controlled. Resources made available by these measures must be channeled to priority sectors and pro-poor expenditure.
Figure ES.2: Composition of Public Spending, 2001–08 33. Burundi could improve its fiscal stance and direct additional resources toward priority 50 sectors. Although the authorities have made a significant effort to improve the country’s fiscal 40 stance, increased and inefficient spending as well 30 as inadequate revenue mobilization could result in larger, unsustainable fiscal deficits. High defense 20 and security spending and the high wage bill are percent of GDP 10 the main obstacles to reallocating public resources toward priority economic and social sectors. 0 2001 2002 2003 2004 2005 2006 2007 2008 34. Making aid more effective. To promote Military and security expenditure Other current expenditure Capital expenditure Exceptional expenditure growth, external resources could be shifted from Source: International Monetary Fund. Note: Exceptional relief and humanitarian aid to investments in expenditure includes spending on DDR and on elections. infrastructure and human capital. An analysis of recent aid disbursements in chapter 4 shows a declining trend in humanitarian and emergency aid, which is consistent with the improved security in the country. Resources freed should be shifted to production-oriented sectors with more growth potential, especially agriculture and infrastructure. This shift does not imply that spending on social sectors should be reduced; to the contrary, investments in these areas are crucial for the long-term development of the country.
35. Aid management of aid must change to promote growth. Because of the conflict and post- conflict status of the country, aid has been primarily allocated to humanitarian and emergency projects; production-oriented sectors, such as agriculture and infrastructure, have received less support than they would have. The unpredictability of aid flows has exacerbated the spending bias toward government consumption and away from investment in production-oriented sectors and in human capital, as shown in section B, chapter 4. Improved donor coordination will facilitate the allocation of aid; in that regard, a common performance assessment framework to make aid more effective must be agreed.
xiv Table ES.1: Priority Recommendations Area Recommendation Time frame Approximate cost Cost as share of total (2010–15) (US$ millions) expenditure projected for 2010 15 Closing the infrastructure gap Electricity Develop Burundi’s domestic hydroelectric potential (Kaganuzi, Mpanda, Kabu 16) Short to medium term 80 1.3% Provide 110 kV lines with necessary substations to all provincial capitals Short to medium term 60 1.0% Finalize the financial restructuring of REGIDESO Short term to be determined Transport Rehabilitate, upgrade, and maintain national highways Short to medium term 600 9.8% Rehabilitate provincial and community road networks Short to medium term 13 0.2% Build human and institutional capacity in government agencies that are Short to medium term 15 0.2% responsible for regulation and management of road transport activities Prepare a business and master plan for civil aviation Short term 1 <0.1% Maximizing the benefits of Burundi’s EAC membership Institutional Recruit competent staff in MEACA Short term 0.2 <0.1% reforms Make institutional arrangements (including capacity building and other regional Short to medium term 0.5 <0.1% integration issues within the new ministry) Implement the coordination, monitoring, and evaluation mechanisms Short term 0.3 <0.1% Implement communication/sensitization campaign Short term 0.5 <0.1% Economic Eliminate remaining NTBs (including equipment at border posts, one stop Medium term 1 <0.1% reforms window at the port, and so forth) Strengthen and operationalize the public private dialogue framework Short to medium term 0.3 <0.1% Increasing the contribution of the private sector to growth Financial Provide technical assistance to banks and MFIs to develop products that meet the Short to medium term 1 <0.1% sector needs of small and medium enterprises Establish units within banks that specialize in lending to small and medium Short to medium term Included in previous enterprises, and units responsible for rural and agricultural customers technical assistance Revise the regulatory framework limiting the development of MFIs; and remove Short to medium term 0.1 <0.1% the ban on MFIs engaging in leasing and issuing mortgages, subject to prior approval from the central bank Improve capacity of the BRB to supervise MFIs Short to medium term 0.4 <0.1% Regulatory Finalize and adopt the revised mining code Short term 0.6 <0.1% environment Finalize the revision of the tax code, including the introduction of the simplified Short to medium term 0.35 <0.1% impôt synthétique (synthetic tax) Governance Promulgate the revised Privatization Law Short term to be determined Adopt and implement the national policy of good governance and its action plan Short to medium term 30 0.5%
xv Encouraging emerging growth sectors Food crops Strengthen input distribution systems (technical assistance) Short to medium term 0.5 <0.1% Rehabilitate, create, and strengthen local infrastructure for storage, conservation, Short to medium term 10 0.2% transformation, and commercialization of agricultural products Organize, structure, and professionalize producers Short to medium term 3 <0.1% Coffee Strengthen the capacity of the regulatory agency (ARFIC) Short term 1.5 <0.1% Launch the second round of washing stations and dry mills privatizing Short term 0.5 <0.1% Promote microcredit and matching grant programs to help farmers finance inputs Short term 30 0.5% or replant trees Strengthen the capacity of the interprofessional organization (INTERCAFE) Medium term 1.5 <0.1% Improve processing practices Medium term 0.5 <0.1% Diversification Develop an action plan to establish facilities to comply with international sanitary Medium term 1.5 <0.1% and phytosanitary standards Improve the availability of cargo space to export fresh produce Short to medium term 0.5 <0.1% Mining Develop and implement a sectoral strategy that explicitly accounts for the Short to medium term 0.15 <0.1% characteristics and constraints of the different types of mineral resources Become a candidate country of the EITI, be validated as a compliant country, and Short to medium term 0.15 <0.1% use available facilities for assistance with negotiations Revise the institutional framework by creating a unit within the MEN to deal Short term 0.3 <0.1% specifically with artisanal and small scale mining Develop a project to support the formalization of artisanal and small scale mining Short to medium term 1 <0.1% Source: Authors’ compilation. Note: Total expenditure projected for 2010 15 refers to the alternative scenario presented in section A, chapter 4, assuming a constant FBu/US$ exchange rate.
xvi
INTRODUCTION
1. This Country Economic Memorandum (CEM) is the first for Burundi since the 1980s. It has been developed in collaboration with the government of Burundi. The CEM has been prepared in cooperation with the African Development Bank and the U.K. Department for International Development. 2. Burundi is one of the poorest countries in the world, and has suffered from many years of civil conflict and its consequences. In the last years, peace has been established and a promising recovery of the economy has started. Economic growth rates, however, are not in line with what has been projected in the latest Poverty Reduction Strategy Paper (September 2006). Real GDP growth had been projected to average almost 7 percent between 2006 and 2009 in that strategy paper, but actual growth will average just above 4 percent for the same period. 3. The report reviews the economic developments in the past and tries to identify the most binding constraints to growth. The CEM then sets out a strategy to address these constraints to promote increased and participatory growth, reduce poverty, and improve the livelihood of the population. 4. The report draws on a number of background studies conducted on various subjects relevant to the country’s economic development and on existing reports and studies from the government of Burundi, the World Bank, other donors, and academics. The CEM provides a synthesis of various recommendations and attempts to prioritize and sequence key actions. 5. The CEM is not meant to offer a comprehensive view of all economic issues in Burundi. It focuses on the main constraints that have been identified for short- and medium-term growth, and suggests strategies to address those constraints. Furthermore, the necessity of a growth-supporting environment is highlighted and the role of government in supporting the economy is addressed. Given the scope of the CEM, some long-run growth-enhancing measures (such as policy reform priorities in social sectors) are not directly addressed. 6. The scope of the CEM, as well as preliminary results of its analysis, have been discussed in Burundi with the members of the CEM steering committee and with a wider audience, including representatives from various ministries, other institutions, and donor agencies.
1
CHAPTER 1 MACROECONOMIC AND FISCAL DEVELOPMENTS
1.1 Burundi is one of the poorest countries in the world. At US$111.3 in 2008, its per capita income (in constant 2000 U.S. dollars) ranks second-to-last among countries whose statistics are compiled by the World Bank, ahead of only the Democratic Republic of Congo. Burundi has not yet started the transition from a traditional society, with most of the population is employed in subsistence agriculture, to a modern society where most of the population lives in urban areas and is employed in manufacturing and services. In most countries that have already undergone such transformation, this transition has taken decades; along the modernization path, subsistence agriculture was gradually displaced as more sophisticated and organized forms of agricultural, manufacturing, and services production increased their relative shares of GDP. 1.2 Several obstacles are critical to this transformation and modernization of the economy. They are political instability and the menace of violence; poor physical infrastructure, particularly the low supply of electricity; low productivity in agriculture; the low level of integration of the economy with the rest of the world and between the rural and urban economies within the country; poor governance and government decision-making capacity; the low levels of education and health of the population; and high population growth. Through strong links and interdependencies, these obstacles cannot be seen independently. The capacity to surmount these obstacles will determine Burundi’s future economic growth trends. 1.3 Political instability and the menace of conflict is still a powerful deterrent to economic growth and poverty reduction. The assassination of the presidents of Burundi and Rwanda in 1993 sparked an ethnic clash and genocide in Rwanda and a civil war in Burundi, which destabilized both countries for years going forward. The Arusha Accords of 2000 initiated a peace process in Burundi, but insecurity remained high during most of the following five years because several rebel Figure 1.1: Per Capita GDP in Burundi and Rwanda, 1962–2008 groups did not surrender their weapons and did not join the peace process until well into the 100 50 decade. 80 40 1.4 The cost of war has been substantial. 60 30 Figure 1.1 illustrates some of the costs of political instability and conflict in Burundi and 40 20 LCU '000 LCU '000 Rwanda. From 1965 to 1992, per capita GDP in Burundi increased steadily; but from 1993 to 20 10
2000, as violence erupted, per capita income 0 0 contracted steadily. As of 2000, the cost of
political violence amounted to about 62 percent 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 of the GDP. In contrast, although GDP declined Rwanda (LHS) Burundi (RHS) by an even wider margin in Rwanda in 1994, the Source: World Development Indicators. Note: LCU = local currency time of contraction was limited and high growth unit; LHS = left hand side; RHS = right hand side rates were experienced in the post-conflict period. 1.5 The following section analyzes the performance of the economy from a historical perspective, section B gives an overview of Burundi’s public finances.
A. ECONOMIC GROWTH: HISTORICAL PERSPECTIVE 1.6 This section starts with an overview of the past five decades, analyzes growth during the period of political turmoil, and discusses the causes underlying slow growth following the Arusha Accords of
2 2000. Also included is a comparison of Burundi’s economic performance with those of Rwanda and Sierra Leone—two post-conflict countries that share similarities with Burundi. Furthermore, an outlook of macroeconomic developments is provided.
A1. Economic Growth in the Past Five Decades and the Cost of Burundi’s Civil War 1.7 Overall, growth has been low in Burundi over the last five decades. From 1962 through 2008, real output grew at an average annual rate of 2.9 percent (table 1.1), while per capita GDP grew at an average of only 0.8 percent. This 46-year period has three different subperiods: a period Figure 1.2: Real GDP Growth and GDP per Capita, 1962–2008 of steady, although slow increase in per capita income from 1962 to 1992; a 25% 180 period of severe economic contraction 160 20% from 1993 to the early 2000s; and a 140 period of slow growth and stagnation of 2 15% per capita GDP after 2000 (figure 1.2). 120 10% 100 1.8 The outbreak of the civil war in 1993 led to a severe disruption of the 5% 80 economy. The inflation rate reached 60 0% unprecedented levels, the exchange rate 40 Constant 2000 US$ depreciated, and capital stock was 5% 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 20 destroyed. In addition, neighboring countries imposed an embargo and donor 10% Real GDP growth rate (left hand side) 0 GDP per capita (constant 2000 US$, right hand side) support for Burundi was suspended. All these factors led to a negative average Source: World Development Indicators. growth rate of real GDP and real GDP per capita. The decline of per capita GDP during the 1990s contrasts with the rest of Sub-Saharan Africa, where it grew steadily. Furthermore, per capita GDP growth in Burundi was considerably more erratic than for Sub-Saharan Africa as a whole: for the period 1980–2008, such growth in Burundi displayed a standard deviation of 4.6 percent, compared with 2.1 percent for Sub-Saharan Africa. Table 1.1: Selected Economic Indicators, 1962 2008 1962 2008 1962 1992 1993 2000 2001 2008 Real GDP growth rate (%) 2.9 4.4 3.1 3.0 Population growth (%) 2.1 2.2 1.1 2.8 Real GDP per capita growth rate (%) 0.8 2.2 4.2 0.2 GDP per capita (2000 US$) 124.6 129.7 120.1 109.4 Consumer price inflation (%) 10.2 8.2 17.7 9.4 Exports of goods and services (% of GDP) 10.1 10.8 8.9 8.7 Imports of goods and services (% of GDP) 22.5 22.4 21.1 36.6 Current account balance (% of GDP)a 5.0 4.6 2.9 7.6 Gross domestic savings (% of GDP) 1.8 2.0 4.9 15.9 Gross fixed capital formation (% of GDP) 7.1 7.7 5.1 8.6 Official development assistance (% of GDP) 16.8 14.7 16.7 40.0 Aid per capita (current US$) 23.8 28.1 24.6 41.8 Revenue excl. grants (% of GDP) 16.9 19.7 Grants (% of GDP) 3.4 12.6 Expenditure and net lending (% of GDP) 24.5 35.7 Source: World Development Indicators and International Monetary Fund. Note: = not available. a. including official transfers; starting in 1962, some data are not available.
2 Nkurunziza and Ngaruko (2002) further divide the time between 1962 and 1992 into two subperiods: 1962–72, characterized by increasing political instability and weakening economic performance, but overall moderate GDP growth; and 1973–92, characterized by somewhat lower political tension (with the exception of the domestic conflict in 1988) but very volatile economic growth.
3 1.9 Peace was only restored slowly after 2000. The signing of the Arusha peace agreement in August 2000 marked the start of a transitional period and return to democracy, as peace was slowly restored. A three-year transitional government was formed in November 2001, and democratic elections were held in August 2005. Reconstruction and rehabilitation efforts began, donors resumed their support, and the economy started to recover. Real GDP growth returned to positive levels, but Burundi has not yet recovered from the steady and deep fall in GDP provoked by the war. 1.10 The cost of war was substantial for Burundi. The cost of the war in Burundi may be gauged as the difference between the actual GDP and the potential GDP Figure 1.3: The cost of war: potential and actual GDP per capita (Constant FBu) (defined as the output that would 31,000 have occurred had Burundi grown Actual GDP per capita between 1992 and 2000 at the 29,000 Potential GDP per capita median rate of economic growth 27,000 experienced in 1962–92. Figure 25,000 1.3 contrasts these two concepts. 23,000 The cost of war can be estimated as the difference between actual 21,000 and potential GDPs. The gap 19,000 amounted to about 62 percent of 17,000 actual GDP as of 2000, and the 15,000 average loss between 1993 and 2000 was roughly 41 percent. The 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 potential GDP would have reached FBu 30,000 in 2008, almost Source: Authors’ calculations based on World Development Indicators data. double the current level. 1.11 External savings declined as a result of the war. The direct negative impact of the war was amplified by the sharp reduction in external savings. The current account deficit narrowed from an average of –10.4 percent of GDP in the years 1980–92 to –2.6 percent of GDP in the years 1993–2000. This reflects a large deceleration of inflows of external savings, which served to amplify the impact of the conflict on aggregate demand. 1.12 Compared with Rwanda, the conflict has had a more severe impact on growth. In Rwanda, the conflict was brief but extremely intense, causing large loss of life. Burundi, however, experienced a prolonged period of political instability and open conflict, which continued to disrupt economic activity for well over a decade. Hence, the destruction of physical capital and the disruption of economic activity were appreciably higher in Burundi than in Rwanda. And although the cost of conflict in Burundi was limited initially, it grew rapidly and accelerated over time. Moreover, the peace accords of 2000 were not immediately able to fully reestablish law, order, or complete government control over the entire country; and the economic impact of political instability, uncertainty, and conflict extended for a prolonged period. Disarmament of the last rebel group did not take place until 2009.
A2. Slow Recovery after 2000 1.13 Per capita GDP stagnated over the past decade. To develop a better understanding of the specific factors underlying Burundi’s slow progress toward recovery in the post-conflict period, we first decompose the rate of growth by (1) sector or origin, (2) component of aggregate demand, and (3) factor of origin (growth accounting). The analysis focuses on the subperiods 1997–2000 (last years of the civil war), 2001–04 (transitional period), and 2005–08 (fragile return to democracy).
4 Supply-Side Analysis 1.14 The Burundian economy is rural and mainly agriculture based, relatively unsophisticated, and undiversified. The primary sector clearly dominates the economy (table 1.2), representing nearly half of economic output and involving the vast majority of the population. The main driving forces of the primary sector are food crops, averaging a share of about 81 percent from 2005 to 2008, followed by livestock and export crops with 13 and 4 percent, respectively.
Table 1.2: Supply Side Components, 1997–2008
1997 2000 2001 2004 2005 2008 Share Growth Contribution Share Growth Contribution Share Growth Contribution of GDP rate of GDP rate of GDP rate Primary sector 53.2 0.2 0.1 50.5 1.4 0.7 44.4 0.8 0.3 Secondary sector 12.3 3.0 0.4 13.6 5.5 0.7 15.5 5.9 0.9 Tertiary sector 26.7 0.7 0.1 29.1 4.5 1.3 33.4 6.8 2.2 GDP at market prices 100.0 0.8 0.8 100.0 2.5 2.5 100.0 3.5 3.5 Source: Authors’ calculations based on International Monetary Fund data. Note: Differences between overall GDP and the total of the sectors are the result of indirect taxes, which are not displayed.
1.15 The share of the secondary and tertiary sectors has been increasing. The secondary sector— which includes manufacturing, processing, construction, and related industries—represents the smallest overall share of the economy. It averaged about 16 percent from 2005 to 2008, with manufacturing and construction as its main subcomponents. The Figure 1.4: Sectoral Contributions to Real GDP Growth, 1997–2008 tertiary sector—services, transport, and commerce—represents a relatively large and 6 growing share of economic output, contributing about one third of GDP 4 between 2005 and 2008. The strength of this sector, however, primarily reflects an 2 increase in public services. 1.16 The primary sector remains the 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 main determinant of overall growth. In percentage points figure 1.4, it is evident that, despite their 2 improving performance, neither the secondary nor the tertiary sector has 4 succeeded in eclipsing the influence of the Primary sector Secondary sector primary sector on overall economic Tertiary sector Real GDP growth performance, particularly given the Source: Authors’ calculation based on International Monetary Fund data. relatively small proportion of the population involved in these activities. Also clear is the extreme volatility of primary sector performance and its contribution to growth. 1.17 The poor performance of the food crop sector is the main reason underlying stagnation of per capita GDP. Food crops are a source of both livelihood and sustenance for most of the population. In the near term, any strategy aimed at accelerating growth will have to focus on improving the performance of this dominant sector.
Demand-Side Analysis 1.18 Despite the recent investment expansion driven by foreign aid, investment and savings rates are very low in Burundi. Gross fixed capital formation reached an average of 15.9 percent of GDP in 2005–08, more than double the average in the 1997–2000 war period (table 1.3). In comparison, gross fixed capital formation in Rwanda averaged 21.0 percent of GDP; and in Sub-Saharan Africa, the average
5 was 20.0 percent. Public fixed capital formation has increased during the post-conflict period as a result of the reconstruction effort. This increase in the stock of productive resources has benefited from a rebound in public investment and a recovery in private investment since 2006. However, this rebound comes after a prolonged period of underinvestment, which is only now catching up to precrisis levels (figure 1.5). Information available on remittances indicates that, in contrast to other African countries, they have been insignificant—averaging less than US$800,000 in the last five years or less than 0.1 percent of GDP.3
Table 1.3: Demand Side Components, 1997–2008
1997 2000 2001 2004 2005 2008 Component Share Contribution Share Contribution Share Contribution of GDP of GDP of GDP Consumption 101.0 1.7 102.1 1.4 110.7 6.3 Public 13.6 1.1 23.4 2.2 24.1 0.6 Private 87.4 0.5 79.2 0.3 86.6 5.3 Investment 6.2 1.8 8.8 2.2 15.9 2.2 Public 5.3 1.8 6.6 1.8 8.4 0.3 Private 0.9 0.0 2.2 0.4 7.6 1.9 Resource gap 7.4 0.2 10.8 1.1 16.6 2.1 Exports 8.1 0.6 8.5 0.4 9.2 0.1 Imports 15.5 0.8 19.3 1.5 25.8 2.2 GDP at market prices 100.0 0.8 100.0 2.5 100.0 3.5 Source: Authors’ calculations based on International Monetary Fund data. Note: The individual shares and contributions may not add up to the total because of changes in investment stock, especially in the 2005–08 period.
1.19 A combination of official flows Figure 1.5: Private and Public Investment, 1997–2008 (percent of GDP) and domestic resources has financed 12 these investments. Foreign investment flows have yet to show an appreciable 10 acceleration. The very low level of 8 foreign direct investment (averaging less 6 than 1 percent of GDP over the last 10 years) has been caused by a number of 4 2 factors, including the protracted war and percent of GDP an unfavorable business environment. 0 The ongoing regional integration process will help create opportunities for regional foreign direct investment. Private investment Public investment 1.20 Public expenditure contracted Source: Authors’ calculations based on International Monetary Fund data. sharply during the crisis, but has recovered considerably since 2000. It contributed about half of the GDP growth between 2001 and 2008.4 In line with the expansion of development assistance, this increase in government spending can be attributed largely to recurrent expenditures (representing about 60 percent of overall expenditures), grant- financed capital expenditure, exceptional expenditures, and military expenditures. This expansion relies heavily on external financing and grants; and it has resulted in a significant deterioration in the budget balance, which calls into question its sustainability over time without a significant improvement in domestic revenue mobilization.
3 These are World Bank staff estimates based on the International Monetary Fund’s Balance of Payments Statistics Yearbook 2008. For more information, see Rhata, Mohapatra, and Silwal (2009). 4 General government final consumption expenditure includes all government current expenditures for purchases of goods and services (including compensation of employees), as well as most expenditures on national defense and security.
6 1.21 The export base is narrow and undiversified. Exports have been low, averaging less than 10 percent of GDP over the last 10 years. The main export crops are coffee and tea, with coffee accounting for about two thirds of the export value. The performance of exports has been erratic, owing largely to the volatility of coffee production. Imports, on the other hand, are increasing, mainly financed by the increasing level of foreign aid. Consequently, the current account deficit has been rising. 1.22 In summary, neither exports nor private investment (except over the past three years) have contributed significantly to economic growth during the recovery period. This fact, combined with the analysis of growth by sector of origin, supports the view that the stagnation of agriculture and low expected return on investment have been at the heart of poor economic performance in Burundi during the post-conflict period.
Growth Accounting 1.23 Productivity growth has been negative throughout almost the entire past five decades.5 Only the 1962–72period displayed positive productivity growth, and the highest growth rate of GDP per worker (which was attributed about equally to productivity growth and capital accumulation). Figure 1.6: Annual Growth of GDP, Capital Stock, and From the onset of political violence in 1993 to Productivity per Worker, 1990–2008 the signing of the Arusha Accords in 2000, the 10 growth of output and physical capital declined Arusha Accords markedly, and productivity growth was negative. 5 The decline is attributable to disruption of domestic economic activity, destruction of 0 physical capital, loss of foreign aid and capital percent flows, the economic embargo imposed by 5 neighboring countries from 1996 to 1999, displacement of the population, and suspension 10 of many government services over the period. Official aid flows helped finance an increase in Growth Rate of Real GDP per Worker Growth Rate of Real Capital Stock per Worker capital stock form 2001 to 2008; even so, the rate Total Factor Productivity of productivity growth continued to decline, Source: Authors’ calculations based on World Development Indicators resulting in negative GDP growth per worker and International Monetary Fund data. (figure 1.6). 1.24 Agricultural productivity has been low and declining. Given the importance of the agricultural sector in terms of its share of overall output and the large proportion of the population involved, the productivity of the sector is especially crucial. The data show that agricultural productivity has declined in Burundi since the early 1990s, with value added falling to less than US$65 per worker (in adjusted 2000 dollars) by 2005. This decline stands in sharp contrast to the Sub-Saharan African average of nearly US$288 that same year.6
A3. Comparison with Rwanda and Sierra Leone 1.25 How does Burundi’s economic performance compare with that of other post-conflict countries? Although such comparisons are clearly difficult and imprecise, given the many qualitative differences across countries (including those with respect to the depth, duration, and impact of crises), some insights can be drawn from the experiences of other post-conflict countries.
5 A growth accounting decomposition can isolate the origins of growth by factor of origin (labor and capital) and total factor productivity (TFP). Dividing the post-independence period into several subperiods, as discussed in section A1, the growth accounting decomposition has been based on a Cobb-Douglas production function. The output elasticity with respect to capital per worker has been estimated as 0.42. 6 Source: World Development Indicators.
7 1.26 Burundi, Rwanda, and Sierra Leone are at a similar developmental level. The three countries share similar levels of human development, life expectancy, literacy and educational enrollment, and poverty. All three compare poorly with most other countries assessed by the United Nations in the context of the Human Development Index: Rwanda ranked 167th, Burundi 174th, and Sierra Leone 180th, among 182 countries in 2009.
Figure 1.7: Real GDP Growth during Post conflict Periods in Burundi, 1.27 Burundi has not experienced Rwanda, and Sierra Leone the post-conflict boom that Rwanda and Sierra Leone have seen. Economic %40 performance during the years following 30 cessation of major hostilities in the three 20 nations—2000, 1994, and 2000, 10 respectively—has differed widely (table 0 1.4). Burundi underperformed its 10 t 2 t 1 t t+1 t+2 t+3 t+4 t+5 t+6 contemporaries over this period with
percent 20 respect to overall and per capita growth 30 and in investment, despite displaying the highest levels of foreign assistance and 40 Sierra Leone 50 Rwanda government expenditure. Rwanda and Sierra Leone experienced explosive 60 Burundi growth (20 and 18 percent, respectively) Source: World Development Indicators. Note: The year during which the crisis ended officially is denoted as “t”—that during the first three post-conflict years, is, 2000 for Burundi and Sierra Leone, and 1994 for Rwanda. This methodology whereas Burundi grew at a modest is also used for related comparisons in this chapter. average of 2 percent (figure 1.7). Although both Rwanda and Sierra Leone experienced more significant reductions in per capita income during the crisis period, per capita income levels returned to well more than twice the level of Burundi only six years after the crisis. 1.28 This is partially explained by the fact that in Burundi, unlike in Rwanda and Sierra Leone, the conflict ended neither rapidly nor completely. The protracted peacemaking process also led to allocation of aid to sectors without a direct impact on growth. A higher share of aid has been disbursed for humanitarian and emergency operations in Burundi than in the other two countries. Aid with a short-run impact on growth, to the contrary, received the smallest share in Burundi, whereas (especially in Sierra Leone) more aid was allocated to short-run growth-enhancing sectors. To determine what other factors explain the differences in post-conflict performance, the origins of growth in each country by sector, component of aggregate demand, and factor endowment or TFP are considered. 1.29 Regarding supply-side decomposition, several important differences are apparent: • First, the services sector has outperformed all other sectors in Burundi by a large margin, which primarily reflects a substantial aid-financed expansion of government services (table 1.4). • Second, the contributions of agriculture to overall output in both Rwanda and Sierra Leone are higher than that displayed by Burundi, despite the fact that the proportion of the population involved in agricultural activity is much lower in Sierra Leone and roughly equal to that of Rwanda—that is, the proportion of the total population involved in agriculture in 2005 was 46 percent in Burundi, approximately 22 percent in Sierra Leone, and 47 percent in Rwanda.7 • Third, the contribution of agriculture to growth in Burundi was negative, compared with the significant positive contributions of agriculture in the other two countries. Agricultural
7 These figures are based on the most recent data for agricultural sector employment available from the World Development Indicators database.
8 productivity has been significantly higher in Rwanda and Sierra Leone, with agricultural value added per worker being two to five times as high as in Burundi in the post-conflict years. 1.30 The demand-side decomposition shows that the most significant divergences relate to the growth contributions of investment and exports: • The average level of gross fixed capital formation relative to GDP in Burundi was less than half of that in Rwanda and about two thirds that in Sierra Leone during the first six post-conflict years. • The relative share and contribution of private investment to output in Burundi was about one sixth the level in Rwanda and one fifth the level in Sierra Leone over the same period. • Measured as a proportion of GDP, overall investment has generated about nine times the contribution to growth in Rwanda than in Burundi, and more than five times as much in Sierra Leone. Those figures underscore the relatively low productivity of capital displayed by Burundi during the post-conflict period. The strong contributions to growth observed might stem from higher levels of productive investment undertaken by the private sector in both Rwanda and Sierra Leone. • The performance and contribution of exports to growth have fluctuated considerably in all three countries. On average, Burundi has underperformed both Rwanda and Sierra Leone. In Rwanda, the contribution of exports of goods to growth was limited; however, exports of services excelled. Exports grew fast in Sierra Leone, but the nature of exports (minerals) implies much lower impact on creation of jobs and value added outside the sector than the isolated figure may suggest.
1.31 Coulibaly, Duffy, Table 1.4: Comparison of Selected Indicators Six Years Post conflict (averages) and Ezemenari (2008) provide a composition of Selected Indicator Burundi Rwanda Sierra Leone GDP growth (%) 2.7 14.4 12.8 growth for Rwanda during Contribution to real GDP growth of agriculture 0.7 6.5 4.3 the post-conflict period. industry 1.2 3.4 2.7 Although somewhat services 4.7 6.0 7.5 different in its approach Agricultural value added per worker (2000 US$) 74 171 401 GDP per capita growth (%) 0.7 8.3 8.4 from the one presented ODA (% of GDP) 38.7 25.9 33.0 above, this study also Gross domestic savings (% of GDP) 13.4 3.1 4.8 estimates TFP growth using Gross fixed capital formation (% of GDP) 10.5 15.8 12.3 Public (% of GDP) 6.8 7.7 4.9 a Cobb-Douglas framework. Private (% of GDP) 3.6 8.1 7.4 In contrast to Burundi, Government final consumption expenditure (% of GDP) 23.9 10.6 14.9 productivity growth in Export (% of GDP) 8.9 6.5 18.5 Rwanda was significantly Imports (% of GDP) 21.3 25.7 30.5 Foreign direct investment, net inflows (% of GDP) 0.01 0.2 3.3 positive. From 1995 to Source: Authors’ calculations based on World Development Indicators and International 2003, TFP growth is Monetary Fund data. Note: Burundi and Sierra Leone 2001–06, Rwanda 1995–2000; estimated to have accounted contribution to GDP figures are consistently available for only five years post conflict. for about two thirds of the observed increase in output, with growth in employment producing most of the remainder. This finding is striking, given that the augmentation of physical capital during the period was quite modest in relative terms, underscoring the tremendous efficiency gains during the period. 1.32 The increase in TFP in Rwanda is explained by several factors, particularly by the development of human capital via investments in health care and hospitals, education and professional training, and knowledge infrastructure; and the formation of civil service. Trade and financial liberalization also are considered to have supported productivity growth by allowing a freer flow of resources and supporting financial deepening and lower costs of capital, as well as increased government revenues that were then put toward productivity-enhancing investments. In this context, it seems clear that an important factor in Rwanda’s success has been the government’s ability to support productivity enhancements via prudent
9 policies, reforms, and public investments in critical areas (see box 1.1). This, along with increasing private investment, has been at the center of the country’s rebound.
Box 1.1: Rwanda’s Growth Experience and Possible Lessons for Burundi In its Vision 2020 document published in 2000, the government of Rwanda established targets for GDP growth and poverty reduction to be achieved by 2020. Rwanda has a long way to go to achieve some of the targets set under Vision 2020—among them, to raise per capita income from US$230 to US$900 and reduce poverty from 60 to 30 percent of the population. The country will face tremendous challenges in meeting these targets. Nevertheless, almost a decade after the establishment of the Vision 2020 targets, it has made substantial progress. A key factor in Rwanda’s progress has been its commitment to market-based reforms, a limited role for government, and support for private sector initiatives. Some reforms were under way prior to the crisis in 1994: liberalization of trade and exchange rate regimes; accession to regional arrangements, including the Common Market for Eastern and Southern Africa and the East African Community; and participation in the U.S. African Growth and Opportunity Act. These and related reforms were critical, given the need to look beyond the small domestic market for sources of growth. These measures also allowed the government to reduce military expenditures and increase spending on education, health, agriculture, and infrastructure needs. Prudent macroeconomic policies and the adoption of a medium-term perspective with respect to planning and budgeting have provided a stable economic environment for investment and reform. A political economy analysis of the causes of the genocide stressed the need for increased voice, inclusion, and widespread participation in the political and decision-making processes. These findings led to adoption of a decentralization program involving legal, institutional, and policy reforms to delineate central and local government roles, responsibilities, and accountability mechanisms. The decentralization was followed by a refocusing of the government’s efforts on service delivery to communities. These and related reforms have supported both political stability and growth. A key dimension of the reform process has been efforts to support the agricultural sector, given its size and importance for the population and economy as a whole. A national strategy for agriculture (articulated in the first postcrisis Poverty Reduction Strategy Paper) was implemented, including policies aimed at increasing productivity through the use of agricultural inputs, improved water management, and soil conservation. Recently, the government has adopted a program to strengthen the private sector–led distribution of fertilizer, which has been successful. These reforms have contributed to increases in per capita agricultural income by 5.6 percent from 2006 to 2007 and by 8.0 percent from 2007 to 2008. Rwanda has also adopted many reforms aimed at improving the business environment, which led to its being named the most improved country in the World Bank report Doing Business 2010. Reforms have focused on reducing administrative costs of doing business, improving transparency, and a implementing a zero-tolerance policy on corruption. Those reforms have led to substantial progress in the key indicators of growth and poverty reduction. Between 1995 and 2005, some of the growth experienced was catch-up growth, following the events that occurred in 1994. But much of the growth was also the product of structural reforms. As a result, GDP growth averaged 10 percent (and per capita income averaged 5.4 percent), with average inflation below 10 percent. Looking forward, the government continues to focus on strengthening the institutional framework for public investment, supporting private sector initiatives, and reducing the cost of doing business. Despite success in reforming the sector, the government recognizes the need to support a transition from heavy reliance on the agricultural sector over the medium term, and it has emphasized the development of the services sector and investment promotion in its current strategies. In sum, Rwanda’s success has been the result of a stable and responsible government focused on reform, sound macroeconomic policies, the progressive liberalization of finance and trade, public financial and expenditure management reforms, agricultural extension and investments, improvements in the business environment, and initiatives aimed at increasing social welfare and inclusion.
1.33 In summary, Burundi’s post-conflict performance does not compare well with that of Rwanda or Sierra Leone. One of the major factors underlying this divergence is certainly Burundi’s slow and sporadic progress from open conflict to peace, relative to the relatively short transitions in
10 Rwanda and Sierra Leone. In this context, the performance of productivity is particularly telling, having stagnated and fallen despite increased investment during the post-conflict period. Clearly, any recovery and increase in growth potential will require productivity enhancements, as well as additional private investment. But these improvements will not occur unless Burundi initiates a comprehensive reform of the agricultural sector and invests in significant infrastructure improvements. The private sectors in Rwanda and Sierra Leone perceive that the investment climate (especially in Rwanda) is much more favorable than in Burundi, and that perception must have had a meaningful impact on the far more dynamic contribution of private investment to recovery in Rwanda than in Burundi.
B. BURUNDI’S PUBLIC FINANCES: AN OVERVIEW 1.34 The structure and management of public finances has an important role for the development of the country, given the weakness of Burundi’s private sector and the country’s dependence on foreign financing. The following section gives an overview of the structure of Burundi’s public finances as well as the public finance reforms that have been undertaken or are currently being implemented.
B1. Structure of Public Expenditure 1.35 Burundi has a narrow tax base, and most revenue stems from domestic activities. More than 90 percent of Burundi’s domestic revenue is derived from taxes, with an average of 71 percent coming from taxes on domestic activity (taxes on goods and services and income tax) and approximately 19 percent from taxes on international trade during the period 2001–08 (table 1.5). Despite its narrow tax base and the reduction of the petroleum tax, domestic revenue roughly has maintained its 2006 level of about 19 percent of GDP over 2007 and 2008. Further increases in domestic revenue will depend on improved revenue administration and the elimination of excessive tax exemptions (see box 2.2 in chapter 2). In this regard, the Table 1.5: Composition of Government Revenue, 2001–08 (percent of GDP) government has plans to revise Revenue source Average 2006 2007 2008 Average the general tax code and unify 2001 05 2006 08 the local and national revenue Total revenue 28.3 36.8 39.5 43.5 39.9 administrations. As a result of Domestic revenue 20.3 18.9 18.6 18.5 18.7 Tax revenue 18.4 17.3 17.2 16.6 17.0 Burundi’s accession to the East Income tax 5.0 4.9 5.0 4.7 4.9 African Community, tax policy Taxes on goods & services 9.0 8.9 8.7 8.8 8.8 will be reviewed to bring it in Taxes on international trade 4.2 3.1 3.2 2.9 3.1 line with other member- Other tax revenue 0.2 0.4 0.3 0.2 0.3 Nontax revenue 1.9 1.6 1.4 1.9 1.6 countries. The introduction of a Grants 8.0 17.9 20.9 25.0 21.3 common external tariff could Current 4.8 13.5 13.2 16.0 14.3 lead to a reduction in Capital 3.2 4.4 7.7 9.0 7.0 international tax revenue. Source: International Monetary Fund. 1.36 Burundi relies heavily on grants. Grants represent an increasing source of revenue, rising to 57 percent of total revenue and thereby raising government revenue to 43.5 percent of GDP in 2008. This indicates that Burundi is heavily reliant on a category of revenue that is subject to significant delays and is less predictable than other sources of revenue. Grants are primarily disbursed as current grants, including budget support as well as support for elections and for disarmament, demobilization, and reintegration. 1.37 Total government expenditure has been rising rapidly in recent years. It has increased from 27.2 percent of GDP in 2001 to 44.1 percent of GDP in 2008 (figure 1.8). This is a very large share of public expenditure, and considerably higher than those in other Sub-Saharan African economies. Recurrent expenditure accounts for more than 60 percent of total expenditure and net lending. The wage bill has been rising over time, from 7.3 percent of GDP in 2001 to 11.4 percent in 2008. The rise since
11 8 2005 reflects the hiring of new teachers and a 15 Figure 1.8: Composition of Public Spending, 2001–08 percent salary increase in 2006—the first increase 50 since 2001. Capital expenditure has more than doubled (from 6.4 percent of GDP in 2001 to 15.1 40 percent in 2008) as a result of a steep rise in 30 foreign grants. When military expenditures, exceptional expenditures, and externally financed 20
capital expenditure are excluded, government percent of GDP 10 expenditure has been rising much more slowly. 1.38 The government has increased public 0 expenditure allocations to priority sectors over 2001 2002 2003 2004 2005 2006 2007 2008 Military and security expenditure Other current expenditure the past six years. As a result of the availability of Capital expenditure Exceptional expenditure funds for heavily indebted poor countries (starting Source: International Monetary Fund. Note: Exceptional in 2005) and increased donor support, the expenditure includes spending on DDR and elections. government increased priority economic and social expenditures from 4.8 percent of GDP in 2001 to about 12.4 percent on average during the period 2006– 08, mainly reflecting an increase in education expenditures. The increased share of expenditure allocated to social spending reflects the government’s commitment to increase pro-poor spending. Demobilization has reduced defense expenditures, but the increase in the police force has more than offset the savings. A reduction in security sector spending is needed to provide more fiscal space for further increases in social and economic priority sector expenditures. 1.39 The fiscal position improved in recent years, in line with increased external financing. The fiscal deficit (on a commitment basis and after grants) shifted to an average of 0.7 percent of GDP in 2006–08 (from an average deficit of 5.0 percent in 2001–05), mainly because of increased grants disbursement (table 1.6). The share of the deficit funded by external partners increased from 27 percent in 2001 to 95 percent in 2008. Shortfalls in budget support in 2006, partly reflecting donors’ governance concerns, were offset by spending cuts in nonwage spending and increased domestic borrowing. Following the 2007 governance incident, the government took mitigating measures (increasing some taxes, cutting expenditures, and postponing until later years increases in benefit payments to the civil service) to maintain the overall fiscal deficit within the International Monetary Fund program.
Table 1.6: Central Government Fiscal Deficit, 2001–08 (FBu billion, unless otherwise stated)
2001 2002 2003 2004 2005 2006 2007 2008 Overall balance (commitment basis) Including grants 28.4 7.9 40.2 43.5 54.1 17.1 5.7 10.6 % of GDP 5.2 1.4 6.2 5.9 6.3 1.8 0.5 0.8 Excluding grants 39.6 33.1 88.7 144.3 144.3 182.2 210.4 355.0 % of GDP 7.2 5.7 13.8 19.7 16.8 19.3 19.8 25.6 Change in arrears (reduction ) 21.3 9.4 2.2 58.5 10.2 13.7 21.9 11.0 External (current interest) 18.5 6.6 4.2 49.0 10.1 1.8 0.4 0.0 Domestic 2.8 2.8 6.4 9.5 0.1 11.9 21.5 11.0 Overall balance (cash basis) Including grants 7.1 1.5 42.4 101.9 64.3 30.8 16.3 21.6 Excluding grants 18.3 23.6 90.9 202.7 154.5 195.9 232.3 366.0 Financing 18.3 23.6 90.9 210.3 165.2 199.8 237.7 366.8 External 5.0 41.3 71.8 143.7 156.3 175.4 212.2 347.6 Privatization proceeds 0.0 0.0 0.0 0.0 0.4 3.1 0.1 0.2 Domestic 13.3 17.7 19.1 66.7 8.5 21.3 25.4 19.0 Banking sector 23.4 13.0 4.9 60.5 14.5 37.0 0.7 19.0 Nonbank sector 10.1 4.7 14.2 6.2 6.0 1.2 9.2 0.0 Financing gap/errors and omissions 0.0 0.0 0.0 7.6 10.6 3.9 5.5 0.8 Source: International Monetary Fund; authors’ calculations.
8 Primary-school enrollment has doubled since primary school fees were abolished in September 2005.
12 1.40 Given Burundi’s risk for debt distress, most external financing is provided in the form of grants. The grant element of external finance has been increasing in recent years, reaching about 90 percent in 2008. The increasing share of grants in external finance is in line with the debt-sustainability analysis of the Heavily Indebted Poor Countries (HIPC) Initiative Completion Point. It stresses that the provision of external financial assistance needs to be heavily weighted toward grants, given the country’s vulnerability to external shocks and its low and volatile export performance.
B2. Public Finance Management Reform 1.41 Since 2005, the government has taken steps to improve public finance management. The government introduced a double-entry accounting system, a new nomenclature that helps identify pro- poor expenditures, and an interim computerized financial management information system. In response to a governance scandal in 2008, the government introduced further changes in the 2008 annual budget law. In May 2008, a new, organic budget law was adopted, and it was promulgated in December 2008. The government is also consolidating its cash management by closing off-budgetary accounts. A new procurement law was promulgated in February 2008, after considerable delays. 1.42 Despite progress, there remain major weaknesses affecting the quality and transparency of public finance management. The government and the World Bank recently completed a Public Expenditure Management and Financial Accountability Review (PEMFAR) to prioritize future reforms. It identifies four main challenges: 1. improving the allocation efficiency of public expenditures by reducing security sector expenditures and increasing public expenditures to education, health, and infrastructure, in line with the Poverty Reduction Strategy Paper (PRSP); 2. maintaining control of the wage bill, including by eliminating ghost workers; 3. continuing public finance management reforms to increase transparency and strengthen fiscal discipline, including by consolidating the budget, streamlining budget execution, reducing the use of exceptional procedures, and improving internal and external audit systems; and 4. professionalizing the public service to improve performance, including by creating staff evaluation and grading systems and reducing the number of political appointees. 1.43 Increasing the share of priority sectors and pro-poor expenditures in planned and executed budgets needs to be supplemented by programs aimed at better tracking the destination and improving the quality of public expenditures. Although the public expenditure tracking survey exercise should continue, it should be complemented by identifying ways and means of associating beneficiaries’ representatives and civil society in evaluating the performance of public services and the quality of public expenditures. An International Development Association health sector project approved in June 2009 will initiate and test mechanisms aimed at involving local communities in monitoring basic health services. Extending the same approach to other key sectors should be encouraged in the context of future development policy operations. 1.44 The size of the wage bill is a serious public finance management issue in Burundi. Representing an unsustainable level of 11.8 percent of GDP in recent years, the size of the wage bill constrains resources that should have been otherwise allocated to growth-enhancing productive activities (such as investment). 1.45 External and internal control and audit institutions have been strengthened, but much remains to be done. Virtually nothing has been done to coordinate their programs so as to eliminate unnecessary duplications and focus their activities on high-priority objectives. The Inspection Générale de l’Etat (State General Inspectorate) is the most appropriate institution to initiate a dialogue between existing internal oversight departments and institutions, with a view to developing work plan for consistent, risk-based audits. In parallel, the operational independence and professionalism of the Cour
13 des Comptes (Audit Court) should be further strengthened because it has a major role to play in enhancing the transparency of public finance management and the accountability of the executive. The state accountability institutions also should progressively learn to rely on the work of internal oversight bodies as they gradually strive to comply with international standards.
14 CHAPTER 2 OBSTACLES TO ECONOMIC GROWTH AND EMERGING OPPORTUNITIES FOR BROADENING THE BASE FOR GROWTH
A. OBSTACLES TO ECONOMIC GROWTH 2.1 Private investment is low because the expected rate of return on capital is low. During the period of conflict, instability, and uncertainty, the ability of the economy to attract external savings fell considerably; and domestically financed investment was constrained by modest levels of income and domestic savings. However, it is difficult to argue that domestic savings have been a constraint since 2005, with a current account deficit (the supply of external savings facing the country) of about 15 percent of the GDP. 2.2 There is not a unique reason why the return on capital is low. In addition to political instability and the menace of conflict, these reasons include (1) isolation from the rest of the world; (2) poor physical infrastructure, particularly the low supply of electricity; (3) poor government policies regarding the largest sector—agriculture; (4) a narrow and undiversified export base; (5) a weak private and financial sector; (6) poor governance and government decision making and tax revenue collection capacity; and (7) the low level of education of the population, especially in the context of the high population growth rate. The main obstacles to economic growth are discussed next.
A1. Geographic and Economic Isolation 2.3 Low integration within the domestic economy and with the rest of the world has deep geographic, historical, and political origins. Burundi is landlocked, nestled among the highly unstable Democratic Republic of Table 2.1: Selected Indicators Comparing Burundi to Sub Saharan Africa and the Congo to the west, Rwanda to World Economy the north, and Tanzania to the Indicator Burundi Sub Saharan Africa World south. Poor physical Per capita GDP (2000 US$) 111 619 6,024 infrastructure, low average Rural population (% of total population) 90 64 50 educational attainment, and Exports of goods and services (% of GDP) 10 33 28 weak channels of interaction Agriculture value added by worker (2000 US$) 64 288 939 with the rest of the world Agriculture value added (% of GDP) 45 16 3 imply that Burundi remains Phones per 100 peoplea 4 25 71 b fairly isolated in a highly Dependency ratio 72 85 55 volatile region. Sources: Latest available data (2005 and 2008, respectively) from the World Development Indicators database and the International Monetary Fund. 2.4 The economic a. Number of fixed and mobile lines. B. Ratio of population younger than 15 and older than 64 to working age population (15—64 years). geography conjoined with the infrastructure deficit lead to high transport costs that affect trade negatively. Poor roads, bridges, and rail infrastructure hinder internal trade and reduce Burundi’s opportunities for exchange with the region and the rest of the world. The internal obstacles are compounded by the high cost of using the ports of Dar es Salaam in Tanzania or Mombasa in Kenya, poor storage facilities, inadequate freight handling capacity, and some customs and administration issues that have proved particularly difficult for Burundian firms to navigate. The poor state and coverage of infrastructure in Burundi imply high transport costs and delays that lower the expected rate of return on capital, discourage domestic and foreign investment, and constrain economic growth. In the agriculture sector, transport costs represent, on average, 35 percent of import prices and 40 percent of export prices.
15 High infrastructure costs and lack of access are major obstacles to improved incomes and well-being for the very large part of the population that depends on agriculture for a livelihood. 2.5 Two economic dimensions of this reality are the small size of the domestic market and the low export share of GDP. The size of the domestic market (US$1.1 billion in 2008) is very small; and in the World Development Indicators database, Burundi ranks 146th of 170 countries for which comparable GDP data are available. Moreover, with the exception of Liberia, countries with GDPs lower than Burundi’s are all small states.9 According to International Monetary Fund data, exports were equal to about 10 percent of GDP in 2008, the lowest in the world after Rwanda10; and significantly below the average for Sub-Saharan Africa and the world as a whole (table 2.1). 2.6 Exports and imports are critical to the development of any country—even more so when the country’s domestic market is small. However, Burundi’s export base is narrow and has not changed significantly in the past 20 years. Coffee is responsible for more than 90 percent of foreign exchange earnings. In turn, Burundi’s imports have been growing rapidly as a result of aid inflows; in the medium term, however, they are constrained by the country’s low capacity to export. The development of Burundi’s export base in the short and medium terms is directly associated with its capacity to overcome stagnation in agriculture, particularly in the coffee sector, and with its development of new products (goods and services). The development in neighboring countries needs to be taken into account. 2.7 In that context, joining the East African Community (EAC11) has been a momentous decision that has presented both opportunities and challenges. However, becoming a member of the EAC will require considerable institutional changes and policy reforms, including with respect to Burundi’s tax and customs regimes, its legal system, and regulatory structures—all challenges that will require considerable effort and time. 2.8 Many potential benefits arise from membership in the EAC. Joining the EAC opens opportunities for trade in goods and services, for development of physical and regulatory infrastructure linking Burundi to a larger regional market and to the rest of the world, and for development of services provided by the regional organization through the pooling of resources from all member-countries. Although there is a risk of trade diversion, the larger economic zone may help open opportunities for the production of new goods and services currently not visible in the trade accounts. 2.9 The importance of the regional agreement for the development of physical infrastructure cannot be overstated. As mentioned above, Burundi depends on Dar el Salaam and Mombasa for its exports and imports. Efficient port management is critical to improve the competitiveness of products coming from the region; the network of roads, railroads, and bridges that connect these final destinations with the rest of the region is also crucial. The EAC membership does not guarantee progress in any of those directions, but it is a clear step toward opening opportunities to integrate the country with the rest of the world and to broaden possibilities of exploiting economies of scale in many areas.
A2. Infrastructure Deficit 2.10 Burundi suffers from extreme infrastructure gaps in road access, power generation, communications infrastructure, and access to water and sanitation. On just about any measure of infrastructure coverage—road density, telephone density, power generation capacity, or service coverage—Burundi and the other EAC countries lag behind most other regional groupings in the world. Burundi also lags behind other EAC member-countries in access to basic infrastructure services (see table A.8 in appendix 4). About 90 percent of the population lives in rural areas. Despite the importance of agriculture, only a relatively small portion of the rural population has access to all-season roads. The road
9 Following the World Bank definition, a small state is a sovereign state with a population of fewer than 1.5 million. 10 However, Rwanda’s exports of services as opposed to goods and services) have grown significantly in the recent past. 11 The EAC is the regional intergovernmental organization of the Republic of Kenya, Uganda, the United Republic of Tanzania, and (since 2007) the republics of Burundi and Rwanda.
16 densities in areas of arable land are substantially lower in Burundi than elsewhere in Africa and in other low-income countries. 2.11 Poor physical infrastructure—in particular, the low supply of electricity—results in high production costs and inhibits new investment in the rest of the economy. About 72 percent of Burundian firms surveyed recently by the World Bank ranked poor access and reliability of electricity as the most significant obstacle to new investment. Despite abundant hydroelectric potential, Burundi displays a severe generation and transmission capacity deficit that began in years of underinvestment and of deterioration and damage during the conflict period. Only 2 percent of households currently are thought to have electricity service, compared with 16 percent for Sub-Saharan Africa and 41 percent for other low-income developing countries. Bujumbura accounts for two thirds of all connections (Benitez, Estache and Niyungeko 2009). As of 2008, the power utility, REGIDESO, had about 31 megawatts of total installed electrical generation capacity, although only about 26 megawatts of this capacity was in use. Domestic electrical production capacity in per capita terms is the lowest in the world. Limited access to electricity is an important barrier to development. Currently, the electricity shortage is met, in part, through imports from the Democratic Republic of Congo, but these are not sufficient to satisfy demand at current prices. A strategy to tackle this is critically needed. It should address both supply-side and demand-side issues, and take into consideration the extended period that such an extensive investment and reform program will require. Burundi is also lagging in mainline and mobile telephone densities, as well as Internet access. Teledensity remains poor at 3 percent of the population, with more than 90 percent of subscribers concentrated in urban areas. 2.12 In general, the following are examples of widespread problems found in the EAC (World Bank 2008c): • Poor roads and bridges—The main regional corridors and core road network are mostly paved. Acute problems are caused by the variable conditions of parts of the core/regional network and the increasing levels of traffic congestion on roads in and around major urban centers; those factors, in turn, affect access to the ports of Dar es Salaam and Mombasa. Figure A.2 in appendix 4 gives the locations of the main road transport corridors in East Africa. • Dilapidated railways—Disrepair of railroad infrastructure and equipment is characterized by aging tracks, shortage of cars, and inadequate locomotives. Most of the railway systems in Kenya, Tanzania, and Uganda have now been contracted out to concessionaires. This is expected to lead eventually to improved operational and financial performance of these rail networks. Figure A.3 in appendix 4 gives the location of the rail transport network in the EAC. • Inadequate ports and inland container freight stations—Problems include inadequate storage and handling capacity; handling equipment (such as cranes) in poor condition; and lack of adequate rolling stock at Bujumbura, Dar es Salaam, and Kigoma. In 2007, for example, about 700 containers were being offloaded each day at the port in Dar es Salaam, although Tanzania International Container Terminal Services was able to deliver no more than 300containers out of the port daily. Inland container freight stations have similar problems. • Insufficient cargo vessel capacity at the Great Lakes—Age and disrepair of the merchant fleet on Lake Tanganyika and Lake Victoria are serious problems. There is no shipyard on Lake Tanganyika that could maintain or renovate the existing fleet. With the current low levels of traffic on the lake, however, a shipyard may not be viable. • Inadequate facilities at the border posts—Problems at the border posts include understaffing relative to the volumes of activities and cargo handled; limited parking space for cargo trucks at most posts; lack of truck scales forcing customs agents to estimate loads; inadequate office facilities, including lack of computer equipment; and inadequate bonded warehouses.
17 • Poor power supply—A number of border posts have no electricity; for others, the power supply is unreliable because there is a lack of fuel for generators. As a result, bookkeeping must be done manually. • Expensive cross-border communication—Counterpart revenue authorities located immediately across a border communicate on landlines charging international rates. Private telecommunication companies (such as MTN, Safaricom, and Zion) are developing special arrangements for cross- border calls at local rates, but these do not cover Burundi. 2.13 Access to infrastructure services is limited, and the poor state of infrastructure leads to substantially higher costs. Prices for services can be two to three times greater than prices in other countries, further undermining the competitiveness of Burundi business in regional and global markets (table A.9 in appendix 4). The cost and adequacy of these services affect commercial opportunities for small farmers, entrepreneurs, and businesses both small and large. Business surveys in Burundi consistently identify the cost of power and the poor reliability of the service as the single most important obstacle to increased business investment.
A3. Poor Agricultural Policy 2.14 Low productivity in agriculture is one of the main barriers to development. According to World Development Indicators data, agricultural output, which is dominated by subsistence farming, represented about 45 percent of GDP in 2005, and involved about 90 percent of the labor force. Value added per worker in the sector is US$64 (in constant 2000 dollars)—well below the average for Sub- Saharan Africa and about 15 times smaller than the global average, indicating very low productivity in the sector. 2.15 Given the size of the agriculture sector and its potential to benefit large segments of the population, addressing the causes of low productivity is of utmost importance. The immediate causes of low productivity are low human capital and low use of modern machinery, fertilizers, productive seeds, and modern methods of production. The underlying reasons are these: poor economic incentives for farmers to introduce modern inputs; poor physical infrastructure and extension services supporting agriculture; market organization that is very poor and heavily distorted by government intervention (as in coffee production); and slight development of farmer associations and cooperatives to provide storage, commercialization, and financing of crops and other supportive services. 2.16 Increasing government intervention in the coffee sector, the main cash crop, has been a failure in the past. The government has tried to improve performance in the sector through liberalization and privatization, but the effort has been partial and there is considerable uncertainty about the roles of the private and public sectors. Furthermore, prices have been distorted by government intervention and have stagnated, providing few incentives to producers to improve quality. 2.17 The government is now moving in the right direction. Liberalizing agricultural prices, defining a subsidiary role for the public sector in the production and commercialization of coffee and tea, and tabling the privatization of washing stations are positive steps. Unfortunately, these programs are being implemented very slowly. Even if privatization occurs in coffee and tea, the challenges to increasing productivity in these crops will not be eradicated. Transitioning from the current situation to one where private organizations are responsible for services supporting farmers is not straightforward.
A4. Narrow and Undiversified Export Base 2.18 Burundi’s export base is narrow and undiversified, making the country vulnerable to external shocks and dependent on foreign financing. This section outlines the composition of Burundi’s current exports, also compared with the other EAC members. Furthermore, Burundi’s potential to diversify its exports is explored.
18 Burundi’s Current Exports 2.19 Burundi very narrow export base (averaging only about 10 percent of GDP) has not changed notably in the last 20 years. In comparison, other EAC members, especially Tanzania and Uganda, have experienced an increase in exports as a share of GDP since 1985, coupled with a significant increase in GDP per capita. The same development can be observed, on average, for low-income and Figure 2.1: HI for EAC Member Countries Sub-Saharan African countries. Burundi’s exports remain very concentrated. One possible 1.0 measure of export concentration is the 0.9 12 0.8 Herfindahl index (HI). The HI lies between 0 0.7 and 1, where 1 represents exports that are 0.6 extremely concentrated (only one product), and 0 0.5 stands for a completely diversified export basket. 0.4 Economies with an index below a threshold 0.1 0.3 0.2 are considered to have a highly diversified export 0.1 basket. The HI for Burundi has remained high 0.0 throughout the entire period considered here 1970 79 1980 84 1985 89 1990 94 1995 99 2000 04 2005 07 (figure 2.1). It averaged about 0.5 from 1985 to Burundi Kenya Rwanda 2004; and it has even risen above 0.8 during Tanzania Uganda recent years, reflecting the higher share of coffee Source: COMTRADE, SITC 2–4 digits; calculation by the Economic Policy and Debt Department of the World Bank. in total exports. 2.20 Compared with most of the other EAC members, Burundi has a relatively high level of export concentration. Rwanda’s exports are similarly concentrated, but Kenya and Tanzania have low and decreasing HIs (figure 2.2). The continued decline of the HI for Uganda demonstrates the country’s Figure 2.2: HI and GDP Per Capita success in diversifying its export base. Its HI 2,500 was as high as 0.9 in the early 1990s, implying El Salvador almost fully concentrated exports, but it has 2,000 Colombia decreased significantly to below 0.2 since Guatemala 2001. 1,500 2.21 The high concentration of exports is Honduras 1,000 correlated with low GDP per capita. Figure Cote d'Ivoire 2.2 illustrates the negative relationship 500 Tanzania Kenya Rwanda between a high HI and GDP per capita for a CAR GDP per capita (2000 US$) MadagascarUganda Sierra Leone Burundi number of coffee-exporting countries. It also 0 demonstrates that Latin American countries 0 0.2 0.4 0.6 that used to be mainly coffee exporters have Herfindahl Index been more successful in diversifying their Sources: World Development Indicators and COMTRADE, SITC 2–4 digits; economies and increasing their income levels. calculation by the Economic Policy and Debt Department of the World Burundi is clearly at the very bottom of the Bank. curve with a very low GDP per capita and strong export concentration. 2.22 Burundi’s economy is highly dependent on primary products, predominantly coffee. Burundi lags behind the development of all other countries that were exporting primarily coffee in the 1980s. All other coffee exporters have reduced their dependence on coffee and established alternative export products. This dependence on coffee makes Burundi vulnerable to fluctuating external demand, price shocks, and unfavorable climatic conditions. In addition, the coffee sector is suffering from
12 ͦ The HI is calculated as ͂̓ Ɣ ?$Ͱͥ ͧ$ , where ͧ$ is the share of the considered variables (in this case, exports) and N is the total number of products exported. Two factors can lead to a lower HI, a more even distribution of shares, and an increase in the number of products.
19 declining output and quality. The sector, however, is undergoing a comprehensive reform (as described in detail in section B2). 2.23 The second most important primary export product is tea, which has remained in the shadow of coffee in the past. The structure of the tea sector has remained largely unchanged since the 1960s when the first tea plantations were established. Constraints the sector faces include structural inefficiencies of the tea factories and plantations, poor incentives to smallholders and estate laborers, limited use of inputs and extension services, and nonexistent research. The tea sector, however, has the potential to contribute significantly to export earnings; and a reform program of the sector is currently under way. 2.24 Horticulture exports are only a small share of Burundi’s total export basket. Horticulture exports from Burundi began in the mid-1980s and picked up momentum into the 1990s, peaking in 1993. Burundi has fallen back since that period, while market opportunities and food procurement systems worldwide have changed radically. The global economic and financial crisis in 2008 has also caused a decline in horticulture exports. Flower exports especially have suffered in recent years, also because they were supplied to only one buyer and no diversified customer base had been established. 2.25 Gold used to be one of the main export products. Gold reached as much as a 30 percent share of exports in the early 1990s, but the civil conflict hampered the development of the mining sector. The country has large deposits of several minerals (such as nickel, coltan, and cobalt) that have not been exploited commercially (see section B4).
Figure 2.3: Composition of Exports in EAC Countries, 1985–89 and 2005–06
1985 1989 2005 2006 100% 100%
90% 90%
80% 80%
70% 70%
60% 60%
50% 50%
40% 40%
30% 30%
20% 20%
10% 10%
0% 0% Burundi Kenya Rwanda Tanzania Uganda Burundi Kenya Rwanda Tanzania Uganda PP PP HV RB LT MT + HT PP PP HV RB LT MT + HT Source: COMTRADE, SITC 2–4 digits; calculation by the Economic Policy and Debt Department of the World Bank and the authors. Note: These are the technology classifications: primary product (PP), high value primary product (PP HV), resource based (RB), low tech (LT), medium tech (MT), high tech (HT).
2.26 Burundi’s export structure is characterized by very low use of technology. Following Lall (2000), exports (excluding primary products and natural resource–based products) can be classified according to the technology level needed to produce them—namely low, medium, and high. High-value primary products represent a subgroup of primary products. The share of primary products—mainly coffee and tea—in Burundi’s export basket has even increased in the recent years, reaching 93 percent in 2005–06. The importance of resource based–products (mainly gold) has decreased since the early 1990s. The share of low-, medium-, and high-tech product has always been minimal, reaching only about 2 percent of total exports in 2005–06.
20 Box 2.1: The Transformation of Uganda’s Exports—A Success Story Uganda was in a situation very similar to that of Burundi in the 1980s, recording very low GDP per capita and exporting almost exclusively coffee after a prolonged period of civil conflict. Like Burundi, the country is landlocked and has access to one of the Great Lakes. Uganda, however, has been successful in stabilizing the country and diversifying its export base. It has a natural comparative advantage in unprocessed, primary exports because of its abundant natural resources, but is disadvantaged by an unskilled workforce and its landlocked location. During the 1960s and 1970s, its export basket contained mostly coffee and cotton; other commodities (copper, tea and mate, hides and skins) had shares of less than 10 percent. The cotton sector collapsed in the 1970s because of heavy taxation; it has never recovered. During the 1980s and early 1990s, the pattern of exports shifted toward a “coffee-only” trend. In the 1990s, the abolition of the export tax and trade liberalization led to a dramatic export recovery. The mid- 1990s marked a structural shift in Uganda’s export pattern. The story line transformed from mainly coffee to mainly coffee, fish, and flowers. Fish was a new natural resource–based discovery. The share of coffee plummeted from around 94 percent of total exports in the mid-1980s to 28 percent by 2004, while the share of cotton remained roughly 4–5 percent. A few traditional exports (such as tea and mate, animal feed, and oilseed) either lost their shares or vanished. Several nontraditional exports emerged. Exports of fish started in the late 1990s and their share accelerated from 4 percent to 26 percent between the early 1990s and 2004 (second only to coffee). The shares of other high-value exports (such as fruits and vegetables, spices, tobacco, leather products, maize, hides, and skins) also gained ground. The graduation of fishery products—which catered informally to the local and neighboring markets—to export products was not a natural outcome of market forces. After repeated bans by the European Union because of failure to comply with its phytosanitary standards, concerted efforts by the government of Uganda in partnership with external donors succeeded in establishing the regulatory standards necessary for the fishery exports industry to take off. The government of Uganda has been generally proactive in promoting export growth and diversification. Policy measures taken by the government include • replacement of the export licensing system with a system of export certificates without value limitations • establishment of the Uganda Revenue Authority, leading to the rationalization of the tax and revenue system • creation of a new Investment Code, specifying duty and tax exemptions to exporters and offering additional protection to foreign investors • The creation of the Uganda Investment Authority, a one-stop investment clearing agency that promotes, facilitates, and supervises investments in Uganda. To catch up with its comparators outside Sub-Saharan Africa, Uganda will need to scale up high-PRODY exports to raise their weight, reduce that of coffee, and boost its EXPY. Uganda’s EXPY has increased considerably in the past, but its more recent trend has worrisome implications for its development. Around 2000, the emergence of high- PRODY fish products augmented Uganda’s EXPY significantly; but the absence of similar products since that time flattened out its EXPY during 2001–05. To transform into a middle-income country, Uganda has few options but to diversify into at least some higher-PRODY products. Catch-up experiences indicate that a few high-PRODY products sufficed in the early stages of diversification, but persistence was necessary for sustained growth. Sources: Boccardo and Chandra 2008; Dijkstra 2001.
2.27 In contrast, the other EAC members have successfully diversified their export portfolio (figure 2.3). Particularly notable is Uganda (box 2.1). Twenty years ago, Uganda was almost exclusively exporting coffee. Now, however, high-value primary products, such as fish fillets or cut flowers, make up more than one third of its exports. Kenya and Tanzania have expanded their shares of high-value primary products, as well as low- and medium-tech products. Rwanda is still mainly relying on primary products and mineral resources, and has not yet been able to establish high-value exports. The case of Uganda shows that the promotion of high-value primary products is a viable strategy to diversify the economy and increase the income level.
21 Burundi’s Export Potential 2.28 Burundi has been only partially successful in introducing high-value primary products, such as cut flowers. High-value primary products as well as resource-based products can contribute to an increasing income level to the same extent that manufactured products can contribute. Hausmann, Hwang, and Rodrik (2007) have defined a measure that ranks products according to their implied income/productivity level—their “PRODY.”13 PRODY is a good measure to assess the potential of a specific product to enhance per capita income. The evaluation of a product according to its PRODY— which is product specific, but not country specific—can help Burundi choose products that can contribute most to a higher income level. Especially high-value primary products (such as cut flowers,14 palm oil, or fish fillets) may be appropriate products for Figure 2.4: EXPY of EAC Countries Burundi to focus on in the medium term. The 4,000 average (unweighted) PRODY of the top 10 3,500 export products from Burundi has more than doubled from US$2,878 in 1985–89 to 3,000 US$6,705 in 2005–06. The share of the 2,500 products with a high PRODY, however, 2,000 remains very low. 1,500 1,000 constant 2000 US$ 2.29 The income potential of Burundi’s 500 export basket remains low. Based on the 0 PRODY, Hausmann, Hwang, and Rodrik 1970 791980 841985 891990 941995 992000 042005 06 (2007) developed another measure, called the Burundi Kenya Rwanda EXPY, which evaluates the income potential Tanzania Uganda of the entire export basket of a specific 15 Source: COMTRADE, SITC 2–4 digits; calculation by the Economic Policy country. They find that there is a strong and Debt Department of the World Bank. positive correlation between the sophistication of the export basket, measured by the EXPY, and subsequent growth. According to their results, the relationship runs from higher sophistication to growth. Although Burundi’s EXPY has increased, it remains low, thus demonstrating the limited income potential of the country’s export basket. Burundi lags behind the development in other EAC member-states—especially Tanzania and Uganda, which also started at rather low EXPY levels in the 1980s but have been successful in exporting larger shares of high-PRODY products in recent years (figure 2.4).
Classification of Burundi’s Exports 2.30 Burundi’s exports are dominated by a few classics, and there is only a very limited number of emerging champions. Based on the development of their revealed comparative advantage (RCA) between 1985 and 2006, products can be classified in four categories: classics, disappearances, emerging champions, and marginals.16 A summary of Burundi’s exports, classified according to this approach, is given in table 2.2.17
13 PRODY is calculated as the weighted average of the GDP per capita of the countries exporting the product. 14 As mentioned above, flower exports have suffered significantly during the crisis. Flowers had only been supplied to one buyer, who cancelled orders in response to low demand. 15 EXPY is computed as the weighted average of the PRODY, with the export shares of the respective country as weights. 16A country has an RCA if its share of a certain product in its total exports is greater than the share of that product in global exports (Balassa 1986). If a product has an RCA less than or equal to 1, it is assigned a value of 0; an RCA greater than 1 is assigned a value of 1. For a more detailed description of the applied framework, see Chandra and Osorio Rodarte (2009) or Boccardo and Chandra (2008). 17 A detailed table is given in table A.2 in appendix 2.
22 Table 2.2: Classification Summary of Burundi’s Exports
Exports (US$ thousands) Shares (% of total exports) PRODY RCA Category 1985–89 2005–06 1985–89 2005 06 (US$, PPP) 1985–89 2005–06 Classics 131,925 58,750 95.53 93.18 3,190 1 1 Disappearances 1,539 22 1.05 0.01 5,036 1 0 Emerging champions 1,030 915 0.62 1.62 4,304 0 1 Marginals 2,655 1,558 1.07 2.03 5,876 0 0 Source: COMTRADE, SITC 2–4 digit; calculation by authors and the Economic Policy and Debt Department of the World Bank. Note: PPP = purchasing power parity.
2.31 The classics, products for which Burundi has maintained an RCA, are dominated by coffee, followed by tea and nonmonetary gold. Smaller shares can be attributed to fresh fish; plants, seeds, fruit used in perfumery and pharmacy; and goat, kid, sheep, and lamb skins (between 1 percent and 2 percent of total exports). The classic products account for more than 90 percent of exports and have a comparably low PRODY, on average. Their performance has been disappointing in recent years, with the value of total exports dropping to about half of the profit realized in 1985–89. 2.32 The only Burundi products that have been identified as disappearances are bovine and equine hides as well as bones, horns, and the like. Those products had an RCA of 1 in the past, but lost their comparative advantage. 2.33 Products are classified as emerging champions if their RCA was 0 in the past, but turned into 1 in 2005–06. Their share of total exports is still very small, but increasing. Only very few emerging champions have been identified, including ores and concentrates of other nonferrous metals; cut flowers and foliage; bulbs, tubers, and rhizomes of flowering or foliage plants; and leather or other hides and skins. 2.34 Many products have been classified as marginals. The share of total exports, however, remains low, at less than 2 percent. A number of marginal products have the potential to develop into emerging champions and contribute notably to the diversification of the export base. The list contains several high- PRODY products, a number of them high-value primary products (such as bananas, other fruits, and vegetables) and resource-based products (such as sawlogs and different alloys).
A5. Weak Private and Financial Sectors 2.35 Burundi has one of the most unfavorable business environments in the world. The World Bank’s 2010 Doing Business survey ranked Burundi 176 out of 183 countries18 surveyed in their Ease of Doing Business Ranking (World Bank 2009a), near the bottom relative to other countries assessed in the areas of dealing with permits, getting credit, protecting investors, trading across borders, and enforcing contracts. Another cross-country assessment of business conditions ranked Burundi 133 out of 133 countries surveyed in its Global Competitiveness Index.19 Survey respondents highlighted corruption, political instability, inefficient government bureaucracy, and crime and theft among their top concerns. 2.36 As a consequence, the country suffers from low productivity growth, thereby impeding private sector efficiency and productivity gains and ultimately slowing overall growth. The results of an enterprise survey carried out in 2006 showed that domestic firms clearly have a negative perception of Burundi’s business climate (World Bank 2008a). The main constraints cited by managers of formally registered firms were access to electricity, access to and cost of financing, political and macroeconomic
18 Burundi is only followed by República Bolivariana de Venezuela, Chad, the Republic of Congo, São Tomé and Principe, Guinea-Bissau, the Democratic Republic of Congo, and the Central African Republic. 19 The World Economic Forum’s 2009–10 Global Competitiveness Report, which assessed and compared 133 countries based on both surveys and quantitative measures. The Global Competitiveness Index is a composite measure of individual indicators, including specific indicators related to the quality of institutions, infrastructure, economic stability, health and education, and the like. For details, see http://www.weforum.org/pdf/GCR09/GCR20092010fullreport.pdf.
23 instability, tax rates, and business practices of informal competitors. Findings from a survey of the urban informal sector suggest the lack of infrastructure, access to finance, insufficient and volatile demand, and lack of professional and technical skills as the main constraints. According to the Heritage Foundation’s 2011 Index of Economic Freedom, Burundi reaches a score of only 49.6 out of 100.20 Poor governance also impedes the development of the private sector. Implementing fundamental structural reforms—which has already begun—is necessary for private sector-led growth. 2.37 The Burundian private sector is underdeveloped. It includes a large number of small enterprises, most of which operate in the informal sector. The formal private sector includes about 3,000 registered enterprises, employs some 37,000 workers, and produces mainly for the local market. Construction, agricultural processing, brewing, energy, and communications are the main activities of the very small industrial sector. The service sector is more important, but it is dominated by growing public services, accounting for about two thirds of the sector. 2.38 The development of the private sector is hampered by several constraints. They can be broadly grouped into four categories: (1) structural constraints, (2) supply-side constraints, (3) poor regulatory environment, and (4) poor governance. The government has recognized the situation and started the reform process, but much remains to be done.21
Structural Constraints 2.39 Structural constraints include the landlocked position of the country, the lack of market links, and the poor condition of the infrastructure. According to the latest Investment Climate Assessment (World Bank 2008a), the lack of reliable infrastructure is seen as the main constraint for enterprises. Eighty-seven percent of manufacturing establishments cite the provision of electricity as a major or very severe constraint, and transport infrastructure is considered a major or very severe obstacle by more than 27 percent of respondents. The severity of Burundi’s infrastructure gap and how it negatively affects the country are outlined in section B of chapter 3.
Supply-Side Constraints: Finance and Workers 2.40 The main supply-side constraints are the limited access to and high cost of finance, the limited availability of skilled workers, and low labor productivity of labor
Access to Finance22 2.41 The financial sector of Burundi is rather small and lags behind those of its neighbors in the EAC. The commercial banks suffer from high numbers of nonperforming loans and internal weaknesses, although there are prospects for changes with the recent entry of foreign banks; the payment system is archaic, and there is no card payment system; the largest microfinance network is facing difficulties; insurance companies operate without a license and without supervision; and access to finance for housing, small and medium enterprises, and agriculture is quite limited. Consequently, the financial sector is in no position to play its role efficiently in Burundi. 2.42 Access to financial services is limited. Available data show that only about 1.90 percent of the total population have bank accounts, 0.42 percent use bank credit services, and 4.00 percent are members of microfinance institutions. Postal checking services manage about 120,000 accounts, mainly those of civil servants. Movements on these accounts are rare, except at month-end when wages are paid. There is
20 This score makes Burundi the 148th freest country (out of 179 countries) and the 31st out of 46 countries in Sub-Saharan Africa (Heritage Foundation 2011). 21 Currently being prepared is a World Bank project that will provide $20 million to support the development of the financial and private sectors. This section draws on the findings and recommendations outlined in the relevant project appraisal document. 22 This section benefits from the results of the International Monetary Fund’s and World Bank’s Financial Sector Assessment Program.
24 also a high concentration of bank branches in urban areas. Given most people’s lack of access to commercial bank services, microfinance institutions have an important role to play. However, nearly 80 percent of members/customers are wage earners. Thus, the industry continues to exclude a majority of the population (nonwage earners) while serving a market that is uncertain for it in the medium term (uncertain because the banking industry could rather easily serve these customers if it makes that its objective). The financing of agriculture or production in general remains marginal. Because of difficulties in obtaining access to credit and the high cost of it, 74 percent of microenterprises and informal firms and 67 percent of enterprises in the manufacturing sector use internal funds or reinvested earnings as their first and foremost source for financing assets. Sixty-five percent of firms that needed financing in 2006 did not even request a loan, mainly because of the high interest rates, the guarantee requirements, and the complexity of the process (World Bank 2008a). 2.43 Burundi’s performance trails that of countries at a similar level of development. With 4.0 accounts per 1,000 inhabitants, Burundi is below the average for such countries as the Central African Republic (5.6 per 1,000), Uganda (5.8), and Cameroon (14.4). This low level reflects the widespread use of cash. Fiat money represented 38 percent of the money supply as of December 31, 2007. 2.44 The distribution of bank credit is not consistent with the structure of the economy. A breakdown of credit by economic sector shows Table 2.3: Sectoral Distribution of Bank Credit a large concentration Sector 2006 2007 2008 in the commerce FBu Share FBu Share FBu Share sector, millions (%) millions (%) millions (%) representing 60 percent Agriculture and related business 22,143 11 8,587 4 27,544 10 of the outstanding Industry 1,976 2 3,66 2 5,431 2 credit of all banks at Construction 9,826 5 6,325 3 10,251 4 Commerce 132,261 67 144,118 70 158,295 60 end-2008 (table 2.3), Miscellaneous services 30,350 15 43,880 21 61,907 24 whereas this sector Total 196,556 100 206,576 100 263,428 100 accounts for less than Source: BRB. 10 percent of GDP. Financing of agriculture increased from 4 percent to 10 percent of total credit, whereas agriculture’s share of GDP was more than 40 percent. 2.45 The financial products supplied by banks are little diversified. Bank credit is generally short term. Other financial services are marginal. On the deposit side, demand deposits are prevalent, for which remuneration can be freely set; and within time deposits, those maturing in less than a year seem the most widespread. In 2007, broad money represented 35 percent of GDP, which is short of the 41 percent average for Sub-Saharan Africa and well below that of South Africa, (where it represents 56 percent). Microfinance institutions have limited products, are not allowed by law to offer leasing services or mortgage loans, and cater mainly to the needs of wage earners. 2.46 The Investment Climate Assessment conducted in 2006 (World Bank 2008a) confirmed that the lack of access to bank credit is a major obstacle. Cost of and access to bank credit appears to be a major constraint for 69 percent of manufacturing firms and 43 percent of firms in the trade sector. For these firms, cost and access rank second as a major concern, behind electricity. Cost and access are the most important constraint for 68 percent of firms from the informal sector. Compared with other EAC countries, Burundi is in the worst situation. 2.47 Fully aware of the problems, the authorities already have taken a number of measures. The country has a new banking law, a new central bank charter, and a new law on microfinance. The central bank has also put into place the supervision of microfinance institutions. The authorities requested that the International Monetary Fund and the World Bank undertake a Financial Sector Assessment Program (FSAP). The program took place in January 2009.
25 Limited Availability of Skilled Workers and Low Productivity of Labor 2.48 Burundi’s labor market regulations are relatively adaptable, according to the Investment Climate Assessment from 2006. In fact, issues related to the labor market (difficulty in hiring, rigidity of hours, redundancy costs, and so forth) are the least important obstacles (World Bank 2008a), and the labor rigidity index is low compared with other Sub-Saharan African countries. Low wages prevail, and that could be seen as a comparative advantage in the short run. 2.49 There is, however, an urgent need to improve technical and professional skills in the labor force. Currently available education does not seem to provide the private sector with enough skilled labor. Nearly all employees surveyed for the assessment report the need for training (94 percent), mostly in technical, professional, and production competencies. Furthermore, language skills (French and English) and information technology skills have been cited as required training. 2.50 There is also a need to improve the flow of information on the labor market, especially regarding available positions, to better match supply and demand. Currently, there is no formal system in place, and positions are communicated only through social networks. To ensure equal access to the labor market and a reliable flow of information, objective channels of dissemination must be established.
Poor Regulatory Environment 2.51 Administrative and bureaucratic procedures in Burundi are cumbersome. Paying taxes in Burundi is complicated by the multiplicity of taxes that discourages the formalization of businesses and leaves room for uncertainty and corruption. Microenterprises see registration formalities as regulatory deterrents to entering formal markets. Thus, small, informal firms established by such entrepreneurs are unable to grow and benefit from economies of scale or to access additional benefits of formalization (such as commercial bank finance and business development services). Investment regulations are burdensome and discourage foreign direct investment.
Box 2.2: Burundi’s Tax Regime Burundi’s current tax regime is riddled with large tax exemptions that make it overly complex and opaque and lead to important distortions that negatively affect producers. Tax exemptions are widely perceived as inefficient, leading to discretionary decisions and delays. The various exemptions (which mainly concern imports) not only reduce revenue collection, but also make tax administration more complex. According to a recent study, 60 percent of imports entering Burundi in 2006 had partial or total exemptions, costing FBu 103.7 billion (that is, 10.7 percent of GDP and 65.6 percent of tax receipts). The government’s decision to introduce a broad VAT, together with membership in the EAC, will lead to broad changes in the tax regime. Under the assumption of a broad-based VAT introduced at 17 percent and without EAC adhesion, it has been estimated that it will lead to a reduction in tax revenue of FBu 1.6 billion, equivalent to 0.2 percent of GDP. However, the combined effect of introducing the VAT and joining the EAC is expected to have a large negative effect on overall receipts—estimated to FBu 15 billion (1.5 percent of GDP). This anticipated negative effect highlights the importance of introducing a broad-based VAT regime, reducing import exemptions, and improving tax administration. 2.52 Tax policy is not conducive for the development of the private sector. The tax burden is one of the highest in the world, and consequently is seen as one of the main obstacles for the private sector. According to the Doing Business 2010 report (World Bank 2009a), the total statutory tax rate over profits ratio is 279 percent, which may lead to tax evasion. The tax regime would need to be simplified and tax exemptions avoided. The value-added tax (VAT) of 18 percent was recently introduced (see box 2.2), but it needs to be accompanied by information campaigns and taxpayer training.
26 2.53 A revision of legislation and codes is necessary. Burundi’s legal, regulatory, and procedural framework is outdated. Many key laws have been revised recently, but are often not accompanied by appropriate regulations, which causes no or arbitrary implementation. 2.54 The government already has taken important steps in the right direction, but much remains to be done. In June 2008, the Council of Ministers approved a new investment code that includes some of the best practices in investment legislation—among them, investor protection, freedom to transfer capital and dividends, and a streamlined approval process. The investment code was declared law in September 2008 and has been implemented, providing additional incentives to investors. Unfortunately, the revision of the tax code has been significantly delayed. A competition law is being drafted, and two insolvency laws were adopted recently. The draft versions of the commercial code and the code for public and private companies were revised, and (in line with international standards and the Organization for the Harmonization of Business Law in Africa [Organisation pour l’Harmonisation du Droit des Affaires en Afrique – OHADA] recommendations) both have been adopted by the Council of Ministers and submitted to Parliament. The mining code is currently under revision. It should prepare the country for potential large-scale investments and clarify the role of artisanal and small-scale mining. 2.55 The capacity to enforce contracts is weak. The economy is characterized by a weak application of the laws for businesses; and by weak capacity to enforce commercial contracts and have commercial disputes resolved fairly, predictably, and professionally in the commercial court. Burundi has only one commercial court in Bujumbura, and it lacks adequate facilities, equipment, and skills despite receiving training and equipment through the Project d’Appui à la Gestion Economique (PAGE). The high turnover of magistrates also contributes to the court’s poor performance in handling its caseload. The government has taken a number of measures to improve the performance of the court and other sector institutions. These measures include salary increases to discourage corruption, increases in the number of magistrates, and a reform of the règlement intérieur (internal rules). The percentage of cases whose duration exceeds 60 days decreased from about 42 percent in March 2008 to roughly 22 percent one year later. Moreover, it fell to 14.3 percent and 10 percent in May and July 2009, respectively. 2.556 In an effort to provide quicker delivery of commercial decisions, an arbitration center was created in 2002 and the appropriate legislation was adopted in 2008. Almost 30 arbitrators from various sectors have been trained, and some public education has been initiated. However, although the center has received some start-up funds from PAGE, it lacks the financial resources to further develop this tool for dispute resolution.
A6. Poor Governance 2.57 After a prolonged period of conflict and ethnic tension, improving governance, government decision making, and the capacity to raise taxes are critical to economic recovery. Poor governance and corruption are also major obstacles to private sector development; and they lead to lower investment, both domestic and foreign. Poor governance is difficult to measure; even so, survey-based assessments of perceptions among both citizens and businesses indicate problem areas.
27 2.58 A survey commissioned by the government of Burundi through the Figure 2.5: Governance Indicators, 1998, 2000, 2003, and 2008 World Bank in November 2007 found corruption, access to and abuses of the judicial system, and security and criminality to be among the respondents’ main concerns (Republic of Burundi 2008a; World Bank 2008b). Poor governance has been identified as one of the main reasons for the prolonged civil conflict in Burundi. The same study also found that the majority of households, as well as many representatives of firms and nongovernmental institutions, had doubts regarding the ability and willingness of the government and its political representatives to promote economic and social development. It is obvious from figure 2.5 that the Source: Kaufmann, Kraay, and Mastruzzi 2009. situation in Burundi has improved since Note: The confidence level is 90 percent. the end of the civil war, but the country still ranks among the worst performers in terms of governance.23 The Ibrahim Index of African Governance yields similar results. Burundi ranks 38th among 53 African countries in the 2009 Ibrahim Index, showing some improvement to previous years (Mo Ibrahim Foundation 2009).
2.59 With respect to corruption, the study cited many specific factors that have exacerbated this problem (Republic of Burundi 2008a). These factors include extreme poverty, low government salaries, and limited access to basic and public services, which have worsened incentives with respect to bribes. The survey also revealed that the police force, customs, tax collection, public procurement, and justice system were prone to corruption, with the police force singled out as the most frequent recipient of bribes. Other issues (such as public employment practices) also were identified as areas of concern because respondents saw political party affiliation or personal connections rather than technical competence as motivations for government hiring. Transparency International’s recently released Corruption Perception Index ranks Burundi 168th among 180 countries, representing a considerable deterioration from rank 131 in 2007 and rank 158 in 2008.24 2.60 Public enterprises dominate the economy. A 2007 study (Republic of Burundi 2007a) found that the government had shares in 48 entities (40 enterprises and 8 financial institutions), 16 of which were fully publicly owned. Public enterprises also dominate the export sector (coffee – in the process of being privatized, tea, cotton, sugar). Several public enterprises are faced with a difficult financial situation. They are burdened by the governance problems typically encountered in many countries, including political interference from the government. Their indebtedness increases faster than their turnover, and debt obligations often exceed cash flow. By the end of 2006, the overall debt of the 21
23 The governance indicators presented here aggregate views on the quality of governance provided by a large number of enterprise, citizen, and expert survey respondents in industrial and developing countries. These data are gathered from a number of survey institutes, think tanks, nongovernmental organizations, and international organizations. The Worldwide Governance Indicators do not reflect the official views of the World Bank, its executive directors, or the countries they represent. These indicators are not used by the World Bank Group to allocate resources. 24 For details. see Transparency International, http://www.transparency.org/policy_research/surveys_indices/cpi/2009/cpi_2009_table.
28 largest public enterprises amounted to FBu 128 billion (about 14 percent of GDP), including debt to suppliers, the public sector, foreign lenders, and local banks. The Service Chargé des Entreprises Publiques (SCEP), a special entity created in the 1990s to oversee public enterprises, does not have the capacity and the resources necessary to bring about substantial improvements in the performance of the sector and to plan and monitor the privatization program envisaged by the government. The government recognizes that it needs to retrench from productive sectors, but the existing analytical base does not provide sufficient information to fully understand the financial situation of many public enterprises, its impact on government finance, and the best options for reforms. 2.61 Poor governance and government decision-making capacity imply lower productivity for all sectors of the economy and inhibit new investment in physical and human capital. Given the importance of government spending in the post-2000 economy, increasing the efficiency of spending is necessary to obtain the highest results from aid spending programs. Poor governance is currently an obstacle to this effect. In addition, the government’s capacity to make decisions regarding policy, spending, and tax revenue is weak. This state of affairs fosters a perception that recovery is weak and causes delays in private sector response. The slow decision-making process regarding the reform of the coffee sector is a clear example of this problem and a major determinant of the slower response of the economy of Burundi, relative to Rwanda or Sierra Leone, following the reestablishment of peace.
2.62 Low government decision-making capacity is a critical barrier to development. Low decision-making capacity is often the result of political dissent within a society and of maneuvering by interest groups. But it also is the result of low human capital and disorganization. One strategy that several small states use to circumvent this barrier is to outsource the provision of some government services to regional organizations. Burundi is moving in this direction by joining the EAC. But there are areas of government decision making that cannot be outsourced. Meeting the challenge requires reorganizing current capacity and prioritizing its use.
A7. Low Levels of Education and Health 2.63 To support higher levels of growth and development, one first-order requirement will be to invest in improvements in health and human capital. By most measures, Burundi lags behind the world and behind other countries in the region in these and related areas. Deficits in these areas clearly stem from the country’s low levels of urbanization, income, and development. In this context, Burundi is not likely to meet many of the Millennium Development Goals (which focus on health and education) by 2015 (World Bank 2008b). An estimated 67 percent of the population lives below the national poverty line; and the country displayed a Gini coefficient of 0.38 percent based on data available in 2008, illustrating the fact that poverty is widespread.25 As noted above, Burundi ranked 174th among 182 countries on the United Nations Development Programme’s Human Development Index in 2008. 2.64 Low educational indicators reveal severe limitations to increasing productivity and growth potential. In 2006, only about half of the population was able to read and write in the national language Kirundi, and only about one third was literate in a foreign language (IMF 2009). Although enrollment rates for primary schools have improved tremendously in recent years (to more than 70 percent), partly as a result of the introduction of free primary education in 2005, completion rates have remained low (at about 40 percent) (World Bank 2008b). Secondary enrollment stood at just 19 percent in 2006–07, reflecting high dropout rates, poor infrastructure and access to resources, and high poverty in rural regions. At the tertiary level, evidence suggests that performance is even worse, with university enrollment rates estimated at 2.6 percent—very low even by regional standards. Although private universities have become more prominent in recent years, a review of the PRSP in 2009 emphasized that
25 The Gini coefficient is a widely used measure of income inequality, with a scale ranging from 0 to 1. A score of 0 indicates perfect equality (every household with equal earnings), and a score of 1 indicates the inverse (a single household earns all of the national income).
29 poor academic infrastructure at the tertiary level is a serious impediment to improving educational outcomes and to increasing productivity and economic potential in Burundi. 2.65 Against this background, education is a key means of broadening opportunities for the population, of strengthening integration of the society as a whole, and of encouraging deeper integration between Burundi and the rest of the world. The challenge is how to accomplish this in a country where there are few qualified teachers and the literacy rate in English and/or French is extremely low. In this context, modern information and communication technology and programs designed to promote the role of women in the household may be crucial ingredients in addressing the education challenge. 2.66 This is especially critical in light of the high population growth rate. The fertility rate in Burundi stands at 5.4 children per woman, and the net population growth rate increased to 2.6 percent between 1950 and 2005. The population of Burundi is expected to grow rapidly in the next 40 years. Rapid population growth produces several problems. First, it prevents countries from building their human capital and achieving their “education for all” target, a prerequisite for economic development. Unless fertility levels decline, attaining the Millennium Development Goals will not be possible. Second, high fertility levels jeopardize poverty reduction efforts, particularly among people in poor quintiles. Poor households suffer most because high fertility is an additional burden with regard to access to quality education and health services (World Bank 2007). Third, rapid population growth elevates the pressures that countries face regarding food security, land tenure, water supply, and environmental degradation. Slower population growth eases the security problems that often result from conflicts over scarce resources—conflicts exacerbated by unsustainably high rates of population growth and widespread youth unemployment. Fourth, high fertility levels have detrimental consequences for the health of women and children.
A8. The Challenge of the Demographic Transition 2.67 Today, it is widely recognized that a country’s accelerated demographic transition and faster fertility decline may positively influence its socioeconomic outcomes (May 2009). The demographic transition or modernization is defined here as the gradual shift from high crude birth rates and high crude death rates, in equilibrium, to low levels of fertility and mortality that reach a new equilibrium. In Sub-Saharan Africa, these demographic changes have not occurred in earnest—hence, the sluggish economic growth experienced in this region, as stressed in the seminal study by Ndulu et al. (2007). The relationship between rapid population growth and socioeconomic development is most noticeable in four different areas, namely, economic growth and human capital formation, poverty reduction, environmental protection, and health status. 2.68 The current population of Burundi is estimated at 8.3 million (mid-2009 estimate) and is very young, with 41 percent of the Burundians below age 15 (Population Reference Bureau 2009). The population of Burundi has increased inexorably since 1950, when it was estimated at 2.5 million. It has almost quadrupled in a period of 50 years. The young age of the population has far-reaching implications for the formation of human capital, particularly for education and health. Given the small territory of the country, the already high population density can only be expected to increase. Greater density will add to the pressure on the agricultural sector and ecosystems unless migratory movements take place to neighboring and/or more distant countries. 2.69 In absolute terms, the Burundian rate of net population growth has increased since 1950. It increased from 1.8 percent in 1950 to 2.6 percent in 2005 (United Nations 2009). However, it is important to note that given Burundi’s history of conflict (which peaked in the early 1970s and in the 1990s), the rate of net population growth (taking migration into account) was not regular at all. The periods of conflict are clearly characterized by much lower rates of population growth as a result of increased
30 mortality rates from direct war and war-related causes, probable postponement of births, and migratory movements. 2.70 Mortality levels and fertility are very high, but appear to be slowly decreasing despite sharp variations. The infant mortality, under-five mortality, and maternal mortality rates remain quite high. Life expectancy for both sexes at birth is low: 49 years, compared with the Sub-Saharan African average of 51 years. However, the HIV/AIDS epidemic appears to be somewhat under control. Fertility is estimated at 5.4 children per woman and the slow trend is generally downward, despite some variations over time. Burundi’s total fertility rate is close to the Sub-Saharan Africa average rate of 5.3 children per woman. However, the contraceptive prevalence rate for modern methods was estimated at only 8 percent of married women. 2.71 As a result of mortality and fertility global trends, the population of Burundi will continue to grow rapidly during the next 40 years. The current rate of population growth is estimated at 2.1 percent per year, leading to a doubling time of 33 years (Population Reference Bureau 2009). The medium variant of the United Nations 2008 population projections indicates that the population Figure 2.6: Population Projections, 2005–50 of Burundi should reach 14.8 million in 2050, 18 assuming that the total fertility rate declines from its low variant current level of 5.4 children per woman to 2.3 16 medium variant high variant children per woman in 2045–50 (figure 2.6). 14 However, the 2050 population would be 13 million if fertility declines to 1.8 children per woman in 2045– 12 millions 50, but almost 17 million it declines only to 2.8 10 children per woman during those years. These 8 United Nations projections assume significant improvements in the mortality levels: the expectancy 6 of life at birth would increase from 50.3 years in 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2005–10 to 65.7 years in 2045–50. Finally, net Source: United Nations 2009. migration is assumed be negative as of 2010 onward. However, one could also assume that net migration could become positive if more Burundians decide to return to their country of birth (United Nations 2009).
2.72 Even if fertility declines dramatically and quickly reaches the replacement level, Burundi’s population will continue to grow and probably double during the next 70 years or so. This situation is caused by the youthfulness of the age structure and the phenomenon called the population momentum. The population momentum occurs when the very large number of people in the reproductive age groups continues to fuel the population growth, even though fertility levels decline. 2.73 The trauma caused by Burundi’s checkered political history has put any meaningful discussion of population issues on hold for almost 20 years. As a consequence, population and reproductive health issues have been neglected. First, Burundi has not undertaken or completed the major data collection operations that would have provided an adequate base of knowledge about the demographic situation in the country. Second, the population and reproductive health issues have not been addressed forcefully in Burundi’s development strategies and poverty reduction documents. The Poverty Reduction Strategy Paper has highlighted the issue of rapid population growth, but has not proposed clear strategies and programs to deal with it. Third, although Burundi is preparing a national population policy (with the support of the United Nations Population Fund), there is no guarantee that this document will go much beyond similar population policies drafted in the past decades in Sub-Saharan Africa. Moreover, given the difficult political context, the population policy’s implementation remains a challenge. 2.74 Several steps are necessary to address the challenges related to the demographic transition, including
31 • gathering additional information on demographic trends and gaining a better understanding of the demographic challenge; • repositioning population and reproductive health issues within development strategies and strengthening population institutions; • developing a comprehensive national population policy; • revamping Burundi’s family planning program to accelerate the demographic and fertility transition; • developing new strategies of family planning distribution; and • strengthening the health system, which could bring substantial gains in the supply of family planning services. 2.75 The experience of other countries shows that successful population policies can be implemented in Sub-Saharan Africa too. Necessary, however, is a high level of commitment and a strong implementation of supply-side family planning programs. In Rwanda, a new national population policy is currently being prepared, superseding the one enacted in 2003; and political support for family planning programs is strong. Huge gains in the supply of family planning services have been made possible by strengthening the health systems. As a result, Rwanda made impressive progress in the contraceptive prevalence rate between 2005 and 2007–08, accomplishing more than a third of what is required to reach the replacement level of fertility.26 2.76 Burundi needs to address its high population growth rate. The attention paid, or unpaid, to population issues will predicate the development prospects of the country for many decades to come. The fertility decline will affect most development sectors positively. In particular, it will facilitate formation of human capital, help reduce poverty levels, mitigate the pressure on the environment, and enhance the health of women and children. At this juncture, Burundi needs support from its development partners to design the needed policies and programs to address this fundamental challenge.
A9. Conclusion 2.77 Identifying the most binding constraint to growth by analyzing the economic dynamics of the country has proved difficult. Given the set of constraints Burundi faces, it is difficult to single out one specific area that can be considered the most binding constraint. The key problem areas are linked, creating negative synergies. Low productivity in agriculture, for example, can surely be seen as one of the main barriers to development. Likewise, the large infrastructure deficit (especially the low supply of electricity and the high transport costs) prevents the development of almost all sectors. The small size of the country, coupled with the high population growth rate, necessitates creation of off-farm employment; development of that employment is hampered, in turn, by the undiversified economy, the infrastructure deficit, and low educational attainment. Poor governance and the dismal business environment aggravate the situation by deterring investment, which could create additional employment and income. Burundi therefore needs to address some of the most binding constraints simultaneously. As a necessary prerequisite, peace, security and stability must be maintained.
B. OPPORTUNITIES IN POTENTIAL GROWTH SECTORS 2.78 Burundi is a mostly rural economy. It is highly dependent on the mainly subsistence-oriented agriculture sector. The country relies on small-scale farming and focuses mostly on food crops. Despite increasing demand from a high population growth rate, productivity is low and has barely improved over the past four decades. Food insecurity is widespread, and a large part of the population is undernourished. As a result, there is an urgent need to improve the performance of the agriculture sector to stabilize it and
26 For more details, see May and Kamurase (2009).
32 increase its growth. The same applies to the currently exported crops (mainly coffee and tea), which are suffering from low production volumes and low quality. The productivity of the food crops subsector urgently needs to be improved to increase food security and improve the nutrition of the population. That will lead, in turn, to a healthier and more productive population. 2.79 Burundi’s economy is not diversified. Burundi’s product and export base is currently extremely concentrated. Efforts should be undertaken now to pave the way for the diversification of the economy to achieve more broadly based growth, to reduce the likelihood of negative growth periods, and to create employment opportunities. Burundi has the potential to diversify in different areas, most notably in agro- processing and mining. Although this is an ambitious objective that can only be achieved over the long term, the course should be set now to transform the economy.
B1. Agricultural Recovery: Food Crops—Performance and Challenges
2.80 The World Bank recently Figure 2.7: Proportion of Individuals Living in Households with a Daily conducted a comprehensive study Caloric Intake per Adult below 1,900 kcal addressing the sources of rural growth, with a focus on agricultural recovery (Baghdadli, Harborne, and Rajadel 2008). In addition, the government of Burundi developed a National Agricultural Strategy in 2008 and a National Food Security Program. The following section provides a summary of the main findings and recommendations of these documents. 2.81 The contribution of food crops to GDP is significant. Food crops contribute between 35 and 40 percent of GDP and 80 percent of agricultural output. Food crops occupy about 85 percent of the cultivated land and 28 percent of total land. There is a high incidence of malnourishment, which seriously affects the health and productivity of the population.27 Consumption of own production accounts for nearly three quarters of the average rural household’s income. Poverty is widespread in rural areas, where most of the population lives. 28 Severe land constraints have led to an Source: Baghdadli, Harborne, and Rajadel 2008. overexploitation of the land, which has further contributed to land degradation and soil erosion. 2.82 Burundi’s food supply is not meeting the increasing demand, leading to food insecurity and the need to rely on food aid. Figure 2.7 shows the proportion of individuals living in households with a
27 According to the 2008 Global Hunger Index, the situation is only worse in Eritrea and the Democratic Republic of Congo. The severity of the situation in Burundi is considered extremely alarming (von Grebmer et al. 2008). 28 With 257 inhabitants per square kilometer, Burundi exhibits one of Africa’s highest population density rates.
33 daily caloric intake below the required minimum of 1,900 kilocalories.29 The World Food Program provides monthly assistance to an average of 635,000 people, mostly in the food-insecure north and northeastern areas of the country. 2.83 Production has stagnated (figure 2.8) Figure 2.8: Food Crop Production, 1990–2007 while the population has continued to 4,000 increase, leading to a constant decline in 3,500 production per capita. The main food crops 3,000 (ranked by production) are bananas, roots and 2,500 tubers, legumes, cereals, vegetables and fruits, 2,000 1,500 and oilseed. The use of land is already at its 1,000 limits (using traditional technology), and the thousands of tons 500 average size of a plot is only 0.5 hectare. The 0 soil is characterized by low fertility, about a third of the soil is acidic, and the hilly terrain is
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 prone to soil erosion. The irrigation potential is Cereal Legumes Roots and tubers Bananas mostly underexploited, with less than 10 Source: Republic of Burundi 2008b. percent of the potentially irrigable area under irrigation. 2.84 The agriculture sector is not well organized. The majority of producers in Burundi are households, and only a minority of them has joined a producer organization. The production and distribution of agricultural inputs (seed, fertilizer, crop chemicals, and machinery) are not well organized and managed and thus contribute to low usage and high input prices. Only limited processing takes place (mainly at the household level), and there is almost no industrial processing. Most of the food crops are marketed only at a local level, and food markets are generally underdeveloped. Because most households engage in subsistence farming, only a modest amount of produce gets to the market. Farmers are poorly organized and have only very limited access to information on market developments, opportunities, and prices. 2.85 The lack of infrastructure and extension services hampers the development of the sector. The insufficient supply of electricity and the poor condition of rural roads are obstacles to an increase of production as well as marketing of the products. Extension services, including agronomic research, are weak, as are training, input supply, and distribution. 2.86 Overall, the food crop subsector in Burundi faces a number of challenges that can be grouped into the following five main categories: 1. Agronomic constraints—The fertility of the soil is poor, and degradation of the ecosystem is progressing. There is only limited use of inputs because of a deficient distribution system and high input prices. Food crops production relies mainly on traditional extensive cultivation methods, whereas commercial agriculture based on modern technology and inputs remains underdeveloped. 2. Climatic constraints—Climate change severely affects agriculture in Burundi. Although the country generally receives more rainfall than most African countries, rainfall shortage has been observed, leading to volatile production. 3. Technological constraints—There is weak agricultural research and a poor system to distribute knowledge about innovative technologies. Specific issues include limited use of improved management practices (water management, irrigation, transformation, and conservation of
29 This estimate is made using the Questionnaire des Indicateurs de Base du Bien-Etre (QUIBB) 2006 household survey data (see Baghdadli, Harborne, and Rajadel 2008). The suggested minimum intake is between 1,900 and 2,100 kilocalories.
34 products); limited promotion of alternative and renewable energy sources; and insufficient human resources and laboratories for quality control of inputs, food and cash crops. 4. Institutional constraints–Extension services (including research, postharvest transformation and conservation, production and distribution of agricultural inputs) are inefficient. There is a lack of skills and human capacity. The involvement of the private sector is limited. Weak coordination and lack of harmonization of interventions persists. 5. Socioeconomic constraints—One of the main constraints is a lack of resources to rehabilitate infrastructure and equipment to support agricultural production, reforestation of the damaged ecosystem, and a reactivation of extension services after the long civil war. Furthermore, the population increase puts pressure on the use of existing land. Access to finance is limited for rural households, and the business climate is not conducive to private investment.
Box 2.3: Land Tenure and Economic Growth Uncertain land tenure rights delay the resolution of land-related conflicts. The legal environment fosters uncertainty over land tenure rights. The first obstacle to the implementation of an efficient legal land system in Burundi stems from the coexistence of customary and statute law. For instance, according to customary law, farmers who have cultivated and managed plots over a certain period of time earn ownership rights to those plots. This system has led to harsh conflict over the status of some state-owned lands, particularly in swamp areas. Moreover, customary law in some regions separates ownership of the land from ownership of what stands on the land; statute law does not recognize such a distinction. This difference has fueled bitter disputes over coffee or oil palm tree plantations. Women’s access to property is another matter of great concern because, according to customary law, family land cannot be transmitted to daughters. Statutory law, however, completely contradicts customary law in this matter. Finally, issues pertaining to the resettlement of long-term refugees provide yet another striking illustration of contradictory legal frameworks. According to the Arusha Accords, long-term refugees should be given the right to retrieve their property; if such retrieval is impossible, they should receive fair compensation. However, statute law stipulates that a 30-year prescription applies to land-related claims. The situation is further complicated by the somewhat unclear answers provided by customary law. It does not recognize the 30-year prescription, but it acknowledges that long-term occupants have earned some rights over the land on which they live. In addition, the efficiency of the judiciary and institutional systems is gravely affected by an unclear delineation of responsibilities and overlapping functions. The judiciary system intervening in land disputes relies on three entities: the Bashingantahe, local authorities who arbitrate conflicts using statue law as a reference, and local tribunals (tribunaux de résidence) who return verdicts based on statute law. The institutional level is as confused because at least four different ministries are involved in land-related issues (although functions and prerogatives are often poorly defined and thus generate unnecessary confusion). As a result, local authorities have been implementing ad hoc solutions to land disputes, especially where refugees are concerned, without prior consultation or coordination with governmental entities. Uncertain land tenure rights hamper productivity at the farm level through two channels. First, it reduces incentives to invest in the property. The economic literature stresses the importance of tenure security and land transferability rights for long-term investments. Second, land titling is generally thought to foster access to financial services, although the extent to which and the mechanisms through which it does so are not always well known. Landownership security thus should be given special attention today as the country considers upgrading its agricultural production through specialty coffee and tea and horticulture. Moreover, land tenure insecurity reduces the attractiveness of the economy for both domestic and foreign investors because it involves higher risks and undermines the reliability and credibility of Burundian economic actors. Finally, the development of sustainable land management policies to fight the dramatic land depletion in Burundi will include reinforcing landownership rights. Sustainable land management is indeed a time-consuming, costly, and long-term process that only farmers who are secure in their property rights will be willing to undertake. A global discussion of land issues is essential to determine how land tenure could better contribute to shared growth. Particularly, it will require thinking about ways to enhance wage labor to reduce tensions over access to land property, specifically by diversifying sources of growth outside the agriculture sector (which could be linked to the development of urban centers). Source: Baghdadli, Harborne, and Rajadel 2008.
35 2.87 In addition, uncertain land tenure rights hamper productivity and delay the resolution of land- related conflicts. A global reflection on land issues is essential to determine how land tenure could better contribute to shared growth. In particular, it will require thinking about ways to enhance wage labor to reduce tensions over access to land property, specifically by diversifying sources of growth outside the agriculture sector that could be linked to the development of urban centers. (See box 2.3 for a discussion of land tenure and economic growth in Burundi.)
B2. Improving the Performance of the Coffee Sector 2.88 Coffee has been the main cash crop and source of export revenue in Burundi for a long time. Coffee is produced by 600,000–800,000 households, and it usually constitutes their main source of cash income. However, the coffee sector is suffering from weak performance, with declining production and quality. The involvement of the government has not always been beneficial to the performance of the sector. Coffee has the potential to contribute significantly to the country’s economy and its population’s livelihood. Burundi has the agro-ecological conditions to produce high-quality coffee. The current reform of the sector, including the ongoing privatization process, is a step in the right direction; but more remains to be done, as outlined in this section.
Past Performance of the Coffee Sector in Burundi 2.89 Despite the sector’s potential and the government’s strong involvement, performance remained weak (particularly in the 1980s). Coffee was first introduced by the Belgians in Figure 2.9: Farm Gate Prices for EAC Members, 1977–2008 the 1930s. Its production was then mostly coercive. After the country’s independence in 200 1962, the sector was operated entirely by private stakeholders until its nationalization in 150 1976. In an effort to boost the industry’s revenues, the government deployed substantial 100 efforts in the early 1980s to increase production; but those efforts were only partially 50 successful. Whereas the share of fully washed Price (US cents per pound) 0 coffee in total output increased steadily, the overall results were disappointing. Annual production fell well below the expected output. Burundi Kenya Rwanda Despite an increase in the production of fully Tanzania Uganda washed coffee, which should have led to higher Source: International Coffee Organization. premiums, prices paid to growers stagnated Note: Prices for 2008 are for the first three quarters only. throughout the 1980s (figure 2.9).
2.90 Prompted by the sector’s weak performances and the steep fall in international coffee prices in the early 1990s, the government launched a reform in 1992 that has shaped the organization of the coffee sector to this day. The reform strived to • improve incentives throughout the value chain to promote the production of high-quality coffee; • increase efficiency at the washing station and dry mill levels by calling in private actors; • enhance sale revenues by fostering competition among exporters; and • rationalize government intervention by converting the public entity Office du Café du Burundi (OCIBU) into a coffee marketing board. 2.91 The 1992 reform entailed a downsizing of OCIBU, as well as an attempt to improve efficiency by promoting private sector participation in the industry. Ownership of the 133 washing
36 stations and two dry mills was transferred to a public entity, which leased them on 30-year contracts that remain in effect to this day. All washing stations have been managed since then by five public-private companies, the Société de Gestion des Stations de Lavage (SOGESTALs, or Washing Stations Management Companies); and the dry mills have been managed by the Société de Déparchage et de Conditionnement (SODECO, or Hulling and Processing Company). The SOGESTALs became responsible for providing inputs to growers and financing the purchase of coffee cherries. To improve sales efficiency and transparency, an auction system was set up to replace the state monopoly on exports of Burundi’s green coffee. 2.92 This partial privatization did not provide proper incentives along the value chain. The remuneration system, controlled by OCIBU, initially was aimed at stabilizing growers’ revenues by guaranteeing a floor price to stimulate production. Farmers were paid on delivery at the washing station and could receive a complementary payment when prices were high enough. But the floor rate usually acted as a de facto ceiling price for producers because washing stations were not in competition (each of them operated in an area designated by OCIBU). This had a negative impact on the farmers’ incentives to produce higher-quality coffee. Washing stations and dry mills also had little incentive to increase quality because their percentage fee was set in advance, without regard to actual performance. In addition, the newly introduced auctions failed to improve the marketing process. The licensing system was set up so that only exporters with adequate counterpart risk would be authorized to operate. But the granting of licenses lacked transparency, and the number of bidders was so drastically limited that collusion was almost inevitable—thus capping earnings of upstream actors. 2.93 The outbreak of war in 1993 put an end to the reform process, and the political and economic collapse that ensued further worsened the coffee sector’s performance. The protracted conflict led to the deterioration of existing trees, which were no longer properly cared for, and to a lower replacement rate for aging trees. Washing stations and dry mills also were neglected during that period, leading to the depletion of part of their assets.
Figure 2.10: Burundi Coffee Production, 1977–2008 2.94 As a result, the reforms, flawed at conception and interrupted by the conflict, 45 failed to improve the sector’s performance. 40 Production levels in the 1990s fell below those 35 displayed during the previous decade (figure 30 2.10). Cyclical variations partly explain these 25 low outputs. But smuggling of coffee cherries 20 to neighboring countries (another testimony to 15 the sector’s failure) is also believed to have 10 contributed to artificially decreasing official 5 Production (thousands of tons) production figures. Prices paid to growers 0 remained low. The erratic evolution of international prices for coffee further contributed to decreased earnings in the 1990s and early 2000s. However, growers in other Source: International Coffee Organization. East African countries managed to fare better than those in Burundi. As shown in figure 2.9, farm-gate prices in Kenya, Ethiopia, Tanzania, and Uganda stood well above those in Burundi. At a time when other countries were adapting to changes in the global coffee market (notably, modification of consumer patterns and the rise of specialty coffee), Burundi trailed behind and failed to seize opportunities arising from these new developments. 2.95 Reforms resumed in 2005 in an attempt to complete the process initiated in 1992; but they failed because of the lack of a comprehensive approach. In the first half of 2005, the transitional government issued a series of decrees to liberalize the coffee sector. Prices paid for cherries and washed
37 parchment were deregulated, and private operators were allowed to invest in the industry. In addition, the government no longer guaranteed loans made by banks to OCIBU to finance the coffee campaigns. That measure, however, was not accompanied by a transfer of property rights over the washing stations to the SOGESTALs, putting them in a difficult situation: they were expected to finance the purchase of cherries and other expenses, but had no access to working capital because they had no collateral to offer banks. OCIBU’s role was also reduced by ministerial decree to that of a regulatory agency. It still remained responsible for holding the auctions through which green coffee was sold. Other issues (such as the suspected collusion among exporters, in particular) were not properly addressed. The series of ad hoc measures signed into law in 2005 thus failed to reform the coffee sector because they did not provide a comprehensive deregulation framework. 2.96 The lack of a comprehensive and meaningful reform has impeded the coffee sector’s revival. Although production soared in 2008, it stood at only 35,000 tons, the average level displayed by the sector in the 1980s (figure 2.10). Moreover, prices paid to growers remain among the lowest of the region (figure 2.9). Aging coffee trees are becoming a major concern, further increasing cyclical variations. Incentives offered to growers remain insufficient to induce an increase in output and quality. The SOGESTALs and SODECO are in a dire financial situation. Low production volumes in the past years have undermined their capacity to repay loans incurred to finance the campaigns. The banking sector is now reluctant to extend new loans to the industry, and recent campaigns were marked by tough negotiations in which the government intervened despite the theoretical deregulation initiated in 2005. In addition, the ill-adapted remuneration grid does not always enable the washing stations and dry mills to cover their operating costs. 2.97 In December 2008, the government adopted a privatization strategy for the coffee sector, following lengthy consultations with stakeholders. According to the strategy, washing stations will be put up for sale in groups of three to six. These relatively small entities are expected to raise competitiveness through closer collaboration with growers. They are believed to be better positioned than the SOGESTALs (each of which manage 20–30 washing stations) to enhance quality and support the introduction of differentiation and traceability systems that are indispensable in accessing specialty markets. By putting washing stations in competition, this new organization is intended to benefit growers. To some extent, growers should be able to choose the stations to which they will sell their cherries, thanks to increased price transparency. The sale strategy also would include two dry mills in good condition, despite their oversized processing capacity and high operating costs. Their production systems are flexible enough to adapt to the handling of smaller volumes of parched coffee. The privatization strategy includes some innovative features that promote more inclusiveness in the Burundi coffee sector. For example, it has opened up 75 percent of the washing stations’ equity to private sector participation, while allotting the remaining 25 percent to producer associations. 2.98 The strategy is currently being implemented, envisaging complete privatization. In June 2009, the government of Burundi decided to launch the tender for the sale of 117 washing stations, without first amending the convention between OCIBU and the SOGESTALs (as had been requested). The government made this exceptional move to avoid derailing the reform process, which had already experienced a long delay. At the closing of the bidding process in August 2009, four potential contenders responded to the tender for the sale; and three of them presented offers that met the technical and financial requirements to acquire 6 of the 26 lots of washing stations. Following lengthy discussions between the government and SOGESTALs, the OCIBU-SOGESTALs convention was amended on August 20, 2009. Based on this amendment, SOGESTALs would continue to manage unsold washing stations. This arrangement laid the legal ground for bid winners to access the washing stations without contest by farmers and their organizations. In November, 13 washing stations (out of 117) were transferred to the bidding winner (leaving 104 unsold stations under SOGESTALs management, based on the terms of the recently amended convention).
38 2.99 The first phase of the privatization process has ended with mixed results. It could be expected that an internationally recognized company effectively managing 13 washing stations would offer an excellent alternative to SOGESTALs, and would force them to improve the remaining poorly managed washing stations. The new private investor could help address the credit and payment issues and deliver additional services in the area of intervention to influence the overall way of doing business in the coffee sector. The new context (with few privatized washing stations) can be viewed as a pilot phase from which lessons may be drawn to improve future transactions. This first phase also can lay the ground for an environment conducive to a second bidding round, particularly after the elections when political pressure on the government is reduced. Even a newly elected government could feel more comfortable to accelerate the reform if the experience is deemed successful. Conversely, poor first-phase privatization outcomes may put an end to the overall process and possibly lead to the return of the government to manage the sector. 30 2.100 The success of privatization will also depend on the government’s ability to set up an efficient regulatory framework. The coffee sector will be able to improve its performances only if rules are adequately designed, clearly communicated, and fully implemented. Investors will be willing to participate in the privatization process only if a proper regulatory authority has been established and market rules have been set. Until today, the industry has been organized primarily by OCIBU. Despite recent reforms, OCIBU fixed floor prices for coffee and was unable to prevent collusion among exporters of green coffee. The dissolution of OCIBU and the creation of a regulatory agency within the Ministry of Agriculture were accomplished in early June 2009. The creation of the Coffee Sector Regulatory Agency (ARFIC) has strengthened the role of the private sector. The first priority for this agency is to determine competition rules to govern the industry. Its responsibilities include establishing quality norms and classification systems, controlling traceability procedures, and delivering certifications. 2.101 Delays in implementing necessary reforms jeopardize the coffee sector’s chances of recovering from decades of weak performance. It is imperative to move forward decisively. Although there appears to be a consensus on the privatization of the sector, disagreements pertaining to the modalities of this privatization have slowed the process. Meanwhile, the industry is suffering from the depletion of its assets (for example, aging trees, outdated processing equipments, and neglected feeder roads connecting farmers to washing stations). But the biggest threat to the sector’s recovery is the risk of a general fatigue among stakeholders. Privatizing and liberalizing are only the first steps in the direction of the sector’s rebound. Much remains to be done— for example, organizing growers into competent producer associations, promoting market intelligence, branding Burundian coffee on international markets, and modernizing production and processing procedures to enhance quality. It is more urgent than ever before to lay the foundations for a thriving coffee sector.
New Opportunities from Recent Global Market Trends 2.102 The world coffee market has changed considerably since the early 1990s. Global consumption increased steadily over the past 15 years, reaching a total of 128 million bags in 2008. In the past eight years alone, it grew by 2.5 percent a year, on average. Consumption in emerging economies and producing countries accounts for part of this upward trend. Demand in major consuming countries has increased—sometimes substantially—in recent years. Nevertheless, Prices have fluctuated widely over the past 15 years because supply did not always meet demand. 2.103 The emergence of specialty markets profoundly modified the global coffee industry. These niche markets include gourmet, fair-trade, and organic coffee for either in-home or out-of-the-home consumption. Targeted customers value high-quality coffee and flavors linked to specific origins. They are usually willing to pay a premium that guarantees that coffee was produced under certain conditions
30 The EAC is the regional intergovernmental organization of the Republic of Kenya, Uganda, the United Republic of Tanzania, and (since 2007) the republics of Burundi and Rwanda.
39 and that growers were offered a fair price—or, more generally, that labor and environmental norms were respected. Specialty coffee markets have developed at a fast pace. In the United States, the industry weighs more than US$20 billion a year and is growing at a 9 percent annual rate. It also has benefited from the expansion of emerging economies, such as China where specialty coffee sales increased by 90 percent between 1998 and 2003 (Clay, DeLucco, and Ottaway 2007). As an example, Starbucks, the world’s leading specialty roaster and retailer, has seen its volumes of roasted coffee rise steadily from less than 100,000 bags in 1995 to 3.5 million bags in 2008. This shift in consumption patterns altered supply and demand dynamics in the coffee sector. Marketing of high-quality coffee is now more and more frequently channeled through direct sales from producers to roasters who develop long-term business partnerships and agree on higher premiums.31 2.104 Burundi has the potential to become a noteworthy player in specialty markets. Commodity coffee, however, should not be neglected because it will continue to account for a significant share of the country’s output for some time. Burundi is endowed with an ideal climate and terrain to grow high- quality Arabica coffee beans. Recent visits from international roasters and retailers, who expressed a strong interest in Burundi’s coffee, demonstrate the country’s potential to access specialty markets. Starbucks offered coffee from Kayanza in very limited release in 2008, after collaborating with the region’s SOGESTAL. In principal, Burundi’s coffee industry has the necessary experience and assets as a basis to shift toward specialty markets. To capitalize on the potential, however, the following measures must be taken: • Quality standards must be raised and stabilized to access specialty markets and increase earnings from the sale of commodity coffee. First, the pricing system should be revised to improve incentives so that growers actually gain by an increase in quality and, importantly, so they fully understand how added value is shared along the value chain. As a second priority, access to finance for farmers needs to be facilitated. This would promote the use of inputs and facilitate the replacement of aging trees, which would positively affect both yields and quality. • New market demands will require processing facilities to pay greater attention to tasting qualities. Burundi’s coffee sector is currently structured around the selection of coffee according to grading characteristics, which depend mainly on bean size, color uniformity, and defect count. Specialty markets assume these criteria are met, but go one step farther to differentiate coffee according to flavor profiles to match consumer demands. Hence, although selecting high-grade beans remains fundamental, greater emphasis needs be placed on high-quality flavors. • Burundi will have to adopt traceability and differentiation systems if it is to access specialty markets. In these two areas, the coffee sector will be starting from scratch. Ideally, output from small groups of growers will be treated separately from the time it is brought to the washing station onward. Processing coffee lot by lot enables washing stations to distinguish excellent batches from poor ones and thus channel the excellent batches to specialty markets and the other batches to the commodity market. Being able to trace coffee to its origin is also important for marketing purposes, notably to communicate the product’s story. Finally, quality control and coffee differentiation will require an expansion of the country’s cupping capacity.32 • To deal with the growing complexity of global coffee markets, Burundi must rethink current marketing practices. Quality improvements will be fruitful only if they are complemented by carefully designed marketing strategies aimed at (1) improving revenues from the sale of commodity coffee, and (2) penetrating specialty markets. Proper gathering of market information and, more important, its dissemination along the value chain are indispensable to understanding
31 Other elements, such as availability and reliability, affect price differentials applied to green coffee according to its country of origin. For instance, Burundi’s coffee is currently offered at a discount of 100 points (that is, minus 1 cent per pound) on the New York Coffee Exchange. 32 To this day, there is only one cupping laboratory in Burundi, at OCIBU.
40 price trends and supply and demand dynamics. Suboptimal decisions could thus be prevented.33 As for specialty markets, they rely increasingly on the development of long-term partnerships between producers and roasters or retailers. Countries that have successfully operated a transition toward specialty markets, such as Rwanda, usually have managed to build a strong image of their produce and industry to attract such partners. This is all the more important for Burundi because its competitors enjoy a 15-year head start in producing specialty coffee.
B3. Broadening the Base for Economic Growth 2.105 Recent research confirms that there is a strong correlation between diversification and long-run economic growth.34 Developing countries that have successfully achieved and sustained high growth rates have typically been successful in diversifying the structure of their economy. Imbs and Wacziarg (2003) established a U-shaped relationship between diversification and GDP per capita, finding that countries first diversify until they reach a certain developmental level; then they start to specialize again. Countries that have achieved a high level of product sophistication relative to income level have been found to grow more than the average (for example, China and India) (UNIDO 2009).
2.106 The key findings regarding the diversification of the economy are the following: • It is not necessary to move directly from primary products to manufacturing of high-tech products. On the contrary, it is advisable to diversify unsophisticated products first and increase their market share before making a move to more sophisticated products. Hausmann, Hwang, and Rodrik (2007) have developed several measures to evaluate the income level associated with a certain product. Their findings show that high-value primary products, such as fresh milk or olive oil, can be associated with a high income level similar to that for motor cars. • It is also neither necessary nor sensible to jump directly to new products. Instead, the volume of products already produced and exported (although currently in small amounts) can be expanded. When considering new products, picking a “winner” (a product that will survive on international markets and contribute substantially to Burundi’s economy) has a chance of success only if there is sufficient information to understand why that specific sector will be doing well (for example, see Brenton and Newfarmer [2007]). • An increase and diversification of exports can lead to higher productivity and growth through “learning by exporting” (Page 2009). Firms that produce goods for the export market have higher levels of productivity than those producing goods only for the domestic market. One reason is the knowledge and skills transfer from international partners and buyers. In addition, a self-selection takes place, allowing only firms with higher productivity to successfully access international markets. Although it is an important and valid option in the short run, a focus on the (small) domestic market is not sufficient for broad-based development over the long run. A strategy that first targets regional markets to acquire the necessary experience before entering the global market seems appropriate. • Ultimately, a shift to manufacturing is recognized to have great potential. High GDP growth rates in low-income countries typically have been supported by high growth rates of manufacturing value added per worker. Although the direction of causality between high GDP and the growth of manufacturing value added is not entirely clear, research suggests that manufacturing growth and structural change generally lead economic growth. Shifting labor and capital from low- to high- productivity sectors increases the overall productivity in the economy and thereby subsequently
33 One example of a suboptimal decision is the 2007 granting of a monopoly on the sale of Burundi’s green coffee to a single New York–based stockbroker. 34 For example, see UNIDO (2009), Hesse (2009), or Lederman and Maloney (2009). It should be noted that correlation does not necessarily imply causality.
41 increases the income level (for examples, see UNIDO [2009] and Hausmann, Pritchett, and Rodrik [2005]). Manufactured exports are often labor intensive, thereby creating off-farm employment opportunities. The labor intensity can also be an equalizing factor, creating employment for the rural population as well as for women.35 • Resource-based products, however, should not be discounted as a potential source of growth and diversification in the short and medium terms. According to Page (2008), resource-rich countries have led Africa’s recent growth turnaround. Although resources can be depleted, they provide an opportunity to diversify exports and generate government revenues (which, in turn, could be invested to improve infrastructure in a way that benefits the whole economy). Natural resource production and exports, however, typically offer less employment opportunities and added value. 2.107 Economic diversification is particularly important for Burundi. Given the currently projected demographic trends and the scarcity of land, Burundi needs to gradually move away from traditional agriculture because it adds little or no value as a main source of income in the long run. In the short to medium term, traditional agriculture will be essential for the development of Burundi; but it will not be able to support the growing population (see section A of this chapter) and contribute to sustained high growth over the long term. 2.108 A diversified economy can create much-needed employment, especially off-farm employment. With a more diversified product base, Burundi could achieve broader and more inclusive growth and reduce its vulnerability to external shocks. Although the transformation of the economy will not be achieved in the next few years, implementing policies that facilitate diversification should be initiated now. 2.109 In this chapter, Burundi’s potential to diversify its export base is analyzed and recommendations are given to ease the process of structural reform. Several policies are needed to support a successful diversification strategy, including addressing weaknesses in infrastructure, business environment, skills, and institutions; and benefiting from regional integration arrangements (discussed in chapter 3).
Which Products Are within Reach? 2.110 Given Burundi’s current export structure, what products within reach could significantly contribute to higher export earnings, increased product diversity, and ultimate development of the economy? To help answer this question, Hausmann and Klinger (2006) have developed the product space methodology that maps the distance between products. Hidalgo et al. (2007) visualize this approach in a network representation. The densest part of the product space is typically made up of manufactured products, whereas unprocessed agricultural goods tend to be in the sparsely populated part. Being in the dense part of the product space—that is, close to a range of other products—makes it easier for a country to venture into new products. Hidalgo et al. find that, unlike other regions, Sub-Saharan Africa exports few product types, all of which are on the periphery of the product space. 2.111 Burundi’s products are placed on the periphery of the product space, complicating the move to new products. The product space map for Burundi (highlighting the position of the products identified in the previous section as classics, disappearances, and emerging champions) is plotted in figure A.1 of appendix 2. It becomes clear that all of these export products lie on the periphery, far from the dense part of the product space. That peripheral location makes it difficult for Burundi to move into new products and develop an RCA in them. 2.112 Several products have been identified as potential export products. Using different measures developed by Hausmann and Klinger (2006), and the location of Burundi’s products in the product space, a number of products can be identified as close to the current classic export products and therefore easier to reach than products that are far from the current exports. These products can serve as starting points in
35 According to Ross (2008), labor-intensive manufacturing is an important source of wage employment for women.
42 addition to the already identified emerging champions. Furthermore, the existing classics, which represent only a small share, can be expanded.36 Products that have been identified include • frozen fish (excluding fillets), crustaceans, and mollusks; • beans, peas, lentils, and other leguminous vegetables; • fruit (fresh or dried); • tobacco; • live animals, including zoo animals (mainly ornamental fish); • precious and semiprecious stones; and • sugars. 2.113 These finding are in line with the products that Burundi has identified as diversification opportunities. In its 2006 PRSP, the following areas were listed as export enhancing: coffee; tea; cotton; mining; fisheries; tourism; handicrafts; and nontraditional exports, such as fruits and vegetables, flowers, ornamental plants, aromatic and medicinal plants, and palm oil. This is also in line with the recommendations of the National Agricultural Strategy from 2008 (Republic of Burundi 2008b), identifying coffee, tea, cotton, cinchona bark, and horticulture products as the main products to be promoted as exports. A similar list (with the addition of sugar cane) can be found in a report evaluating the requirements to meet the Millennium Development Goals (Republic of Burundi 2007b). A strategy for Burundi’s commercial and industrial development, drafted by the OTF Group (2008) with the support of PAGE, suggests that the country should focus in the short term on five main export subsectors: coffee, tea, tourism, horticulture, and mining. In their study on sources of rural growth in Burundi, Baghdadli, Harborne, and Rajadel (2008) identify three subsectors with high growth potential: coffee, tea, and horticulture. Among horticulture options, promising products include specialty vegetables and fruits. 2.114 Detailed information about the prospective sector and product is essential. It is important to note that government intervention to identify new export products is unlikely to work if it is not based on good and detailed information about why the sector and the product is likely to be successful. Moreover, the chance of success is higher if Burundi already has some experience in producing and exporting the product (as is the case for marginals and emerging champions). A sequenced and gradual transition needs to take place. 2.115 In summary, the following product groups emerge as the intersection of the product space analysis with the strategies developed by the government of Burundi, if the main classical exports (coffee, tea, gold, cotton) are excluded: fishing, horticulture, and mining.
Fishing 2.116 Burundi currently produces fish mainly for domestic consumption, and it exports only small amounts. The country has favorable natural conditions, and fish can become an important contributor to exports just as it is a source of nutrition for the domestic population. 2.117 Fishing supports a substantial number of people, but fish production is stagnating. The fishing sector supports about 300,000 people who live primarily in the poorest communities (Republic of Burundi 2008b). Fish consumption is currently below that of the sub-region (FAO 2003) and decreasing as a result of stagnating fish production and population growth. The fishing sector is divided into semi-
36 Based on the concept of proximity, Hausmann and Klinger (2006) further construct a product-specific measure (called path) that is the distance-weighted number of products around a certain product. In addition, they develop a country-specific measure (called density) that measures how close a specific product is to the current exports of the country and how easy it is for a country to develop and maintain an RCA in that product.
43 industrial, artisanal, and traditional fishing (with artisanal fishing accounting for more than 90 percent of the fish caught).37 The amount of fish produced in fish farms is very small. 2.118 The fishing sector faces several constraints. The main constraints are the weak institutional capacity of all actors in the sector; degradation of shipping equipment resulting from fishermen’s lack of access to finance; noncompliance with sanitary standards by unloading, drying, and smoking facilities; pollution; overfishing; and generally inadequate management of fishing resources.
Horticulture 2.119 The horticulture sector is an important source of diversification and of income and employment creation. Products that have been identified include fresh and dried fruits and vegetables, cut flowers, ornamental plants, and medicinal plants (mainly cinchona bark). 2.120 The horticulture sector has great potential for a gradual transformation from primarily unprocessed products to agro-processing. During the transition, the necessary know-how, skills, and equipment can be acquired. Agro-processing includes the transformation, preparation, and preservation of agricultural products (following harvest). It is one of the main sources of employment and income creation in the manufacturing sector, especially in developing countries that depend predominantly on the agriculture sector (UNIDO 2009). 2.121 A move into agro-processing can be especially advantageous for a country like Burundi, where the majority of poor people live in rural areas. Agro-industries can be decentralized and are primarily based in rural areas. Therefore, they are an important tool to promote socially inclusive growth by creating farm and nonfarm employment and income and by raising the living standard of the rural poor. Creating nonfarm employment is essential for Burundi, given its high population growth rate. 2.122 The potential of the horticulture sector has been recognized. In all strategic documents produced by the government and in World Bank analyses, horticulture has been uniformly identified as having the potential to contribute significantly to the diversification of Burundi’s product base. The products can be grouped into the two main categories: fruits and vegetables, and flowers and plants. The following specialty fruits and vegetables have been identified: pineapple, avocado, passion fruit, small bananas, papaya, mango, cherimoya, green podded peas, baby vegetables, leafy greens, selected roots and tubers, and macadamia. Flowers and plants include roses, dracaena, and cinchona bark, among others. 2.123 The horticulture sector faces several constraints. The National Agricultural Strategy (Republic of Burundi 2008b) lists a number of constraints to the development of the horticulture sector, including • the lack of “market knowledge,” leading to suboptimal strategic positioning of products on international markets; • insufficient knowledge of international norms and standards, particularly phytosanitary standards; • the lack of storage facilities and availability of air transport; • producers who are not well organized; and • lack of extension services and training for producers.
Mining 2.124 Mining can be an important source of income—especially the mining of ores and concentrates of nonferrous base metals, including nickel, coltan, and cobalt. The mining sector is discussed in more detail below.
37 On average, only about 13,000 tons of fish have been caught annually between 2001 and 2005, a significant decline from about 21,000 tons in 1990–1995 (Republic of Burundi 2008b).
44 B4. The Potential of the Mining Sector to Contribute to Growth 2.125 Burundi is believed to have considerable mineral resources. The mining sector, however, is dominated by some 20,000 artisanal producers, with no industrial-scale production. It performs below its potential. The development of the sector has been held up for years by the civil war. The government recently has undertaken to reform the legal framework to make it more conducive to the development of the sector. A number of international companies have been interested in investing in commercial exploration of the mineral resources. If these reform and exploration efforts prove successful, one or more mineral deposits could be developed on an industrial scale. This would provide the economy with a significant boost in terms of production value, tax receipts, employment, related infrastructure, and local economic development. However, all of this depends on international market prices and requires that Burundi overcome infrastructure obstacles. Even if major investments seem unlikely in the short term, mining deserves special attention from the government because of the potential contribution of large-scale mining to the country’s future growth, and because current artisanal and small-scale mining needs to be improved.
Situation of the Mining Sector 2.126 Mineral production accounts for a small part of Burundi’s GDP.38 Production is limited to gold, colombo-tantalite (coltan or tantalum ore), and relatively small quantities of cassiterite (tin ore), kaolin (china clay), limestone for cement, construction stone, sand and gravel, and peat for fuel. The most important mineral potential is in nickel, gold, and coltan; but the country also has resources of other metals and industrial minerals like vanadium, phosphate, and rare earth metals. 2.127 The number of nickel deposits known in Burundi suggests good potential. Those deposits are located all along a strip of land 20–40 kilometers wide, stretching from Mugina in the extreme south to Muremera to the northeast. Deposits have been found from the two main types of nickel deposits that can be found globally—namely sulphide and laterite. The recent price collapse has had a major impact on nickel supply, and many projects have been either put on hold or cancelled. However, the global demand for nickel, closely correlated with global industrial production, is expected to resume and continue to grow over the long term. 2.128 Musongati is the main laterite nickel deposit in the country. Located in central Burundi, some 85 kilometers east-southeast of Bujumbura, it is the most explored laterite nickel resource. Musongati is among the 10 largest undeveloped laterite deposits known worldwide. It has an inferred resource of some 180 metric tons. Musongati can be compared in size with other laterite projects recently put forward: Ravensthrope (Australia), Ambatovy (Madagascar), or Goro (New Caledonia).39 A comprehensive mining convention, negotiated with an international company in 1999, was cancelled by the government after the company declared force majeure for security reasons in 2000. The exploration license was granted to another international mining company at the end of 2008, and several international companies are actively involved in discussions with the government about the further development of the mine site. 2.129 Muremera is the most well-known nickel sulphide prospect in Burundi; and may be the same ore body as the Kabanga deposit in Tanzania, a couple of kilometers across the border. A cross-border mining operation might be possible, even though the feasibility of the Kabanga mine as a stand-alone project is being evaluated and the ownership of both projects is separate. At the time of its discovery in the 1980s, Muremera was said to have reserves of as much as 30 million tons of nickel sulphide ore. An international company secured 100 percent ownership of the Muremera project in March 2009, and continues exploration for nickel sulphide.
38 Mining and energy accounted for about 1 percent of Burundi’s GDP in 2008, according to the latest figures from the International Monetary Fund. 39 This comparison is theoretical because resources/reserves of those projects do not present the same level of certainty. In the case of Musongati, contrary to the three others, we believe resources are still classified as “inferred.”
45 2.130 Nickel exploration has been quite dynamic on other prospects. Important lateritic nickel resources include Nyabikere and Waga, which have combined inferred resources of 81.4 metric tons. The holder of the Musongati exploration permit also has the permit for those two potential deposits. One of the world’s largest producers of primary nickel recently has acquired permits to explore the nickel sulphides of Rutovu, Buhoro, and Bukirasazi, which could be seen as an encouraging sign in the current global outlook. 2.131 Gold is only exploited on an artisanal and small scale. This has been true since colonial times in many places across the country. A few small-scale companies, associations, or comptoirs exploit gold legally. However, most of them work outside the legal framework set by Law 1/015 of 2000, the most recent law dealing with artisanal mining, mineral trading, and exportation. Gold production represents a significant but informal contribution to the economy. Most of this gold does not pass through the formal customs system, and a large part of the exported gold comes from the Democratic Republic of Congo (Global Witness 2009). If gold prices can be expected to remain at historically high levels, above US$750 an ounce on a nominal basis over the coming 5–10 years (CPM Group 2009), the contribution of gold to Burundi’s economy can be significant. 2.132 No large-scale gold operation is currently being planned. So far, the only gold exploited in Burundi derives from alluvial deposits found in the riverbeds. These deposits are said to be of relatively low grade. However, the government has prospected attractive indexes in the northeast. The Burundi Mining Corporation (BUMICO), a government-private venture, was exploring the possibility of producing gold on a commercial basis at Muyinga. The government’s efforts to promote the country’s gold potential have attracted several exploration junior companies. One international company has an exploration permit in the Mabayi-Butara region; but as part of its cost reduction measures, it has put its Burundi operations on hold. 2.133 Coltan has been exploited by artisans or small companies in Burundi since the 1930s. Deposits are known in the northeast and north of the country. Informal mining still occurs in rivers or alluvial areas in several provinces. Comptoirs Miniers des Exploitations du Burundi (COMEBU), which has a 25-year concession around the Kabarore mine, is the only legal operation. The coltan artisanal production level is closely related to the tantalite price. Artisanal and small-scale coltan mining has contributed modestly but constantly to Burundi’s economy. Development of small-scale mining will be limited by the bad image of “conflict coltan” in the Great Lakes region. Indeed, major users of tantalum ore have committed not to acquire any material containing tantalum from countries in the Great Lakes region. Development of large-scale tantalum mining is unlikely because there is little room for newcomers in the supply market. 2.134 Cassiterite (tin ore), and wolframite (tungsten ore) often accompany coltan. Unlike gold miners, a single company is said to organize the production of artisanal tin miners. COMEBU is the only company that has a mining concession, and it operates small mechanization. Some unorganized cassiterite artisanal miners still exist. According to the Ministry of Mines, declared production has been very erratic, from less than 10,000 tons in 2003 to more than 100,000 tons in 2008. 2.135 Other metals and industrial minerals include vanadium, rare earth metals, phosphate, platinum-group metals, carbonate materials of the type needed to produce cement, kaolin and peat. However, further exploration work must be carried out for most of those materials. Hydrocarbon research operations were conducted in the 1970s and 1980s in Lake Tanganyika and in the Ruzizi plain. Hydrocarbon indicators were detected in the depths of the basin. Promotion activities are ongoing to attract companies to pursue exploration in Lake Tanganyika.
Potential Contribution of Mining to the Development of Burundi 2.136 The mining sector has been identified as a potential source of growth. The disparity between the contribution of the mining industry to the national economy and its larger potential is partly caused by
46 Burundi’s landlocked location, the inadequate export infrastructure in surrounding countries, and the protracted war. However, the government has identified the mining sector as a potential source of diversified growth and income creation. The 2006 PRSP (Republic of Burundi 2006) envisages activities to (1) reinvigorate research, (2) favor the development of artisanal and semi-industrial activities, (3) develop a mining code attractive to investors, and (4) attract international investors to exploit some of the mineral deposits. In 2008, the government launched a revision of the mining legislation and regulation. This legislative reform should allow the submission of a new mining code with a new model convention agreement in 2010. 2.137 Substantial benefits can accrue to Burundi from a properly structured and administered mineral industry. In addition to providing foreign exchange earnings, mining activity will produce additional revenue through taxes and royalties, improve the professional and technical skills of nationals, and provide a nucleus for regional and local economic development. Successful large-scale development of Burundi’s mineral potential would require investment in energy production and export roads, but artisanal and small-scale mining can already be improved to provide more benefit to the economy.
Contribution of Industrial Mining 2.138 Mining could become the second most important sector after agriculture. Assuming that a commercial deposit exists and that mining companies decide to go ahead, mining could have a major impact on growth and would probably become the second most important sector after agriculture. A few years after the investment decision, export revenues and government tax revenues would jump to new levels. A more in-depth projection would have to assess the percentage of growth that this would imply: it would have to evaluate the added value of each mining project—not only the production value, but also all possible economic, social, and environmental benefits and costs related to such an investment. 2.139 Two different scenarios have been considered—namely, development of a large-scale nickel mine and of a large-scale gold mine. The assumptions and projections are summarized in table 2.4. The macroeconomic contribution of a gold mine would be significant when operating at full capacity. It would represent between US$39 million and US$48 million of production value, and between US$7 million and US$12 million in tax revenues annually. The main part of production would likely be exported, so it would have a direct effect on export earnings. In terms of direct employment, such a project could create a limited number of jobs, possibly several hundred. Indirect employment could vary a lot and typically can be increased by helping local entrepreneurs create businesses and supply the mine. The life span of such a gold mine is also likely to be limited to 5–15 years, compared with the minimal 20–25 years of a nickel mine.
Table 2.4: Annual Projections with One Large Scale Nickel or Gold Mine
Nickel mine Gold mine Assumptions Capacity 50,000 metric tons 2 metric tons Operating costs US$3.2 per pound US$400 per ounce Taxation royalty 3%, income tax 30%, participation of state 15%, dividend tax 15% Price US$4 per pound US$7.3 per pound US$600 per ounce US$750 per ounce Projected outcome (annual cash flow, US$ millions) Sales 441 800 39 48 Operating profit 88 447 13 23 Net distribution to investor 38 214 6 11 Total government cash flow 50 233 7 12 Effective tax rate (%) 57.0 52.1 54.0 52.7 Sources: Authors’ calculations based on AfDB (2009) and on a model created by Robert Parsons in work presented to the government of the Lao People’s Democratic Republic in September 2008. Note: The projections of the nickel mine exclude the potential contribution of cobalt as a by product.
47 2.140 The Musongati nickel deposit has the greatest potential to be exploited on an industrial scale in the near future. Two possible options for the actual mining operation have been proposed at one time or another: one is the export of nickel ore; the other is the export of metal from a refinery at the mine site. The first option would involve transporting some 4 million tons of ore each year from the mine site to the Port of Dar es Salaam for shipment overseas to a refinery. These large volumes would require access to the rail transport network of Tanzania. This scenario therefore depends on completion of one of the current options for the proposed public rail extension from Tanzania into Burundi, and construction of rail spurs to the mine sites. The other option is to refine the ore at the mine site and transport refined metal to the coast for shipment abroad. If the ore is refined at the mine site, the quantities of metal to be shipped are estimated at about 50,000 tons a year. In this scenario, transporting the metal by road to the railhead at Kigoma is seen by industry analysts as the preferred option. The working assumptions are that the Musongati ore field will be developed and brought into production by 2017, nickel and cobalt will be refined at the mine site, and the mining company will ship the refined metal by road to the railhead at Kigoma. 2.141 The macroeconomic contribution of this nickel mine and processing plant would be huge. It would represent between US$397 million and US$800 million of production value40; and between US$50 million and US$233 million of tax revenues annually when operating at full capacity, depending on the nickel price assumption (table 2.4). It is assumed that production starts by 2017. The production would likely be totally exported, so it would have a direct effect on export earnings. In terms of employment, such a project could create 1,000–2,000 direct jobs, and 2,000–5,000 indirect jobs. To those permanent jobs should be added the numbers of short-term jobs that would be required during the construction phase. A portion of that workforce inevitably would come from abroad; but if training campaigns can be organized in advance, the local population could benefit from a large number of those employment offers. The effective tax rate illustrates the “fairness” of the deal between the investors and the government. It is given here only as an indication because the assumptions are too broad to provide a realistic rate in absolute terms. It is still possible to interpret the decrease of the tax rate when prices increase as a reason to introduce a possible windfall tax41 to better share exceptional benefits in periods of high commodity prices. 2.142 In addition to the macroeconomic contribution, mining development can stimulate local economic development. The construction of one or more large mines and the infrastructure to support them would produce many positive local and regional effects on employment, the development of supply industries, and community development around mining. The government must take measures to actually optimize this contribution to the local economy by encouraging corporate social responsibility or training and capacity building of the workforce. Without such measures, mining development may have several negative effects on the local population, including disruption of the indigenous land usage, inflation, pollution, prevalence of certain diseases (HIV/AIDS), political friction, and the like. Moreover, measures must be taken to ensure that local communities receive part of the benefits—for example, channeling part of the mining taxation revenues to those communities, either through decentralized authorities or through dedicated instruments like mining foundations, trusts, or development funds.
Contribution of Artisanal and Small-Scale Mining 2.143 Mining employs a substantial part of the population. A recently conducted survey of the entire artisanal and small-scale mining sector finds that the sector currently employs about 20,000 people; approximately 10,000 of those workers are mining metals and the other 10,000 are mining construction material (Midende 2009). Assuming an average family size of 10, this implies that roughly 200,000
40 The value added is not estimated here; therefore, we cannot estimate the percentage of the GDP that such a production would represent. 41 A windfall tax is a higher tax rate on profits that ensue from a sudden windfall gain to a particular company or industry. Many governments explored the possibility of including such a tax for the mining sector following the 2007–08 commodity boom.
48 Burundians are affected by mining. Income from mining usually complements income from agriculture; and mining is typically carried out in the dry season, when the intensity of activities in the agriculture sector decreases. Mining also offers possible employment for numerous demobilized soldiers. 2.144 Official and survey production statistics differ widely, making it difficult to assess the impact of artisanal and small-scale mining on the economy. Table 2.5 contrasts official production statistics with those determined by the survey of the industry. It becomes clear that most of the official gold Table 2.5: Comparison of Annual Official Exports exports are not being produced in Burundi, but rather are in and Survey Production, 2008 transit mainly from the Democratic Republic of Congo. Material Official exports Survey production Large amounts of coltan and wolfram, on the other hand, (approximate) leave the country without being officially recorded as Gold 2,170 kilograms 500 kilograms Coltan 84 tons 150 tons exports. Prices offered directly at the mining site are also Wolfram 608 tons 1,000 tons well below those offered in the parallel market or by small Source: Midende 2009. exporting companies. 2.145 The contribution of mining to the government budget is negligible, and its contribution to the balance of payments is limited. Most minerals are channeled through comptoirs en transit, foreign companies that are exempt from paying both the VAT and the mining tax and do not have to repatriate foreign exchange earnings. Formalization and supervision of the sector as well as revision of the tax exemption could improve the contribution of mining to the budget and increase the income of miners and intermediaries. 2.146 If not addressed properly, artisanal and small-scale mining in Burundi can have numerous negative implications. Those implications include the following: • Use of basic techniques: Extraction techniques are basic, using primarily primitive tools. Artisanal exploitation of a mining site usually reaches only half of the resources and leaves the rest behind, leading to quickly accumulating losses. • Health, hygiene, and security problems: Mining sites are not secured, and accidents occur frequently. The sites are typically far from any infrastructure, such as health facilities and schools. Temporary housing set up near the site does not provide proper protection, sanitary conditions, and clean drinking water. In addition, the population mix of the makeshift communities and the presence of many demobilized soldiers lead to frequent conflicts. • Child labor: Children who work illegally at the mining sites are prone to be injured and to develop pulmonary infections and eye and skin diseases. Many are also malnourished and not able to attend school regularly. • Environmental impact: Mining that is not well controlled and organized can lead to deforestation, destruction of vegetation and soil, erosion, and water pollution resulting from the treatment of minerals. Abandoned sites are a real danger to people and animals.
Scenarios for Low, Medium, and High Contributions of Mining to the Economy of Burundi 2.147 What follows are three scenarios for the future of mining in Burundi. They are not intended to be accurate calculations, but rather to illustrate the wide spectrum of impacts in terms of government revenues. To maximize the chances that Burundi benefits from its mining sector, as illustrated by the medium- or high-contribution scenario, the country will have to implement significant policy reforms. 2.148 The low-contribution scenario assumes that no policy reforms would be undertaken other than the ongoing mining legislation and regulation update; and the market outlook does not permit Burundi’s potential to be fully realized. The following outcomes would result:
49 • Artisanal and small-scale mining would remain primarily informal, providing the government with erratic revenues of US$0.1–0.5 million, based on the royalty revenues coming from artisanal and small-scale mining in recent history. • The contribution of formal mining to GDP would probably remain around 1 percent. • The share of mineral ores in official exports would vary between 1 percent (2001) and 5 percent (2006). • However, the contribution of informal mining would be a lot more significant (with a stable number of 70,000 artisanal miners exploiting gold, coltan, and other minerals, including construction material); but it would offer very little benefit in terms of government revenues. 2.149 The medium-contribution scenario assumes that enough policy reforms are undertaken to formalize artisanal and small-scale mining, to attract investments, and to prepare the country with strengthened institutions for industrial mining to make a successful contribution to the economy. It also assumes that the market outlook is good enough to favor the launch of a medium-scale gold mine with circumstances corresponding to our low gold price projection (table 2.4). It would yield the following: • Artisanal and small-scale mining would progressively formalize and contribute up to a stabilized US$2 million per year to government revenues. • The gold mine could be committed around 2013, arrive at full production around 2015, and operate for five to seven years. • Government revenues would reach US$9 million in 2017–18, but could quickly decrease thereafter if no other mining operation is launched (figure 2.11). 2.150 Such a scenario would still need significant policy reforms to improve the geology data infrastructure, the capacity of institutions to monitor the operations, and the overall governance of the sector. 2.151 The high-contribution scenario assumes that major policy reforms are undertaken to formalize artisanal and small-scale mining, attract investments, prepare the country with strengthened institutions, and properly integrate good management of mining all along the value chain. It also assumes that the market outlook is good enough to favor the launch not only of a medium-scale gold mine, but also of a large-scale nickel mine operating for 25 years. Circumstances are supposed to correspond to the conservative nickel and gold price projections (table 2.4). Results would be as follows: • In addition to the impacts of the medium-contribution scenario, the nickel mine would make a major contribution to the economy and annual revenues could exceed US$50 million (figure 2.12). • Policy efforts to prepare for such an eventuality should be more consequent as they probably would have to adopt an integrated approach like EITI++. • Experiences of World Bank assistance in other countries suggest that the related public investment could range from US$8 million to US$10 million for a technical assistance project lasting four to five years. • More costly public investment in infrastructure related to nickel mining will be necessary as well (see section A in chapter 3).
50 Figure 2.11: Medium Contribution Scenario— Figure 2.12: High Contribution Scenario— Government Revenue Government Revenue
10 60 9 50 8 7 40 6 30 5 4 US$ millions 20 US$ millions 3 2 10 1 0 0
20102011201220132014201520162017201820192020 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2040 2041 2042 ASM Gold mine ASM Gold mine Nickel mine
Source: Authors’ calculations. Source: Authors’ calculations. Note: ASM = Artisanal and Small Scale Mining Note: ASM = Artisanal and Small Scale Mining
B5. Conclusion 2.152 Burundi has the potential to improve its growth performance considerably. Given the structure of the economy and the constraints it is facing, the low performance of the agriculture sector requires immediate attention. In the medium term, more broad-based growth and increased off-farm employment are necessary to sustainably reduce poverty and improve the population’s living conditions. Sectors that have been identified in this chapter as encouraging new sectors are primarily horticulture, mining, and fishing. 2.153 To support growth, Burundi’s policy reform and implementation efforts should focus on four main areas: (1) closing the infrastructure gap, (2) maximizing the benefits from the ongoing regional integration process, (3) improving the business environment in which the private sector operates, and (4) promoting encouraging new growth sectors. The government can further contribute to promoting growth by strengthening the country’s fiscal position through improved revenue mobilization, prudent spending policies, and reformed public financial management. These issues will be addressed in detail in the following chapters.
51 CHAPTER 3 THE GOVERNMENT’S OPTION TO TACKLE OBSTACLES AND REAP OPPORTUNITIES FOR ECONOMIC GROWTH
3.1 As described in chapter 1, Burundi’s economic recovery has so far been sluggish and below what can be expected for a post-conflict country. Only in recent years has economic growth improved slightly, but it has remained volatile. A number of reasons for the current situation and its origin were outlined in the chapter 2. The challenge now is to turn the economic performance around, consolidate the recent upward trend, and lay the foundation for high and sustainable growth in the future to significantly reduce poverty and improve the living conditions of the population. 3.2 How can the government help create an environment that is conducive to accelerated growth? Although there are many areas where improvements are needed, Burundi should focus on the following main objectives: (1) closing the infrastructure gap, (2) maximizing the benefits from the ongoing regional integration process, and (3) improving the environment in which the private sector operates, and (4) promoting new growth sources. Furthermore, the government needs to strengthen its fiscal position, which will be discussed in chapter 4.
A. AN INFRASTRUCTURE ACTION PLAN42 3.3 There is a broad consensus that infrastructure plays an essential role in growth (for example, see Ndulu et al. [2007], UNIDO [2009], or Foster and Briceño-Garmendia [2009]). Burundi has a large infrastructure deficit in almost all areas, including transport, electricity, and communications. For example, only 2 percent of the population has access to electricity, compared with 16 percent for Sub- Saharan Africa and 41 percent for other low-income developing countries. To make matters worse, the country is currently plagued by frequent power outages that hamper all areas of economic activity. 3.4 The poor state of the infrastructure also negatively affects Burundi’s trade. Domestic roads as well as transport corridors to the main ports in Dar es Salaam and Mombasa, on which Burundi depends, need to be upgraded and expanded to lower transport costs and improve access to regional and international markets. According to Naudé (2007), one of the main reasons for the poor performance of many African countries is the “proximity syndrome,” which he defines as the cumulative result of long distances to markets, landlocked locations, and suboptimal agglomeration patterns. Because of the cross- border nature of the required infrastructure projects, improved regional coordination and cooperation are essential to overcome the proximity gap. 3.5 The infrastructure action plan, developed by the African Development Bank and presented in the following section, lays out a detailed strategy for improving the infrastructure in Burundi, taking into account the importance of the regional context. It focuses on the power, transport, and communications sectors; and proposes actions to be taken, their timing, and a possible funding structure. If realized, the action plan will have a significant positive impact on the whole economy.
A1. Main Features of the Action Plan 3.6 The focus of the proposed infrastructure action plan is on power, transport, and communications. Discussions with the government of Burundi have confirmed the key objectives for the program.43 3.7 One of the objectives of the program for the power sector is to ensure, through increased investments in domestic and regional generation capacity, that the business community and households
42 This action plan has been developed by the African Development Bank; see AfDB (2009) for details of the proposed program. 43 Details of the proposed plan are given in appendix 5.
52 have access to reliable power supply 24 hours a day. Furthermore, the action plan proposes to establish a national transmission and distribution grid by 2015, with all 15 of the provincial capitals linked to this grid and supplied with electricity 24 hours a day at reasonable cost. With the grid in place, the electrification rate is supposed to increase from the current 2 percent of households to 25 percent by 2020 and to at least 40 percent by 2030. By 2020, 85 percent of urban households would have continuous access to the national distribution network; and by 2030, one third of all rural households would be linked to the grid. 3.8 The key objectives of the proposed action plan for the transport sector are to lower the costs of transport for the entire economy and to improve access to local and international markets. The proposed program focuses on road transport and civil aviation, and provides for further investigation of the possibility of a rail extension from Tanzania into Burundi. The main component of the transport program is the upgrading and expansion of the country’s road network over the next decade. The international airport in Bujumbura would be expanded and modernized. The options for extending the Tanzanian rail network into Burundi have also been reviewed as part of this study. 3.9 The proposed communications program aims to improve access to the international communications network substantially and to lay the foundations for a national communications grid that will provide communities and business through Burundi with low-cost voice and data communications. A high priority is attached to the immediate development of a national communications grid of fiber-optic cable and digital microwave that would be linked to the regional network. 3.10 The three main sources of funding for the program are the government of Burundi, including the electricity utility and the airport authority; the donor community; and the private sector. The government and donor community each fund major portions of the power program, but the strategy is to look to the private sector to own and operate the Figure 3.1: Funding Arrangements for the Core Infrastructure Program new domestic hydroelectric stations that are proposed. In the case of the roads 3,000 program, the donor community would Government 2,500 Donors fund about 80 percent of the program. 10% Private The civil aviation program would be 2,000 financed by the private sector, except for 23% a small amount of donor and 1,500 government funding for the human and 37% 74% institutional capacity building that will 1,000 be required. For the program as a whole, 500 the government would fund about 27 40% US$ millions (2007 constant prices) 16% percent, the donor community about 56 0 percent, and the private sector about 17 Power Transport Communications percent (figure 3.1). Funding for investments for the nickel mine project Source: AfDB 2009. would be provided, to a large extent, by the mining company involved in the project.
A2. Key Policy Objectives 3.11 A concerted program of action will be needed over the next two decades to overcome the country’s current serious infrastructure deficiencies in power, transport, and communications. The proposed infrastructure action plan is built around three key policy objectives: 1. Stepping up spending on infrastructure will ensure improved access to services while reducing the cost and improving the reliability of these services, thereby removing some of the main obstacles to sustained strong economic growth. The core infrastructure action program to be
53 implemented over the next two decades would cost US$4.6 billion (at 2007 constant prices). In addition, US$1.2 billion would be spent on maintenance of assets in these three sectors. 2. The US$5.8 billion program will create substantial opportunities for the development of domestic business activities. To generate a strong domestic supply response to this ambitious program and ensure sustained economic growth of 6–7 percent a year, the government will need to design and implement comprehensive programs for development of small and medium-size businesses and for skills development in the labor market. Both of these programs are essential complements to the infrastructure action program. 3. Implementation of the proposed action plan will require increased attention to coordination of the power, transport, and communications programs within the government. It will also require close cooperation between the government and the donor community in designing and implementing these programs. Emphasis on coordination on these two fronts can improve the alignment of investments with national and regional priorities and the overall efficiency of the development process in Burundi. It is important to emphasize that, in fact, an expenditure of some US$5.8 billion over the next two decades is required to meet the current national objectives for infrastructure development. These programs will also lay the foundation for a major investment in nickel mining in the decade ahead—an endeavor that can bring substantial additional benefits to the country.
A3. Economic Impact of the Action Plan 3.12 A number of scenarios are considered to assess the economic impact of implementing the action plan.44 In addition to the base case scenario, five alternative scenarios are considered (table A.10 in appendix 4). In all of these scenarios, it is assumed that internal security in Burundi continues to improve and that there is social and political stability; that the government continues to adhere to sound macroeconomic policies; and that Burundi, with the help of the international donor community, continues to make steady progress in reducing its vulnerability to debt distress. It is also assumed that action is taken to improve the business and investment climate and that there is a strong domestic supply response to the proposed infrastructure action plan.
Economic Impact in the Near Term 3.13 For the short and medium terms, the recent encouraging recovery in economic growth can be maintained. Even if no further domestic policy initiatives are taken, there will be further sharp increases in public investment expenditures until about 2011, because of the ongoing major donor-funded program of infrastructure rehabilitation. However, these projects come to a close by about 2014; and as a result, spending under these programs declines sharply from 2012 onward. Analysis of the impact of the ongoing infrastructure rehabilitation program suggests that GDP growth may be in the range of 4 percent in real terms in 2009, and that it would increase to about 5 percent a year in 2010 and 2011. As the ongoing infrastructure program phases down after 2011, economic growth could decline to about 4 percent a year in 2013–14. Serious delays in implementing the ongoing donor-funded program would result in slower economic growth in the immediate future. 3.14 The key point that emerges from this assessment of the economic outlook for the near term is that a strong push to improve the basic infrastructure services in Burundi can sustain economic growth in the near term and lay the foundations for a period of strong economic growth over the next two decades.
44 These scenarios have been developed by the African Development Bank, as document in AfDB (2009).
54 Impact of the Base Case in the Longer Term 3.15 The long-term economic benefits that accrue to Burundi under the base case scenario are substantial. The base case scenario includes the core infrastructure program and the nickel mining project. Benefits include the following: • Sustained growth of the domestic economy that creates new business opportunities and increases incomes—GDP grows in real terms by about 7.2 percent a year over the next two decades in this scenario. GDP per capita increases by 4.5 percent a year, reaching roughly US$324 by 2030 (at 2007 constant prices). Sustained growth of incomes in this range begins to have a significant impact on the incidence of poverty in the country, with a substantial number of people at or just below the poverty line moving out of “official” poverty (although many of these people would still be vulnerable to downturns in the economy resulting from droughts or other disruptions). • Increased opportunities for productive employment—Employment in the nonfarm sector grows at 6 percent a year. Over the next two decades, about 1.3 million jobs are created in the nonfarm sector, mainly in urban areas—equivalent to almost half of the 2.7 million new entrants into the labor force in that period. The share of employment in agriculture declines steadily to about 70 percent of the workforce by 2030. By 2030, the industrial sector, including mining, is projected to account for almost 10 percent of employment; and the services sector is projected to account for about 20 percent. • Improved access to infrastructure services and lower costs for them—The sustained strong growth in the economy and employment stems from the major investment in basic infrastructure proposed under the action plan. More reliable power supplies, improved Table 3.1: Selected Indicators for the Base Case Scenario transport and communications Indicator 2010 2015 2020 services, and lower costs for these Total electricity supply (gigawatt hours) 196 471 1,981 services enhance the business Export volume (thousands of metric tons) 37 50 124 Investment (% of GDP) 31.8 85.3 16.7 environment and investment Public 24.1 22.9 10.7 opportunities for the private sector and Private 7.7 62.4 6.0 raise Burundi’s international Source: AfDB 2009. competitiveness. (For selected indicators, see table 3.1.) • Increased tax revenues and expanded public services—The combination of strong economic growth and the start-up of the mining operation have significant implications for the revenue position of the government.
Alternative Scenarios 3.16 Given that funding for all components is not guaranteed and that private participation in particular cannot be enforced, a number of alternative scenarios have been developed. Table 3.2 summarizes the outcomes for each of the five alternative scenarios considered and compares them with the base case. Several key points emerge from this analysis: • In the event that the nickel mining project does not proceed, but the core infrastructure program is implemented in full (scenario B), GDP grows at about 6.4 percent a year over the next two decades—sufficient to create a substantial amount of additional productive employment, improve incomes and productivity in urban and rural areas, and contribute to a significant reduction in the incidence of poverty in the country. The main economic impact is a substantial drop in export income and government revenues. • If private investment is not available for the power and aviation sectors (scenario C), there is a further modest decline in growth performance, with GDP increasing by an average of 6.2 percent
55 a year over the next two decades. The decline in the growth rate is limited because it is assumed that all the additional power that is needed to meet demand is imported from the East Africa Power Pool grid. The implication is that by 2030, imported power accounts for 90 percent of total supply in Burundi. • In scenarios D and E, the level of public investment in the proposed infrastructure program declines from US$3.8 billion over the next two decades to less than US$1 billion. The growth performance of the economy declines sharply. In scenario E, the GDP growth rate is more than two full percentage points below that of the base case. The economy has difficulty in absorbing new labor force entrants into productive employment opportunities. In addition, much larger numbers of people remain in low-productivity agricultural pursuits. Although there may be some reduction in the incidence of poverty in the country in this scenario, the total number of people in absolute poverty would increase substantially over the next two decades. • Scenario F assumes that one of the rail extensions into Burundi goes ahead, but that the nickel mining operation is based on refining ore at the mine site. In the absence of the 4 million tons of ore exports each year, the economic impact of the rail extension is modest and GDP growth is not much higher than in the base case. Moreover, in this scenario there is a risk that large public subsidies would be needed for the operation of the rail network in Burundi.
Table 3.2: Key Outcomes for the Base Case and Alternative Scenarios (GDP at 2007 constant prices)
Scenarios A B C D E F
Indicator Implement Private Only 50% of Only 20% of Implement Base core funding for public funding public funding base case case infrastructure action plan available for available for and rail program not available action plan action plan extension
GDP in 2030 (US$ millions) 4,560 3,895 3,745 3,313 2,868 4,721 GDP growth rate (annual % for 2010–30) 7.2 6.4 6.2 5.6 4.8 7.4 GDP per capita in 2030 (US$) 324 277 266 236 204 336 Source: AfDB 2009.
A4. Key Policy Issues for the Government 3.17 The following important policy issues for consideration by the government and the donor community have been identified: • Early action on the core infrastructure program—It is important for Burundi to maintain the current growth momentum generated by the ongoing donor support for infrastructure rehabilitation. An early launch of the proposed infrastructure action plan requires a series of decisions by the government and donor community regarding the design of particular programs and projects and the funding arrangements for these activities. There is a degree of urgency associated with some of these decisions because implementing a number of key components of the program must begin in 2010 if the above-mentioned targets for the power and transport sectors are to be realized and if the current growth momentum in the economy is to be sustained. • Private funding mobilized for infrastructure—The proposed action plan calls for increased use of public-private partnership (PPP) arrangements to mobilize private sector funding for the program. In general, accessing private funding under these types of arrangements enables governments to avoid or defer public spending on infrastructure without forgoing its benefits. To manage the risks associated with the proposed PPPs and to ensure that they provide high-quality infrastructure services in an efficient manner, the government will need to give careful attention to the following three broad sets of concerns: (1) the legal framework governing the PPPs; (2) the processes for selecting and implementing PPPs, including the roles played by relevant
56 government agencies; and (3) the contractual obligations on which the PPPs are based that directly determine the fiscal risk incurred by the government. • Viability of a rail extension into Burundi—An important point that emerges is a series of questions about the extension of the Tanzanian rail network into Burundi. If the rail is unable to carry freight generated by the nickel mining operation in Burundi, then the volumes of freight available from regular commerce appear to be too small to justify the investment at any time within the next two decades. If refining metal at the mine site is the most attractive option available to potential investors, there may be little appetite for the alternative (perhaps higher- cost) option of using a rail system in Burundi for the freight services for the mine. More analysis is needed on these options before final decisions can be made by the government. In any event, there will be a need for close cooperation and coordination with Tanzanian authorities if the mine is to go ahead, because the rail system would be used to transport metal from Kigoma to the port at Dar es Salaam. • Degree of dependence on imported power—This analysis points to the reemergence of a power supply deficit by about 2024. The key policy question for the longer term is whether to develop other potential domestic sites to keep dependence on imported power at prudent levels, or to allow increased dependence on imported supplies from the East Africa Power Pool network. A related question concerns the likely cost of new sources of domestic supply, compared with imports from a low-cost producer such as Ethiopia. 3.18 A 20-year program of this magnitude inevitably faces risks and uncertainties, large and small, foreseen and unforeseen. Many possibilities can be considered, including, for example, major political risks such as deterioration in internal security, or civil disturbances in neighboring countries that affect the overall performance of the Burundi economy and its attractiveness as a destination for private investment. There are also risks that stem from the international environment, including sharply higher petroleum or raw material prices that may adversely affect the attractiveness of investing in Burundi. 3.19 The risks and uncertainties of most interest at this stage relate to the design, funding, and implementation of the proposed program. To manage these risks, the government and donor community will need to strengthen coordination mechanisms for the infrastructure sectors, starting with early completion and adoption of the proposed master plans. Regular meetings with donors may then be required to monitor progress in program implementation. Risks include shortfalls in available funding, delays in implementation, and macroeconomic difficulties arising from the high investment levels. A further key risk is the inadequate supply response. Aid-financed projects usually create a strong demand for imports because the domestic economy is not able to supply the required goods. Action should therefore be taken to ensure an adequate supply response to the infrastructure plan.
A5. Priority Interventions in the Short and Medium Terms 3.20 Based on the detailed analysis conducted in the context of this study, the following priority recommendations concerning the power, transport, and communications sectors could be identified for the short and medium terms (2010–15). 3.21 Improving access to electric power: • Improve demand management through ° installation of prepaid meters; ° distribution and promotion of compact fluorescent lights; ° utility energy audit; and ° promotion of energy efficiency investments by large consumers.
57 • Accelerate electricity supply through ° development of Burundi’s domestic hydroelectric potential, commissioning Kaganuzi (5 megawatts), Mpanda (10.4 megawatts), and Kabu 16 (20 megawatts) plants; ° development of a small-size run-of-the-river hydropower plant; ° participation in regional hydroelectric projects. • Expand the transmission grid through ° upgrading and expansion of regional network to 220-kilovolt network to establish regional links; ° provide 110-kilovolt lines with necessary substations to all provincial capitals; ° develop system protection and control functions for the network. • Strengthen financial capacity of REGIDESO through ° financial restructuring of REGIDESO, including a revision of the tariff structure; ° separate REGIDESO into water and electricity companies. 3.22 Improving transport infrastructure: • Develop and maintain road infrastructure through ° rehabilitation of the road network, with selection of roads according to economic, social, and connectivity aspects; ° development of a program for regulating and managing road transport; ° capacity building and technical studies (such as transport industry surveys, passenger and freight information, and assessment of road conditions); and ° a clear strategy for maintenance, addressing maintenance works, funding of the required program, and involvement of the private sector and the community. • prepare a civil aviation business plan that addresses ° the role of the civil aviation authority (Régie des Services Aéronautiques- RSA), with particular attention to measures needed for compliance with the International Civil Aviation Organization (ICAO) and Civil Aviation Safety and Security Oversight Agency (CASSOA)45; ° training and other capacity-building programs to meet these standards; and ° the outline of a commercial plan for the development of Bujumbura airport. 3.23 For accelerating the development of communications, in addition to already ongoing projects financed mainly by the World Bank, the following measures are recommended: • extension of ongoing capacity building to support the regulatory and policy environment; and • additional support for the development of new applications throughout the country, including the development and/or expansion of e-education, e-health, and e-commerce services for communities, schools, hospitals, and business entities.
45 This agency was founded by the East African Community (EAC) member-states Kenya, Tanzania, and Uganda in April 2007.
58 B. MAXIMIZING THE BENEFITS OF BURUNDI’S EAC MEMBERSHIP 3.24 Regional integration is another critical component of Burundi’s growth strategy. Being landlocked, and with long distances to the nearest seaport, the government recognizes that increased economic integration is key to achieving greater competitiveness in regional and global markets and in laying the foundations for an extended period of strong growth. In addition to easing the current high transport costs and improving access to goods and services, regional integration will also provide a larger market for Burundi’s exports and help attract foreign investment. True regional integration would help Burundi achieve returns to scale, especially with increased labor mobility. The EAC agreement can also create a push for economic and structural reforms. In addition, negotiating trade agreements as a group may give more incentive for other countries and regions to enter into a free trade agreement with Burundi. For example, the EAC is negotiating an Economic Partnership Agreement (EPA) with the European Union (EU). A final gain from the EAC may come from its power serving as a stabilizing factor in the region. 3.25 Like many other developing countries, Burundi has been a member of several regional agreements in the past, with an overall unsatisfactory experience.46 There are various reasons for the poor results, including missing prerequisites, such as stability and political will, and the lack of common interest. The objective of this section is therefore to provide a coherent strategic framework to guide Burundi’s actions toward a successful membership in the EAC, building on lessons learned. Gains and risks of Burundi’s accession to the EAC are discussed, and policy recommendations are developed to help Burundi maximize the benefits of the already ongoing regional integration process. 3.26 One of the biggest challenges of the integration process will be in developing and implementing a sustainable regional integration strategy and its road map. Because the moratorium (requested by Burundian authorities in order to be prepared) ended in July 2009, it is critical to set up such a strategy to ensure Burundi’s successful participation in the EAC. The strategy should focus on the priorities and maximize the contribution of development partners and other actors. Suggestions include the following: (1) prepare an action plan, including sequencing and prioritization of actions; and (2) create or reinvigorate communication and coordination mechanisms. 3.27 There are a number of potential benefits and costs to joining a regional integration process. A summary of those benefits and costs and of guiding principles for an effective integration process is given in box A.1 in appendix 3. A number of actions, however, are necessary to maximize the benefits from the regional integration process. In the concluding section, recommendations for policy action are proposed as well as an action plan matrix suggesting the way forward. 3.28 The following section offers recommendations to help Burundi maximize the benefits of its accession to the EAC. It proposes a road map and highlights critical actions that need to be taken immediately and in the medium term to move forward with the regional integration agenda. Areas of focus include institutional reforms, capacity building, increased involvement and facilitation of the private sector, and a clear coordination and communication mechanism.
B1. Brief History of the EAC: Objectives, Achievements, and Challenges 3.29 The EAC targets a deeper form of regional integration. Two decades after the collapse of the previous East African Cooperation, the Treaty establishing the EAC was signed in November 1999 by Kenya, Tanzania, and Uganda. The EAC intends to go beyond trade and economic arrangements, commonly known as a market-oriented integration process; and to implement a strategy aimed at
46 Since its independence in 1962, Burundi has engaged in various regional agreements and initiatives, including: the Economic Community of the Countries of the Great Lakes (CEPGL) (1976), the Kagera Basin Development Organization (KBO) (1979), the Economic Community of Central African States (ECCAS) (1983), the Cross-Border Initiative (CBI) (1992), and the Preferential Trade Area (PTA) (1984) that became the Common Market for Eastern and Southern Africa (COMESA) (1993). A synthesis of the performances of the different agreements is presented in table A.5 in appendix 3.
59 developing and improving regional infrastructure, managing regional commons, and producing regional public goods, thereby moving toward a deeper form of regional integration. 3.30 The characteristics of the EAC member-states vary considerably. Since July 1, 2007, the EAC has comprised five member-states. Burundi and Rwanda were officially accepted as new members of the EAC in November 2006, and were effectively admitted by July 1, 2007. The area of the EAC has a total size of 1,702,000 square kilometers, has a GDP of about US$75 billion,47 and represents a population of 131 million. There is considerable variation in the economic performance of the member- states (table 3.3). Kenya is far ahead, with a GDP about 20 times as high as that of Burundi, which has the smallest economy in terms of GDP.
Table 3.3: Selected Socioeconomic Indicators for EAC Members, 2008
Indicator Burundi Kenya Rwanda Tanzania Uganda Total GDP (current US$ millions) 1,163 23,507 4,457 20,490 14,529 75,146 GDP per capita (2000 constant US$) 111 464 313 362 348 370a Real GDP growth rate (2002–08, annual %) 3.2 4.5 7.2 6.9 7.8 6.1 a Population (millions) 8.1 38.5 9.7 42.5 31.7 130.5 Land size (thousands of sq. km) 25.7 569.1 24.7 885.8 197.1 1,702 Population density (population per sq. km) 314 67 394 48 161 n.a. Life expectancy (years) 51 54 50 56 53 54 a Source: World Development Indicators. Note: n.a. = not applicable. a Weighted average.
3.31 The ultimate objective is to establish a political federation built on a single market with free circulation of goods, services, and factors of production and harmonized economic policies and regulatory arrangements. So far, the EAC has progressed relatively quickly. It has adopted an approach of parallel implementation of programs regarding all steps of the market integration process. The integration process of the EAC benefited notably from the COMESA free trade agreement48 and the membership of most countries in the Cross-Border Initiative for regional integration.49 Substantial progress has been made in a relatively short period of time and with limited institutional capacity at the regional and country levels. This is a signal of strong political commitment to building an institutionally coherent and economically integrated core group, which could inspire the rest of Africa. 3.32 Encouraged by the good performance so far, negotiations started for creating a free trade area and a customs union among Kenya, Tanzania, and Uganda. In contrast to a number of other initiatives in Africa over the past three decades, the EAC has moved quickly to eliminate a large proportion of tariffs on intra-EAC trade as an integral part of establishing a free trade area among the founding members. The common external tariff (CET) was adopted Table 3.4: External Tariffs in the EAC, 2000–08 on July 1, 2009, and is fully Country 2000 2001 2002 2003 2004 2005 2006 2007 2008 effective. (The development of the Burundi — — 25.3 — — 22.5 15.7 13.4 — CET for the EAC member- Kenya 20.5 21.9 — — 18.5 15.0 15.0 15.3 15.7 Rwanda — 11.7 — 11.0 — 18.5 15.9 — 16.8 countries is given in table 3.4.) Tanzania 18.2 — — 15.2 — 15.1 14.5 14.3 16.4 The schedule of accession allows Uganda 10.0 9.8 9.3 9.7 9.3 9.0 13.8 13.3 16.1 commensurate action on tariffs by Source: United Nations Trade Analysis and Information System. Burundi and Rwanda to be Note: — = not available.
47 That figure is based on 2008 current prices, according to World Development Indicators. 48 Kenya has been a member of the COMESA free trade agreement since its beginning in October 2000, Burundi and Rwanda became members in January 2004, and Uganda joined in 2005. However, a tariff of 20 percent was still in place for Uganda relative to the COMESA countries. Tanzania withdrew from COMESA a few years ago to focus on its membership in the South African Development Community (SADC). 49 This initiative was launched in early 1992, mainly with the support of the World Bank, and aimed at accelerating the liberalization of economies in the subregion to strengthen regional integration in eastern and southern Africa.
60 undertaken until 2010. Equally, negotiations started in 2006 to establish a common market by 2010 (including allowing for free movement of labor by 2009) and a monetary union by 2012. Furthermore, efforts are under way to fast-track the establishment of a political federation, the final stage of the integration process, by 2015. 3.33 The strategic agenda of the EAC, however, is very ambitious and might need to be reconsidered to avoid undermining both the efficiency and the credibility of the process. The development strategy of the EAC has many quite ambitious objectives and policy actions covering a very wide range of issues, without a good sense of priority (World Bank 2008d). Lack of prioritization means spreading EAC’s scarce human, institutional, and financial resources extremely thinly over too many activities. As a result, such a broad and overloaded agenda frequently leads to slow progress, missed deadlines, and a lowering of credibility of the EAC secretariat—while it may also be interpreted as weak political commitment by the member-states. 3.34 The challenge of maintaining the momentum created in the past few years will require prioritization of objectives and sequencing of planned actions, including securing adequate resources. To achieve a deeper integration, the EAC secretariat could focus on seeking and setting up agreed policy priorities. Given the previous progress and the necessity of keeping the common interest in mind, suggestions for immediate intervention could focus on deepening the trade integration process and dealing with regional public goods in infrastructure. This includes, in particular, (1) finalizing free trade area issues within the group, (2) consolidating the customs union, (3) developing a common trade policy, (4) streamlining overlapping commitments, and (5) establishing and starting the implementation of a regional infrastructure road map. Definitely, a realistic and gradual agenda could make maximum impact and build confidence in the EAC secretariat and the member-states.
B2. Burundi’s EAC Membership: Benefits and Risks 3.35 There are several potential benefits and risks of regional integration. The following section outlines the challenges of Burundi’s accession to the EAC, the possible noneconomic gains (especially those related to the reform agenda), and the trade gains that could result from EAC membership. It further explores how the industrial and services sectors could benefit from foreign direct investment (FDI), the role of regional infrastructure, and the possibility of Burundi becoming a regional financial and transport hub.
Burundi’s Accession to the EAC: Challenges and Strategy 3.36 Burundi authorities have shown strong commitment to the EAC membership. Since Burundi was officially admitted as a member of the EAC on July 1, 2007, strong efforts have been made by the authorities to align with the EAC requirements, especially in political, institutional, and economic areas. The political will resulted in the signing of the treaty, the ratification of protocols and conventions, and the quest for development partners’ support alongside the progressive stabilization of the country and (although late) a communication campaign aimed at sensitizing the population to support the integration initiative. 3.37 An institutional structure has been put in place to further the integration agenda. In November 2007, a new government included (for the first time) a ministry in charge of regional integration and EAC affairs, including a vice-ministry exclusively in charge of EAC affairs. In early January 2009, these institutions were restructured into a full ministry in charge of EAC affairs, as practiced within the other EAC member-states. This constituted a strong signal of political commitment to enable the dedicated ministry to play its role fully. 3.38 Burundi, however, still faces institutional weaknesses regarding the challenges ahead (both internal and external). First of all, although the working language in the EAC is English, most Burundians speak only French. This language misalignment has severe consequences in terms of common
61 understanding and follow-up on agreed decisions. Beyond that barrier, the government experienced difficulties in paying member contributions and allowing staff to fully attend the frequent regional meetings. This fact has raised questions about Burundi’s capacity to engage in a regional program, and it could undermine its credibility. Furthermore, compared with the other member-states, current staffs’ capacities, skills, and experience are limited and do not always meet the technical requirements of the regional integration process. 3.39 Moreover, Burundi cannot move ahead without addressing the structural constraints facing its economy. Although Burundi remains the smallest economy within the region (table 3.3), a process of catching up will have to deal with several constraints, including among others (1) poor physical and energy infrastructure (including transportation system); (2) lack of dynamism and competitiveness of the domestic private sector as well as a weak financial system (including the payment system); (3) a nonconducive business environment (including the lack of an encouraging statutory framework for a favorable investment climate); and (4) more generally, poor governance and the absence of a holistic approach to benefit from the regional economic opportunities. 3.40 The biggest challenges will be prioritizing and sequencing the next steps within a sustainable regional integration strategy and its road map. Given the delay in implementing the integration process, it is critical to set up a comprehensive strategy to ensure Burundi’s successful participation. The necessary steps would be described in a coherent road map for the accession, including a sequenced action plan reflecting immediate priorities. Such a strategy would enable the government to plan accordingly, move on quickly with the most urgent measures, and maximize the contribution of development partners who can focus on specific areas of the road map.
The Need to Push the Reform Agenda 3.41 In addition to offering economic gains (see below), a regional integration process also provides noneconomic gains through regional commitment mechanisms. These gains can be grouped into two categories: (1) those addressing political and security issues and (2) those pushing economic and structural reforms. Regarding the first category, for example, Burundi could benefit considerably from a “regional” political stabilization effect. Furthermore, regional integration mechanisms tend to strengthen an individual country’s bargaining power as well as the credibility of its policies (Baldwin, François, and Portes 1997). In addition, they tend to seriously lower security risks through regional agreements and other common arrangements. Moreover, there is a hope that the land issues that are a permanent source of conflict in Burundi could be addressed at least partially through the establishment of the EAC common market, expected in 2010. 3.42 The potential noneconomic gain the Burundian strategy should focus on is advancing the ongoing reform agenda. Because the accession is subject to an alignment on some regional prerequisites, the regional mechanism is an excellent tool to advance the reform agenda. This will be particularly true in the following five main areas: 1. Implementation and harmonization of fiscal instruments—As far as Burundi’s accession to the customs union is concerned, there are numerous measures that need to be implemented without delay. They include the harmonizing the legal framework, including the fiscal and customs codes; computerizing the tax department; introducing the value-added tax in lieu of the sales tax by July 1, 2009; adopting the common external tariff in 2009, including the negative list (and eventually the timeline for alignment); aligning with the EAC budget calendar in 2010; and establishing the Burundi Revenue Authority, which is expected to be in place by January 2010 and fully effective by March 2010. Many these fiscal reforms are under way or have been completed. 2. Development of an integrated financial market—The Central Bank of Burundi needs to accelerate the financial sector reform to comply with EAC standards and norms. A recent study showed that information on current volumes of cross-border trade in financial products is generally sketchy
62 and incomplete, making further integration problematic by obscuring the extent of cross-border links.50 Specific actions would include promoting market access for investors from the EAC, setting up a specific unit responsible for systems and methods of payment, and defining a strategy and policy modernizing the payment systems and necessary infrastructure in line with the plans of the EAC (EAC 2006). This can be done in cooperation with the Central Bank of Rwanda, which already has started such a reform. 3. Reduction of tariff and nontariff barriers (NTBs) and development of a common trade policy— Burundi’s substantial progress in dismantling its tariff barriers and accessing the COMESA free trade area since January 2004 will be instrumental for its accession to the EAC free trade area. However, as in the other EAC member-states, NTBs in place in Burundi remain a serious concern for intraregional trade—especially customs and administrative procedures, including the length of the clearance formalities and the multiplicity of institutions of control at the port in Bujumbura. Proposed measures to remove Burundi’s NTBs are described in box 3.1.
Box 3.1: Proposed Measures to Remove Burundi NTBs within the EAC Establishing a mechanism for identifying and monitoring the disposal of NTBs. Preferably, this mechanism would include the public sector and private sector representatives concerned with issues of trade facilitation. Consolidating institutions in charge of customs clearance and administrative control at the port of Bujumbura in a one-stop window to streamline the transit and customs formalities. Improving layout and equipment of border posts. Introducing competition in the selection of companies for preshipment inspection. Gradually establishing an infrastructure for quality control and certification. Source: Adapted from World Bank (2008d).
4. Streamlining of overlapping agreements—Burundi is part of multiple arrangements that may negatively impact the efficiency of its programs. In addition to the fiscal cost of being a member of several organizations, multiple memberships imply multiple administrative procedures; and that poses problems for a country with limited human, institutional, and financial capacity. As mentioned above, as a member of the EAC, Burundi is negotiating an EPA with the EU. In September 2009, a road map for the finalization and eventual signing of the Framework EPA was approved by the respective council of the EAC.51 5. Promotion of an environment conducive to business—Regarding the business environment, Burundi will need to underpin recent considerable progress. Although Burundi recently drafted several business instruments,52 thereby taking into account the issue of harmonization with the regional existing tools, the duty exemption, the tax incentives to export procedures, customs regulations, and transit cargo regulations have not yet been harmonized on the EAC level; and limited progress has been made in harmonizing standards. In the meantime, it will be instrumental for Burundi to push forward in terms of implementation to allow regional companies and investors to operate efficiently and stimulate investment and growth in the near future.
50 World Bank Financial Integration within EAC: Mission Report, May 2009. 51 Currently, however, the lack of consensus between the EAC and the EU on the sequencing of meetings is the main roadblock to finalizing negotiations. 52 These instruments include revision of the investment code and a presidential decree on the setting up of an investment promotion agency; revision of the commercial and public-private company codes; drafting of a competition law; and drafting of a presidential decree on the public-private dialogue framework.
63 Trade Gains from Burundi’s EAC Membership 3.42 What are the potential gains from trade owing to Burundi’s accession to the EAC? To answer this question, necessary to examine how Burundi’s trade structure has changed over time and to consider the importance of the regional trade agreements to which it belongs. It is not surprising that Burundi is likely to reap much larger traditional gains from trade from the Economic Partnership Agreement with the EU than through the EAC. Burundi’s export structure is very complementary with the EU import structure, and vice versa. However, there are a few interesting features of the EAC. In particular, trade with the EAC has enabled Burundi to move into differentiated goods, which it does not export to Europe. In addition, the EAC has been a very important market for expanding into new goods. This offers a new channel through which regional agreements between similar countries can be beneficial in Africa.
Are Burundi and the Other EAC Members Natural Trade Partners? 3.43 Regional trade agreements are more likely to be welfare improving if they are among countries that are major trade partners prior to the agreement (see Summers [1991] and Krugman [1991]).53 The logic is that countries that trade disproportionately more with each other are less likely to cause trade diversion. In an extreme case—for example, if all of Burundi’s trade were with EAC countries—the agreement would be identical to free trade from a Burundian perspective. Regional trade agreements among natural trading partners are also more likely to facilitate deeper integration, which further expands welfare gains. Large trade flows provide bigger incentives for deep integration, and similar tastes will facilitate institutional harmonization. In addition, proximity is required for integration of many transport-related systems, such as railroad or trucking. 3.44 The share (measured in value) of Burundi’s exports that go to members of the EAC has increased since 2000, but has remained below 15 percent in recent years. This is very low, compared with the Southern Cone Common Market (Mercosur) and the North American Free Trade Agreement (NAFTA) in which the share is more than 25 percent and the EU in which it is above 50 percent. In terms of imports, the share from EAC members has also increased and is now around 20 percent. 3.45 Export similarity and complementarity indexes can be used to analyze Burundi’s trade pattern. To understand how similar or complementary trade is between Burundi and other regions or countries, export similarity and complementarity indexes are calculated between Burundi and other countries in the EAC, other regions such as COMESA (non-EAC), and the EU. In a first step, the Finger and Kreinin (1979) export similarity indexes (which describe how similar two trade structures are) are calculated.54 If the product share distributions of the exports of two countries are identical, the index will be 100; if the export patterns are completely different, the index will be 0. Another way to gauge the importance of complementarity versus similarity in trade is to examine whether Burundi’s main export products are imported on net by its trade partners. The complementarity index is the total share of Burundi’s exports that a trade partner imports more than it exports. For example, Burundi’s biggest export is coffee; therefore, net importers of coffee will be natural trading partners.55 A number close to 0 implies that the country is a net exporter of the products that Burundi exports.
53 Krishna (2003) finds little empirical support for the natural trade partners hypothesis. However, Krishna’s results are highly dependent on the model he is estimating. In addition, he examines only 24 bilateral pairs, and it is possible that this is too limited a sample to find evidence. 54 The index of export similarity is defined by ͍̿̓ Ɣ ʜ?$ ͇͢͝ʞ͍$Q͍$ʟʝ SRR, where ESI is the similarity between exports of country A and country B. ͍$%is the share of product i in j’s exports to the world. 55 First, we create an indicator that is 1 if a country is a net importer (imports – exports > 0) of a given product. Next, we calculate the share of Burundi’s exports that is on net imported by other countries and regions as ͇̓͊̿͒͊ Ɣ ʜ?$ ͍$0-0) $ ̻̓͝·SRR, where ͍̼ͩͦͩ͘͢͝͝ is Burundi’s share of exports in industry i, and ̻̓͝ is an indicator for whether country or region A is a net importer of that product.
64 Table 3.5: Similarity and Complementarity Indexes for Burundi, 1995–2007 (Averages)
Index Export Similarity Export Import Similarity Import Export Similarity Kenya 18.3 11.2 36.8 Rwanda 47.1 10.8 13.9 Tanzania 16.8 12.0 26.1 Uganda 50.2 12.3 20.7 EAC 26.5 11.8 31.4 Other COMESA countries 4.7 15.0 41.1 EU 4.2 91.0 68.9 World 4.2 n.a. n.a. Source: Authors’ calculations based on COMTRADE HS 1992 6 digit data. Note: n.a. = not applicable.
3.46 The following findings can be derived from an analysis of the export similarity indexes (table 3.5, “Export Similarity” column): Within the EAC, exports are most similar to Uganda, where about half of exports, on average, are in the same products. • Overall exports are more similar with countries in the EAC than with the other COMESA countries, with the EU, or with the world. • Burundi and other EAC countries tend to be competing exporters. • Although trade diversion remains a concern, it does not seem likely to have a large impact on Burundi, and all positive effects of regional integration mentioned above will outweigh any negative effects. 3.47 The complementarity index (or export-import similarity index) shows clearly that the EU’s trade patterns are very complementary with Burundi (table 3.5). The EU is a net importer of 90 percent of Burundi’s exports. In contrast, EAC countries and other partners in COMESA are net importers of only as much as 12 percent and 15 percent of Burundi’s exports, respectively. 3.48 Finally, the results of an analysis of potential import/consumer gains suggest that Burundi is far more complementary with the EU than with EAC countries. Burundi is treated as the importer, and the share of the region’s exports of which Burundi is a net importer is examined. In other words, this index looks at whether there are consumer gains from the various agreements. Again, results show that there is a lot more complementarity between Burundi and the EU than with the EAC (table 3.5). Burundi is a net importer of roughly 70 percent of the products that the EU exports, whereas it is a net importer of more than 30 percent of EAC exports. This suggests that in the absence of differentiation and/or diversification of the product base, the trend is not likely to change and the country could not expect to maximize the gains resulting from its commercial policy.
Are There Gains from Trade in Differentiated Goods? 3.49 Although export similarity indexes and export-import complementarity are good ways to model traditional trade, these relationships break down with differentiated goods. In theory, trade in differentiated goods is still likely to be abundant, even when both partners export the same goods. Differentiated goods are goods for which consumers appreciate variety. To measure the extent of differentiation, data from Rauch (1999) is used. Rauch splits all traded goods into three categories: (1) those traded on an organized exchange, (2) reference priced goods, and (3) differentiated goods. The first two categories are meant to capture goods that are roughly homogeneous; the third category covers goods for which brand and producer are relatively more important.56
56 About 10 percent of goods were not classified and therefore were excluded from the analysis.
65 3.50 Whereas the extent of trade in differentiated goods highlights an important avenue for growth with the EAC, trade in differentiated goods is quite volatile; and it is too early to determine how the EAC membership will affect it. Overall, differentiated goods account for less than 10 percent of Burundi’s exports, though their share has grown in recent years. In contrast, Burundi’s exports of differentiated goods to the EAC make up about two thirds of trade. Most of Burundi’s imports are in differentiated goods—though as a share of total imports from its trade partners, differentiated goods are less important among the members of the EAC.
How Important Is the Extensive Margin of Trade? 3.51 Finally, a decomposition of trade into the extensive and intensive margins is conducted to understand their importance for Burundi. The extensive margin is defined as exports of new products and/or to new partners, and the intensive margin is products that were already exported in the second half of the 1990s. Specifically, the shares of trade arising from new and existing goods were calculated. The base period is average trade from 1995 to 1999.57 3.52 The EAC offers a good opportunity for Burundi to export new products. On average, the extensive margin accounted for about one quarter of exports from 1999 to 2007. The extensive margin is more important for exports to the EAC, where it accounted for about three quarters of exports. Only exports of new products were considered to determine whether the EAC is used to test new products.58 New products accounted for about 7 percent of trade in 2007 overall; in the EAC, this amount rises to 28 percent. This implies that the EAC may offer some advantages in terms of getting into new products. These results also imply that entering new markets with existing products is more common than finding new export products. On the import side, the extensive margin is important because it may indicate consumer gains from variety. Whereas about one third of Burundi’s imports are new country-product pairs, the figure rises to more than one half for the EAC. Gains are likely to be even greater in completely new goods; however, the share of completely new imports is very low. 3.53 Burundi, however, might not be able to successfully maintain and grow new exports. The extensive margin is measured as products/markets that are new since 2000. One issue with the figures, whether for new product markets or new products, is that they tend to be very volatile from year to year. This is surprising because, over time, one would expect the new exports to become more important (assuming they grow) because the base year is not changing. If Burundi enters new markets and grows its exports in later years, a larger extensive margin over time should be observed. The persistent volatility of the extensive margin suggests that Burundi may be struggling to maintain and grow these new exports.
Benefits for the Industrial and Services Sectors from Regional Foreign Direct Investment 3.54 The flow and sustainability of FDI, sources of technology transfer and know-how, as well as job creation will depend on the capacity of Burundi to offer economic externalities. According to the theory of new economic geography, externalities include (1) the size of the market; and (2) the existence of intermediary services generated by the presence of public goods that reduce transaction costs (Krugman 1995; Fujita and Thisse 2002), as well as auxiliary services. Whereas market size refers to the possibilities of generating economies of scale (through openness and competition), intermediary services are derived from public goods such as infrastructure for transport and energy, hospitals, and schools.
57 ? ? The decomposition of trade is as follows: ͕͕͙ͨͣͨͨͦ͘͠ Ɣ $j ͐$%/ ƍ $j ͐$%/, where Vit is the value of exports of product i to country j in year t, I is the set of products that exists in both the first period and period t, and N is the set of new products in year t that is not in the country’s 1995–99 export bundle. This is an identity where total trade is decomposed into two parts: (1) products that were exported (imported) in periods, the intensive margin; and (2) exports (imports) of new products. The share of trade resulting from the extensive margin is defined as the new goods relative to total trade. 58 In this case, new products are counted as exports of products that were not exported between 1995 and 2000. New markets for existing products are not counted.
66 There is, however, a minimum threshold below which these externalities do not produce economies of agglomeration (Hugon 2003). 3.55 Concerning the regulation, overall, the reforms under way have made encouraging, but insufficient, progress in a short time. It will be unlikely that, in a competitive market, companies are positioning themselves in areas where the business environment remains difficult, leads to additional costs, or is simply ineffective. 3.56 Regarding industry and service sectors and companies, the strategic regional partnership seems to be the best strategy for an effective integration, and the potential to attract regional FDI exists. Provided that the prerequisites mentioned above are gradually put in place, the interest of this partnership for regional FDI will be twofold because firms can (1) increase profit margins resulting from economies of scale generated by the enlargement of the market; and (2) relocate to production centers to lower production costs, particularly those of labor. In Burundi, wages are low and labor is available— potential incentives to relocate. 3.57 This partnership could take the form of joint ventures with local small and medium enterprises, equity participation in the shareholder, or subsidiaries from which the economy would take the maximum benefit. There are several benefits of FDI for the domestic market, including large- scale production with lower prices, higher standards of quality, technology and know-how transfer in the industry, expansion of the tax base, and more investment flows and job creation with a multiplier effect on the national income.
The Role of Regional Infrastructure 3.58 Public infrastructure plays a critical role in promoting economic growth. Here, infrastructure is understood broadly to include transport (via roads, rail, ports, and airports), energy, telecommunications, and other externalities (financial markets, commercial and institutional basic services). In a regional context, infrastructure influences the spatial concentration through two channels: (1) lowering transport and transaction costs, and (2) reducing fixed and variables costs. The development of such infrastructure on a regional level accelerates the achievement of a threshold effect necessary to realize positive externalities, and thus leads to the stimulation of regional growth. 3.59 Burundi faces a substantial infrastructure deficit. An infrastructure action plan that has been prepared by the African Development Bank reports that, because of the limited access and the poor state of infrastructure, the cost and adequacy of these services affect commercial opportunities for the whole business community (farmers, entrepreneurs, and small and large businesses). Such a situation leads to substantially higher costs, undermining the competitiveness of Burundi’s business in regional and global markets. (The infrastructure action plan is described in more detail in section A of this chapter.)
The Possibility of Burundi Becoming a Regional Integration Financial and Transport Hub 3.60 The role of a transshipment hub using the port of Bujumbura could be a viable option for Burundi. Because of its geostrategic location, one expects that the port of Bujumbura has the potential to return to its former role as traditional gateway to providing goods and transport services to countries bordering on Lake Tanganyika, especially to the eastern portion of the Democratic Republic of Congo. In the context of regional integration, development of intense mining and other economic activities in Burundi, eastern Democratic Republic of Congo, and Rwanda could efficiently use the services of the port of Bujumbura. In addition to its location, it is adequately equipped and currently operating at less than 50 percent of its capacity, according to statistics from the port authority. Mining activities could include exploitation of the mineral potential in eastern Democratic Republic of Congo, with all related possible economic activities and economies of agglomeration. 3.61 The future role of the port of Bujumbura, however, primarily depends on developments regarding the Tanzania Railways Corporation. Until recently, the bulk of cargo movements between
67 Burundi and Dar es Salaam was carried by rail to and from Kigoma (Tanzania) or Mpulungu (Zambia), and by vessel from these ports to and from the port of Bujumbura. In recent years, lack of suitable rolling stock and the age of the cargo-handling equipment at the port of Kigoma, and the deterioration of the Tanzania Railways Corporation have seriously undermined the activity of the port of Bujumbura. The volume of trade entering and leaving Burundi through the port has declined sharply and is now a third of what it was in 2000. Possible scenarios range from a disappearance of the need for transshipment from rail to vessels at Kigoma if a proposed rail extension into Burundi is constructed to carry freight and passenger traffic direct to Bujumbura, to an improvement in Tanzania Railways Corporation’s performance and a move back to rail transport for bulk cargo. Even in the most optimistic case, though, only modest investments to upgrade equipment would be necessary. 3.62 Concerning the financial sector, recent studies recommend focusing on domestic issues in the short and medium terms. Priorities are strengthening the existing structure, improving the access to financial services, and lowering their cost. Certainly, the proximity of an economic area such as the eastern Democratic Republic of Congo, with intense economic activity operating entirely in the informal sector, can be seen to have large potential. Burundi, however, first needs to strengthen its domestic financial, regulatory, and institutional base, and bring them to the level of regional standards in the EAC. This means that Burundi is facing two constraints: (1) without an efficient and solid financial sector that serves a large proportion of the population, Burundi will not attain a satisfactory level of growth and private sector development, and poverty reduction will be slow; and (2) Burundi needs to fill the regulatory and institutional gap that exists between it and the four other EAC member-countries. Consequently, it is too early for Burundi to seek a specialized niche within the EAC where the country may develop a comparative advantage and assume a regional leadership role, including expanding services to the eastern part of the Democratic Republic of Congo. Over the next five years, Burundi should focus on these domestic issues, with the EAC as a benchmark that guides regulation, supervision, and institutional strengthening.
B3. Risks and Risk Mitigation 3.63 In a country that has a large deficit of technical and institutional capacity, the challenge posed by a regional integration policy is enormous; and there are many risks that need to be mitigated. These risks include the following: 1. Capacity—Weak technical and institutional capacity may delay or compromise the implementation of the agreed program. In this regard, a capacity-building program is being developed with assistance from the U.K. Department for International Development (DfID); but permanent technical assistance will be needed, at least in the first years. 2. Trade diversion—Although trade diversion remains a risk, the possible negative effect is expected to be limited, given the small range of products traded in the EAC. 3. Loss of revenue resulting from adoption of the common external tariff—A study of the fiscal impact estimated a loss of 1.7 percent of total revenue. This is not a substantial loss, given existing alternatives of compensation, mainly through tax reform (expanding the tax base, eliminating exemptions) and the mechanisms of compensation agreed between COMESA and the EAC for Burundi and Rwanda. 4. Indirect costs—Indirect costs of regional integration are related to the capacity of regional hubs to attract and combine the best opportunities (research, education, employment, and so forth), leading to brain and capital migration. Other indirect costs are related to the likelihood of criminal activity taking advantage of free movement in a fragile area. The first problem is a political economy issue and needs to be addressed on the community level. The latter poses a security problem and also needs to be addressed at the community level; it further requires the setting up of a monitoring mechanism and the establishment of regional security.
68 5. Other risks related to persistent insecurity at the western boarder of Burundi—Although establishing technical committees for monitoring and evaluation and donor support should mitigate such risks, the government should remain focused on its regional commitment to keep the momentum. 6. Timing of the implementation of priority measures—The timing is considered very ambitious for the newest member-states—especially Burundi. It may undermine the states’ implementation. In this regard, the measures are regularly negotiated according to the principle of variable geometry, recognized within the EAC.
B4. Policy Recommendations 3.64 The adoption of a strategy based on a comprehensive approach to regional integration is the appropriate way to maximize the chances of success in the accession of Burundi to the EAC. It is clear that membership in the EAC is a significant challenge for the country. Weak technical and institutional capacities are major constraints. However, these problems can and should be addressed. The main recommendations to advance the regional integration process are these: 1. Comprehensive approach—Lessons from past experiences show that the basic principles of regional integration are important. The authorities must give them the consideration they deserve. For Burundi, adopting a strategy based on a comprehensive approach to regional integration seems to be the appropriate way to a successful accession to the EAC. 2. Improvement of infrastructure—This strategy requires that national and regional infrastructure (energy, transport, and communications) and regulatory reforms form the backbone of a successful integration with the EAC. 3. Sequenced process—A successful accession strategy will require a sequenced process that gives high priority to immediate measures while gradually implementing those for the medium and long terms. To this end, the policy matrix gives a comprehensive and sequenced action plan where immediate priorities are clearly identified (see table A.7 in appendix 3). The immediate priorities and objectives of the action plan are the following: • Continue institutional reforms—Implementing institutional reforms is important; and the government needs to keep the momentum, particularly by organizing training sessions and mobilizing the budget required to fully participate in the regional meetings. Measures include recruiting qualified staff for the Ministry of East African Community Affairs, acquiring adequate equipment, training staff in English and in negotiation skills, and implementing a communications strategy.59 • Economic reforms—The deepening of economic reforms and the enhancing of private sector economic activity will be cornerstones of success during the first stages of accession to the EAC. This involves the following considerations: ° For trade and investment, (1) support for the diversification and discovery of new markets (Ministry of Trade and the Chamber of Commerce); (2) elimination of remaining NTBs (including those in the financial system); (3) an effective and practical support to promotion of the private sector and improvement of its competitiveness by enhancing the framework for public-private dialogue; and (4) improvement of the business environment by ensuring compliance with EAC standards, and the implementation of an attractive tax code.
59 See table A.6 in appendix 2 for a matrix of strategic and indicative outputs for the Ministry of East African Community Affairs.
69 ° For the industry and services sector, a strategic partnership is a viable option. Burundi’s private sector should try to benefit from FDI, particularly in terms of technology and know-how transfer. ° For services and service delivery, the priority is to reform and develop the financial system, including the strengthening of banking sector activities, to improve the port services and develop the information and communication technologies sector. ° For infrastructure, special emphasis should be placed on energy, transport, and communications. Given the required capacities, the government should anticipate the strengthening/training of its human resources and the request for adequate technical assistance, when necessary. 4. Projects for the medium and long terms—In the meantime, the government could start or continue discussions on the medium- and long-term projects to prepare feasibility studies and mobilize resources from donors and the private sector. 5. Coordination mechanism—A coordination mechanism is critical at this stage. It should include the government, the donor community, the private sector, and civil society. It can clarify the roles of the potential actors and help maximize their contributions. 6. Communication mechanism—The communication mechanism is essential and needs to be finalized and implemented as soon as possible. (An action plan is currently being finalized with technical assistance from the U.K. Department for International Development.)
C. REFORMS NEEDED TO INCREASE THE PRIVATE SECTOR’S CONTRIBUTION TO GROWTH 3.65 Burundi has one of the most unfavorable business environments worldwide. The business environment and, more specifically, the investment climate is considered essential for a successful diversification of the economy (Page 2008). The private sector in Burundi, however, faces several major constraints (discussed in section A of chapter 2). The regulatory framework is poor: administrative and bureaucratic procedures are cumbersome, the tax burden is one of the highest in the world, legislation is outdated, and the capacity to enforce contracts is weak. The economy is dominated by public enterprises, several of which face financial problems. Moreover, the financial sector is small and not well developed; and access to finance is a major obstacle for companies in the formal and informal sectors. 3.66 A number of measures are necessary to enable the private sector to contribute more to economic growth. Several regulations are already under revision or have been revised, and the momentum must be sustained. Given below are recommendations to overcome the main obstacles the private sector in Burundi is facing. They are in line with the areas of activity outlined in the country’s PRSP (Republic of Burundi 2006) regarding private sector development: (1) short-term emergency actions, (2) ongoing reforms of the legal and regulatory frameworks, (3) reinforced incentives for growth in private investment, and (4) government disengagement from the productive sector and privatization of public enterprises. 3.67 In this section, the way forward is outlined to improve the business environment in Burundi. Recommendations are given to overcome supply-side constraints, continue the improvement of the regulatory environment, increase competition, enhance the country’s image as a business and investment destination, and strengthen the dialogue between the government and the private sector.
C1. Overcoming Structural Constraints 3.68 To provide the private sector with the necessary means to operate its firms and increase productivity, it is essential to fill the main infrastructure gaps. An infrastructure action plan— specifically addressing the energy, transport, and communications sectors—provides details on how these
70 obstacles can be overcome (see section A). Burundi can also benefit from the regional integration process that is currently occurring. Potential gains for the private sector include improved links between the domestic and regional markets, regional infrastructure projects, and an incentive to accelerate the reform agenda (as outlined in section A).
C2. Addressing Supply-Side Constraints 3.69 Regarding the financial sector, Burundi is facing two constraints. First, without an efficient and solid financial sector that serves a large proportion of the population, Burundi will not attain a satisfactory level of growth and private sector development, and poverty reduction will slow. Second, Burundi needs to fill the regulatory and institutional gap with the four other member-countries of the EAC. Clearly, as a matter of priority, Burundi needs to focus on strengthening its financial sector, improving access to financial services, and lower their cost. Over the next five years, the country must focus on domestic issues and view the EAC as a benchmark that guides regulation, supervision, and institutional strengthening. It is too early for Burundi to seek a specialized niche within the EAC, where the country may develop a comparative advantage and assume a regional leadership role. 3.70 Recommendations for the financial sector are articulated around two main schemes: shoring up stability and improving access. This means taking actions on both the supply of and the demand for financial services. A summary of the necessary measures to be taken to improve the performance of the financial sector is given in table A.13 in appendix 5. Key recommendations include the following: 1. Increased stability • state banks—privatize state-owned banks and improve supervision; • payment system—introduce a clearing and real-time gross settlement system; • microfinance—strengthen supervision and institutions; • insurance sector—establish independent regulatory/supervisory agency; and • pension system—change management structure, internal restructuring, and parametric reform. 2. Improved access • small and medium enterprises and rural areas—develop new products and lending mechanisms, build the capacity of the Banque Nationale de Développement Économique (BNDE) to lend in rural areas; • microfinance—change regulation to broaden the scope of microfinance institutions to develop new products and reach more customers; • housing finance: authorize microfinance institutions to offer mortgage loans, decentralize land registries; • insurance sector—diversify products, develop insurance brokers; and • availability of information—strengthen existing information centers, offer assistance to improve financial statements and business plans, strengthen the accounting and auditing profession. 3.71 To promote technical and professional skills and facilitate transparent procedures in the labor market, policy recommendation include the following (World Bank 2008a): 1. In the short term, • a needs assessment of required skills for private companies;
71 • evaluation of the possibility of joint programs between public institutions and private firms to improve technical and professional skills; • facilitation of the learning of English; • reexamination of labor market regulations. 2. In the medium term, • improvement of employees’ skills by making schooling available to a larger part of the population, providing more opportunities for technical and professional education, and strengthening existing skills; • revision of certain elements of the labor code. 3.72 In addition, the flow of information on the labor market needs to be improved by supporting organizations that facilitate the match of supply and demand for labor by circulating relevant information to a wider audience. It would be possible to create a national employment fund with a focus on improving the availability and flow of information, setting up programs to improve technical and professional skills based on market needs, and supporting employment creation in general.
C3. Continued Improvement of the Regulatory Environment 3.73 Continually improving the regulatory environment is necessary. The currently ongoing reforms need to be completed as soon as possible and the appropriate regulations implemented. Specific suggestions include the following: 1. Adoption by Parliament of the revised Commercial Code and the Code for Public and Private Companies. 2. Finalize the revision of the tax code. Furthermore, a study of the reform of the tax code completed in 2007 suggests replacing the impôt forfaitaire (lump-sum tax) and other small taxes paid by small enterprises with a simplified impôt synthétique (synthetic tax) based on verifiable and mostly quantitative criteria.60 The new tax would then be incorporated in the general tax code under revision. 3. Finalize the revision of the mining code. 4. Prepare the accompanying regulation for commercial codes. 5. Further improve the efficiency of the commercial court and the arbitration centers.
C4. Improved Governance 3.74 The government recognizes the important role of improvements in governance. A National Governance Strategy and Action Plan were officially launched in early 2009, and the government committed to approve it before the end of the year. The design of the plan follows a systematic and participatory approach aimed at integrating the different dimensions of governance. A Guide des Consultations was prepared, outlining the context; main recommendation of previous studies on governance and corruption in Burundi; and the framework, methodology, and organization of thematic focus groups (Republic of Burundi 2009).
60 This reform would have several benefits: (1) it would help draw into the formal economy those informal entities whose primary motivation for evading tax is not fraud, but an inability to meet the complex and burdensome rules and regulations; (2) newly formal businesses would have better access to credit and government contracts, thus increasing their contribution to the Burundian economy; (3) the reform would expand the tax base and provide the government with a more predictable revenue stream; and (4) it would reduce the possibility of corruption in the present method of estimating business turnover. The government is expected to seek technical advice (from the Foreign Investment Advisory Service) on the structure of the new impôt synthétique.
72 3.75 General recommendations include • implementing a policy of good governance and fighting against corruption; • involving civil society and the private sector during the entire elaboration process, and following up and evaluating policies and projects; • strengthening the capacity (training and equipment) of public institutions, civil society, and private sector organizations that are involved in promoting good governance and the fight against corruption; • reforming the government and the public administration by putting in place or improving existing institutional and legal frameworks; and • disseminating relevant judicial tools for good governance and the fight against corruption. 3.76 Several specific recommendations are given as well, grouped into three categories: (1) socioeconomic governance; (2) political and administrative governance; and (3) governance related to justice, peace, and security. 3.77 Burundi could benefit from an increase in competition. A draft Privatization Law was prepared, approved by Parliament, and promulgated by the government; but the new text is considered unsatisfactory by the private sector, donors, and many public institutions. The government has agreed to undertake a revision of the new law. A consultant was identified to complete the revision and enable the government to approve a new draft and submit it to Parliament by the end of 2009. The objective of the new law would be to modernize the legislative framework to ensure that the government’s withdrawal from productive sectors will not result in noncompetitive market outcomes (creation of private monopolies). The Privatization Law needs to be in line with modern international practices. 3.78 The ongoing privatization process of the coffee sector is an encouraging sign (see section B2 of chapter 2 for details). Whether the economy will benefit, however, depends on the continued implementation of the privatization agenda. In general, the performance and corporate governance of public enterprises need to be improved.
C5. Improvement of the Country’s Image as a Business and Investment Destination 3.79 Burundi needs to improve its image and create incentives to attract investment. The credibility of the country as a business destination for domestic and foreign investors must be strengthened. Policy recommendations in this regard include the following: • develop a communication strategy to inform the public about the measures taken to improve the business environment and about progress made regarding political and social stability; • regularly update the investment guide for potential foreign investors; • adopt an action plan aimed at reforming business regulation and the practice of creating and operating businesses through a substantial reduction of the time and costs associated with these procedures
C6. Strengthened Dialogue between the Government and the Private Sector 3.80 The dialogue between the public and private sectors in Burundi historically has been very weak. Limited consultation and partnership with the private sector have tended to result in poorly informed public sector decision making that often fails to foresee the impact of decisions on the private sector. A general lack of mutual accountability between the public and private sectors has also resulted in very weak policy implementation. In addition, many of the private and public enterprises and organizations (including the Ministry of Commerce, the National Bureau of Standards, the BRB, the Chambers of Commerce, and the accounting profession) need support to strengthen their capacity to
73 provide policy leadership and create the enabling environment for businesses to grow and be competitive domestically, regionally, and internationally. 3.81 Private enterprises believe that one of the most important initiatives to improve the business climate in Burundi is to strengthen the policy and operational dialogue between the government and private enterprises. In June 2008, a presidential decree established a public-private sector consultation framework—including a general assembly, technical groups, and a permanent secretary; to day, however, no meeting has been organized by the government. The private sector also needs to decide which organizations will represent its interests in this dialogue with the government.
D. ENCOURAGING EMERGING GROWTH SECTORS 3.82 The strategy to encourage emerging growth sectors in Burundi over the short and medium terms proposed in this Country Economic Memorandum is based on two pillars. First and foremost, the country must address the low performance of the agriculture sector, including food and export crops. Priority should be given to eliminating widespread food insecurity and malnutrition. Furthermore, revenues from current exports (mainly coffee) need to be stabilized and increased. Second, in the medium term, diversifying the economy and strengthening exports could lessen the nation’s vulnerability to external shocks and create new opportunities for economic development and employment—especially off-farm employment. The measures described in the previous sections on infrastructure, regional integration, and the business environment will enable the development of the emerging growth sectors.
D1. Agricultural Recovery: Food Crops—Prospects and Priority Interventions 3.83 Given the projected increase in population, the land situation is likely to worsen, reinforcing the need for off-farm employment (see section A of chapter 2). Although additional marshland can be developed, it will not be sufficient to meet the needs of the growing population. There is, therefore, a need to shift from extensive to intensive agriculture. Given the current low productivity, the potential for intensification is real. Improving input delivery and soil and water management is essential. 3.84 Burundi has the potential to produce a wide variety of crops. It has sufficient rainfall in general and water resources for extended irrigation systems to withstand localized droughts. There are significant deposits of mineral resources, such as phosphates or limestone that can be used to produce fertilizer— especially for the acidic soil. The country also has an abundant workforce and a number of experienced technicians who can be trained in the use of new technologies. Strategic investments are necessary to implement improved technologies (irrigation and drainage systems, terracing, and so forth). 3.85 The recent accession to the EAC offers a range of possibilities for Burundi. Extending import and export markets, however, requires a lowering of the transport costs to enhance the competitiveness of Burundi’s products (see section A of this chapter). 3.86 To use the potential of the country and increase the performance of the agriculture sector, several priority interventions are necessary. Cross-cutting interventions include rehabilitating rural infrastructure, improving delivery of financial services to the rural sector, and improving the business climate. Specific measures are listed in table 3.6. In addition, a mechanism for monitoring and evaluation must be put in place.
74 Table 3.6: Specific Measures to improve the Performance of the Food Crops Sector Objective Measure Sustainable growth of agricultural Intensify food crop production, by productivity and production - strengthening input distribution systems - improving soil and water conservation and management - reinforcing technology development and transfer Rehabilitate tools of production and infrastructure rationally manage natural resources, by - restoring, improving, and conserving soil fertility in particular and natural resources in general - rehabilitating, creating, and strengthening local infrastructure for storage, conservation, transformation, and commercialization of agricultural products - improving and developing road infrastructure to facilitate access to markets and distribution of inputs Promote lasting production systems and develop existing potential Professionalization of producers and Organize, structure, and professionalize producers development of private initiatives Involve the private sector in modernizing the sector Strengthening of management Revive, professionalize, and decentralize support structures for production capacities and development capacities Contribute to the protection of agricultural workers against nutritional of the sector deficiencies, malaria, and HIV/AIDS Improve and strengthen the capacities of the government to project/forecast market developments and opportunities Sources: Baghdadli, Harborne, and Rajadel 2008; Republic of Burundi 2008b.
D2. Coffee—The Way Forward 3.87 Burundi’s coffee sector could improve its performances considerably within the next five years if it undertakes key measures to increase production and enhance quality. According to Clay, DeLucco, and Ottaway (2007), production could grow by 25–35 percent in five years if incentives to producers are rethought, access to inputs is improved, a greater focus of processing practices on quality is ensured, and selling strategies are adapted to target specialty markets. Another study (by the OTF Group [2008]) estimates that Burundi’s coffee export revenues could more than double within five years, thanks to targeted interventions. Promoting the use of fertilizers could raise production by 20 percent, while developing good agricultural practices (such as mulching) could generate an additional increase of 12 percent. Improving washing stations’ processing methods would enhance quality, which could translate into a 15 percent price increase. Finally, modifying marketing practices and branding Burundian coffee could help prices grow by 10 percent. To achieve such results, the industry will need to (1) foster an enabling environment; and (2) introduce improvements along the value chain, both immediately and in the medium term.
Immediate Actions 3.88 Setting the market rules and the structure of the industry is indispensable to ensure a proper development of the coffee sector and attract the type of investors required to enhance its performances. In addition, improving the industry’s practices should start at the production level.
Fostering an Enabling Environment 3.89 To improve the business environment in the short term, the following actions are recommended: 1. Strengthen the capacity of the regulatory agency (ARFIC)—A regulatory agency was created by presidential decree in June 2009. It should swiftly establish an efficient regulatory framework for the industry, a prerequisite to attract first-class investors. Privatizing and liberalizing of the sector
75 will not guarantee that competition prevails after the reform; it will be up to the agency to introduce rules fostering competition and to ensure that they are followed. 2. Pursuing the privatization agenda—Bidding invitations for the sale of the washing stations and dry mills were published on June 15, 2009. The government lifted some of the obstacles to the divestiture (for example, amendment of the 30-year lease contract with the Sociétés de Gestion des Stations de Lavage [SOGESTAL - the washing stations management companies] and the Société de Déparchage et de Conditionnement [SODECO - the hulling and processing company] on August 21, 2009). But it will need to address further issues in the upcoming months. Reforming the coffee sector is a sensitive matter in post-conflict Burundi, and trying to build a consensus is essential. The ministry responsible for governance and privatization launched a communication campaign to inform the general public about the process and consult with all stakeholders of the industry, some of whom are voicing strong concerns. The employees of the washing stations, dry mills, and the Office du Café du Burundi (OCIBU), for instance, are understandably preoccupied with how privatizing will affect their jobs. The dissatisfaction of producer organizations with the proposed ownership structure of the washing stations also will need to be addressed. The credibility of the process depends on the government’s ability to stay the course and deal with these obstacles. This steady focus will assure stakeholders that the privatization will proceed and be run smoothly, which is essential to retain serious bidders in the process. The intricacy of the process and the upcoming elections should not be allowed to interrupt the ongoing privatization because the coffee sector cannot afford further confusion. 3. Launching the second round of privatizing washing stations and dry mills—With about 10 percent of privatized assets, managing the sector could be more complicated than is anticipated. The government would have to launch a second round of bidding to enable new owners to be operational during the 2011/12 crop season. The best period to launch such a tender of sale would be eight months prior to the beginning of the crop season.
Improving Production Practices 3.90 In the short term, production practices can be improved by the following efforts: 1. Developing and strengthening producer organizations—Strengthening producer organizations should accompany the privatization process. Existing organizations are believed to represent merely 20 percent of Burundi’s coffee growers today. Because the sector relies on smallholders, enhancing quality and increasing output greatly depend on the development of producer organizations. Such associations are also an indispensable means through which growers can participate in the industry, as shareholders of the washing stations or as promoters of their own coffee. 2. Alleviating growers’ financial constraints to improve their access to inputs and replace aging trees—Two of the industry’s immediate priorities should be to increase output and improve quality, starting at the production level. Some major obstacles to raising yields and minimizing cyclical variations are growers’ low use of inputs, poor agronomic practices, and their inability to replace aging trees. Promoting microcredit and matching grant programs would be instrumental in helping farmers finance the purchase of inputs or the replacement of trees. Such programs should be set up within a year, and should be designed to foster producer organizations.
Medium-Term Actions 3.91 The regulatory framework and proper market structure should be complemented by other measures taken within the next two to three years to promote the industry’s development. Improvements at the processing stage should also be introduced within that time frame.
76 Fostering an Enabling Environment 3.92 In the medium term, the following actions can promote an enabling environment in the coffee sector: 1. Improving financial services—As is the case for the rest of the economy, the scope of financial services offered to the coffee industry is relatively limited. Processing factories, for instance, have little access to working capital because the banking system in Burundi has little experience (and interest) in offering such services to small businesses. Incentives should be designed to encourage banks to develop such products, and technical assistance should be provided to key institutions to disseminate good practices. In addition to financing issues, the industry suffers from a lack of financial advisory services. Although these services are virtually nonexistent in Burundi, they would substantially benefit the coffee sector, notably by helping it optimize revenues from the sale of green coffee on commodity markets. Building such capacity should be a priority in the upcoming years. 2. Strengthening the capacity of the interprofessional association (INTERCAFE)—The coffee sector needs an entity devoted to (1) representing it in exchanges with the government and the regulatory agency, (2) coordinating actions aimed at improving performances along the value chain, (3) promoting Burundian coffee on global markets, and (4) fostering market intelligence. An interprofessional association is instrumental for an industry intent on accessing specialty markets. Financed by a small tax on coffee exports and/or membership fees, for instance, it can represent all stakeholders in the sector (producers, washing stations, dry mills, roasters, and exporters) and enable them to voice their concerns and needs. It can be a valuable partner for the state and the regulatory agency, especially in the phase immediately following privatization. It can also be a forum through which the industry could coordinate efforts aimed at improving quality. The government recently has supported the creation of an interprofessional association whose capacity needs to be reinforced to contribute to the management of service delivery.
Improving Processing Practices 3.93 To improve processing practices in the medium term, the following actions are recommended: 1. Introducing traceability processes—Improving washing station and dry mill practices will go along with the privatization process, which offers stakeholders more visibility. The first priority in the next two to three years should be to introduce mechanisms to ensure proper traceability throughout the value chain. It is indispensable to separate high- and lower-quality coffees to channel produce to the right markets and obtain premiums for quality. In addition, traceability procedures will help improve incentives along the value chain, enabling the sector to reward better quality. 2. Initiating coffee differentiation—When well-functioning traceability mechanisms are in place, a priority will be to promote coffee differentiation to effectively penetrate specialty markets. Accomplishing this will require investing in cupping laboratories, initially in regions most likely to produce specialty coffee; and then, if appropriate, investing in relevant washing stations. Developing appropriate know-how and expertise, both at cupping facilities and washing stations, will be crucial and will rely on the sector’s ability to bring in expertise from other coffee- producing countries. 3. Promoting collaboration along the value chain to enhance quality—The coffee sector’s development is currently impeded by a lack of coordination along the value chain to enhance quality. This is primarily the result of to ill-designed incentives—which deregulation and privatization are expected to improve. But programs intervening at each key stage of the value chain could prove invaluable to support Burundi in catching up with its competitors. Such
77 programs are useful in disseminating best practices and can demonstrate to potential investors Burundi’s capacity to produce high-quality coffee. These interventions could benefit from a scaling-up and could be complemented by financial support to replace outdated equipment and invest in new assets.
D3. Diversification of the Economy 3.94 Burundi currently has a very narrow export base, but has the potential to notably diversify its economy. A successful diversification of its economy is essential over the long run to be able to better withstand external shocks, generate income, and create employment (especially off-farm employment). Research has shown that there is a link between increased diversification and higher GDP per capita, where diversification usually leads to higher income levels. 3.95 To realize this transformation, a number of steps must be taken by the government and the private sector. Here are some general suggestions: • Focus on existing products. Burundi should focus on existing products and those well within reach to start the diversification process. The selection of products and sectors needs to be based on detailed information and thorough analysis. Possible products are those that have been identified as emerging champions, as classics with a currently low export share, and as marginals that have the potential to develop a revealed comparative advantage. Existing strategies, such as the National Agricultural Strategy (Republic of Burundi 2008b), should be used to guide the diversification process. • Diversify gradually. Diversifying should be a gradual, sequenced process that starts with the expansion of the current market share, a diversification of unsophisticated primary products, and a subsequent move to more sophisticated and eventually manufactured products. Sectors that have the greatest potential are mining, agro-processing, and fishing. • Benefit from regional integration. The ongoing regional integration process should be used to maximize Burundi’s benefits from its accession to the EAC. Reduced tariffs and NTBs will enable Burundi to have easier access to a larger market, facilitating an increase in exports to the regional market. This market also can be used to introduce new products and build expertise and skills before attempting to access international markets. To a large extent, Burundi can also benefit from regional infrastructure projects that could improve transport and logistics significantly. • Close the infrastructure gap. The large infrastructure gap needs to be closed. This applies to a range of areas, including roads, air transport, and electricity. Regarding transport infrastructure, the insufficient availability of cold storage and a cold chain, in general, needs to be addressed. (A detailed infrastructure action plan is discussed in section A). • Improve the business environment. The business environment must be improved to attract foreign investors as well as to provide incentives for domestic investors and enterprises. The business environment in Burundi is one of the most unfavorable in the world. Reforms need to be implemented without delay (as discussed in section C). • Comply with sanitary and phytosanitary standards. Burundi needs to develop an action plan and establish the necessary facilities/laboratories to comply with international sanitary and phytosanitary standards. Without compliance with these standards, it will be extremely difficult to export outside the region. Burundi could benefit from the experience of other EAC members in this area. 3.96 Below is a discussion of specific sectoral recommendations.
78 3.97 Necessary interventions for the fishing sector include the following: • developing a comprehensive strategy for the sector, involving all stakeholders; • improving the capacity to manage the natural resources—implying, for example, the implementation of annual fishing quotas to avoid overfishing and to guarantee a responsible, profitable, and optimal exploitation of the existing fishing resources without conflict between the main economic actors; • setting up a regional coordination mechanism to ensure proper implementation of programs and projects in the sector; • promoting the participatory management of fish and developing comprehensive legislation/regulation of fishing and fish farming; • promoting commercial fish farming; • improving access to finance; and • complying with sanitary standards. 3.98 A number of interventions can be suggested to increase the performance of the horticulture sector, including the following61: • Overcome logistical gridlock. Burundi has only limited and inflexible cargo capacity available. Possible solutions are chartering planes and negotiating with airlines to make additional cargo space available. The infrastructure action plan (presented in section A), also addresses the expansion and modernization of the international airport in Bujumbura, with the objective of obtaining the International Civil Aviation Organization’s “Certificate of Aerodromes” (which could attract major airlines and freight companies). • Identify potential markets. There is a trade-off between accessing developed markets and emerging and regional markets. (A summary of opportunities and constraints is presented in table A.3 in appendix 2.) The ongoing regional integration process has to be taken into account as well. As pointed out in the section on Burundi’s benefits risks of membership in the EAC, an analysis of the extensive margin of trade shows that more new products are exported to the EAC than to the rest of the world. Products therefore can be tested regionally before they are introduced to the global market. • Identify products with growth potential. A number of products (listed above) can be identified as having substantial growth potential. • Develop an action plan. The proposed action plan consists of the following components: ° development of a comprehensive strategy for the horticulture sector, involving all stakeholders; ° implementation of a competitiveness assessment of a range of products based on profitability thresholds in promising markets; ° increased production through technical assistance, strengthening sanitary and phytosanitary management capacity, development of management capacities of producer cooperatives and export enterprises, and establishment of a small investment fund to address needs for seed capital; and ° development of efficient supply chains, including a substantial improvement of postharvest logistics and transport infrastructure (cold storage facility, collection/packing facilities that
61 These recommendations are mainly based on Baghdadli, Harborne, and Rajadel (2008).
79 comply with sanitary and phytosanitary standards, and so forth), increased availability of agricultural inputs and packaging material, and the creation of a favorable business climate.
D4. Development of the Mining Sector 3.99 The geology of the country is overwhelmingly the most important factor for all investors, but the government has the potential to make the sector more attractive. Surveys of investors’ decision parameters indicate that the key determinants for engaging in mining development rank in the following order: (1) good geological prospects, mining tradition, and potential; (2) clear and secure mining rights and title (mining legislation); (3) attractive and competitive fiscal conditions (tax legislation); (4) ownership and control of operations (mining legislation); (5) political stability and transparency of governance (government institutions); and (6) availability of infrastructure. 3.100 Policy measures include improving the following: 1. Availability of geological information—This could mean undertaking new geological mapping, with the support of international organizations such as the World Bank. Burundi actually benefited in the late1960s from the United Nations Development Programme’s assistance in achieving an inventory of mineral resources with an evaluation of the mining potential. As a result, most of the deposits, including nickel, were identified. This type of undertaking might need to be updated through newly available technologies. If enough information exists, access to current and pertinent information can be improved by setting up modern geologic data infrastructure. 2. Mining legislation—In 2008, the government launched a reform of the mining code to be completed in 2009. A key aspect of this reform is ensuring the continuity of tenure, because an investor will not invest in the necessary capital-intensive exploration phase without certainty that he or she has the right to exploit the reserves that are found. 3. Mining taxation—Relevant questions have been raised worldwide in recent years concerning the fairness of profit and benefit sharing between the investor and the government. Interestingly enough, tax considerations only come in third position, which tends to show that the government should not be tempted to lower its tax rate too much; rather, it should make sure the level of taxation is fair, predictable, and sustainable. 4. Government participation—The suggestion to have a 15 percent free share for the government should be explored cautiously. Government’s involvement in projects through equity participation may not result in significant dividend flows under normal circumstances (in the case of gold in our projection). Such dividends are sometimes more reliably collected through appropriate royalty and tax regimes, which are consistently applied every year (as opposed to dividend payments that may be forgone during years of expansion or other discrete corporate decisions). The merit of dividends is further questioned by the additional risk that the government must accept by being a shareholder, because the government not only shares the profits of the good years but also shares the potential losses during difficult times—thus putting scarce government funds at unnecessary risk. 5. Governance and transparency—With good governance, the exploitation of mineral resources can generate large revenues to foster growth and reduce poverty. However, when governance is weak, it may result in poverty, corruption, and conflict. The Extractive Industries Transparency Initiative (EITI) supports improved governance in resource-rich countries through the verification and full publication of company payments and government revenues from oil, gas, and mining. Such an initiative is now seen as a positive signal by the industry, which has to commit to the same transparency standards.
80 6. Infrastructure—Government has three options for the terms on which it could offer to provide infrastructure for prospective mining projects: (1) no provision, (2) provision in return for an equity share equal to the value of the infrastructure, and (3) provision in return for user charges. Circumstances will dictate the choice, but it is desirable that government policy be spelled out clearly for investors. 3.101 Although the mining sector can make a positive contribution to the economy when it is properly managed, there also are many countries that do not succeed in turning their mineral wealth into development. This “resource curse” that sometimes prevents mineral-rich countries from reducing their poverty should call for a particular effort in building capacities—not only in geology and mining; but also in mining legislation, taxation, accountability, negotiation, and inspection. This will be even more relevant in a country like Burundi, where the mining tradition is weak. 3.102 Assistance may be needed in negotiating with private investors. If private investor interest in the Musongati nickel mining project is to move forward, the government will need to engage in negotiations with one or more potential investors the terms and conditions for the investment. The government will need to assemble an experienced legal and technical team for these negotiations, and it may need advice and assistance from a donor in this regard. Assistance for negotiations is available, for example, through the recently created Extractive Industries Technical Advisory Facility, funded by the World Bank. 3.103 Decisions on exploitation of a depletable natural resource and use of the proceeds from exploitation must take future income into account. International mineral commodity prices, especially for nickel, have been subject to wide fluctuations during the past 10 or more years. Many countries have experienced major economic management problems as a result of both precipitous declines and rapid increases in prices. Proper stabilization policy is needed. 3.104 The World Bank has proposed a new, integrated approach that can help the government ensure the exploitation of nonrenewable mineral resources is turned into sustainable development. Taking stock of the EITI,62 and tentatively calling it EITI++ because it goes one step farther upstream (where do mining revenues come from?) and one step farther downstream (what are mining revenues used for?), this approach tends to promote sound management of extractive industries all along the value chain, transforming mineral wealth into poverty reduction and sustainable development.63 3.105 Several measures need to be taken to improve the contribution of the artisanal and small- scale mining sector, not only to the government budget and balance of payments, but also to the livelihoods of a large part of the population. The institutional, legal, and regulatory frameworks need to be revised to formalize artisanal and small-scale mining in Burundi. (A project supporting the formalization of the sector should be developed, involving all stakeholders. Detailed recommendations are given in table A.4 in appendix 2.)
E. SUMMARY OF PRIORITY RECOMMENDATIONS 3.106 Based on the analysis carried out in the previous chapters, the following priority recommendations have been identified for the short and medium terms (see table A.1 in appendix 1 for a detailed action plan indicating the time frames, responsible agencies, and approximate costs): 1. Peace and stability must be maintained. Without a stable political and economic situation, a successful economic recovery is not possible. Peace and stability enable national reconciliation, continued democracy, and more effective security expenditure.
62 See details at http://www.eitransparency.org. 63 The EITI++ approach could help the government prepare for mining and for petroleum exploitation.
81 2. Following through on the improvement of infrastructure is critical for development. As detailed in the infrastructure action plan, different scenarios are possible for its implementation. Given the importance of improved infrastructure (especially electricity) for the development of all sectors in the economy, the government should be guided by this action plan and follow through on its implementation. Specific recommendations include the following: • Electricity: The electricity supply must be improved, more provinces connected to the network, and REGIDESO financially restructured. • Transport: Given the geographic isolation of Burundi, the country could greatly benefit from rehabilitating the national highways and the provincial and community roads. Strengthening of the role of civil aviation could attract international airlines and increase the availability of cargo space. 3. Burundi should maximize its benefits from the ongoing regional integration process. To do so, a comprehensive approach that prioritizes and sequences actions needs to be implemented. Most important, it should address institutional constraints and economic reforms. Key recommendations are these: • Institutional reforms: It is essential that the Ministry of East African Community Affairs (MEACA) is to be staffed with competent personnel, and the necessary institutional arrangements must be in place, requiring extensive capacity building. It is important to implement a coordination, monitoring, and evaluation mechanism for all EAC-related issues. Furthermore, a communication campaign aimed at sensitizing the population to support the integration initiative should be implemented. • Economic reforms: Eliminating all remaining NTBs is critical. Furthermore, a strong framework for public-private dialogue is advisable. 4. Economic growth has to be led more by the private sector. A recently conducted Financial Sector Assessment Program, an Investment Climate Assessment (World Bank 2008a), and the regularly compiled Doing Business Indicators show that Burundi’s business climate is not conducive for engaging domestic and foreign investors. Here are the priority recommendations: • Access to finance in the rural areas needs to be improved. Therefore, the regulation should be changed to broaden the scope of microfinance institutions to enable them to develop new products. Similarly, technical assistance should be provided to banks to develop new products and lending mechanisms to better cater to the needs of small and medium-size enterprises and rural areas. The capacity of BRB to supervise microfinance institutions needs to be strengthened. • The regulatory environment should be improved to attract domestic and foreign investment. Critical components are the finalization of the tax code revision, and the finalization and adoption of the revised mining code. • Increase competition and improve governance. Competition could be increased by finalizing and adopting the revised Privatization Law. Adopting and implementing the national policy on good governance and its action plan would demonstrate the recognized importance of improved governance. 5. Burundi needs to improve agriculture productivity. As shown by Baghdadli, Harborne, and Rajadel (2008); the National Agricultural Strategy (Republic of Burundi 2008b); and an analysis of the coffee sector, low agricultural productivity of food and cash crops is one of the main obstacles to growth. As a consequence, Burundi still relies on regular food aid, has limited external resources (other than donor funds), and is vulnerable to external shocks. Key recommendations include the following:
82 • Food crops: Critical for increasing production and productivity is strengthening input distribution systems; rehabilitating production tools and infrastructure (including rural roads); and organizing, structuring, and professionalizing producers. • Coffee: Most important is the complete implementation of the ongoing privatization agenda. Next steps include strengthening the regulatory agency (ARFIC), and interprofessional association (INTERCAFE), facilitating access to finance, and improving processing practices. 6. Burundi could benefit from diversification of its economy. A product space analysis shows that Burundi’s products currently lie on the periphery of the product space, complicating the move to new products. Nonetheless, the analysis identifies sectors and products that have great export potential and are in line with Burundi’s Poverty Reduction Strategy Paper and its National Agricultural Strategy. Key interventions to facilitate diversification and the increased contribution of other sectors, such as horticulture and fishing, are availability of cargo space to export fresh produce and compliance with sanitary and phytosanitary standards. 7. The mining sector could play an important role in Burundi’s economy. Given the potential contribution of the mining sector, a sectoral strategy accounting for the characteristics and constraints of the available mineral resources is essential. Burundi could greatly benefit from joining the EITI and using available facilities for assistance in negotiating with international companies. To account for the large share of artisanal and small-scale mining, a unit could be created within the Ministry of Energy and Mines to deal specifically with this subsector. Developing a project to support the formalization of artisanal and small-scale mining is advisable.
83 CHAPTER 4 IMPLICATIONS FOR THE MACRO-FISCAL SITUATION
4.1 The government has an important role in the recovery of the country. Given the weakness of the private sector and the large needs of a post-conflict country in stabilizing and reconstructing the economy, the government plays an essential role in promoting growth and reducing poverty. The government, however, has only limited domestic funds available and relies heavily on foreign assistance (primarily in the form of grants) financing most of the budget deficit and current account deficit. 4.2 To maintain macroeconomic stability and promote growth, it is critical to improve the country’s fiscal stance. Reforms, however, are complicated by the challenges of rebuilding the country’s social and physical infrastructure after years of conflict. The importance of fiscal discipline is well illustrated in the 2008 Public Expenditure Management and Financial Accountability Review (PEMFAR) (World Bank 2008e). Although some savings may come from reductions in defense spending, they will not be sufficient because expenditures for public security (police) are expected to increase. Consequently, external financing (preferably grants and highly concessional loans) will be needed to finance the fiscal deficit, at least in the short and medium terms. However, given the volatility of aid to Burundi over recent years and the unpredictable nature of external capital inflows, the authorities should find ways to rely less on foreign financing over the long term, particularly by (1) improving the tax administration and (2) strengthening the effective use of public resources for development goals. To do so, it is necessary to establish a sound public expenditure management system aimed at promoting fiscal discipline and the efficient and effective operational performance of the public sector. 4.3 High levels of defense and public security spending are the main obstacle to reallocating public resources toward priority economic and social sectors. Therefore, to the extent possible and keeping in mind the need for security and stability, it is essential to reduce the share of defense and security expenditures over time to create fiscal space for increased spending in the priority economic and social sectors. 4.4 Although the authorities have made significant efforts to improve the country’s fiscal stance, increased spending and inadequate revenue mobilization could create larger, unsustainable fiscal deficits. There is a real risk that a larger deficit may lead to increased borrowing from the domestic financial market. This problem could raise the cost of funds, potentially crowding out private investment and dampening growth. Therefore, effective revenue mobilization and efficient expenditure management are indispensable to maintain growth-supportive macroeconomic policies. 4.5 The allocation and coordination of external aid needs to be examined. In the past, because of the country’s fragile nature, aid was allocated primarily to humanitarian and emergency projects; whereas production sectors, such as agriculture and infrastructure, received limited funds. Unpredictable aid flows have complicated the allocation of aid. When resources previously allocated to humanitarian causes are freed up by stabilization of the political situation, how those resources are allocated must be carefully decided. 4.6 This chapter assesses the macroeconomic prospects for Burundi in the medium term and discusses the medium-term outlook for public finances, considering two different scenarios. The chapter also evaluates the impact of resource allocation on growth and on efforts to achieve the Millennium Development Goals (MDGs), and addresses the question of how aid can best contribute to the development of the economy.
84 A. ECONOMIC PROSPECTS, RESOURCE ALLOCATION, AND ACHIEVING THE MDGS 4.7 This section presents macroeconomic projections and an assessment of the projected public finances. The results are presented according to two different scenarios, a baseline scenario and a reform scenario that assumes the priority recommendations identified in previous chapters are carried out.
A1. Medium-Term Macroeconomic Outlook 4.8 The proposed baseline medium-term macroeconomic framework (2010—15) reconciles the macroeconomic framework of the International Monetary Fund’s Poverty Reduction and Growth Facility program with the Poverty Reduction Strategy Paper (PRSP) targets of the authorities. It is based on recent economic trends and assumptions about future developments (anticipating continued progress of the reform process).64 The baseline scenario reflects the more “voluntarist” policies of the authorities, spelled out in other sources of information for assessing the medium-term growth profile. These sources include sector strategy papers in priority areas, such as education, health, rural sector, energy and public works, the results of a study on the sources of rural growth undertaken jointly by the World Bank and the government (Baghdadli, Harborne, and Rajadel 2008), and the recent performance of Burundi’s economy presented in this report (chapter 1).
Table 4.1: Baseline Scenario—Projections for the Main Economic Variables, 2009–15
Variable Actual Est. Projected Average 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2010 15 Annual percentage change Real GDP growth 5.1 3.6 4.5 3.5 4.0 4.4 4.9 5.5 5.9 6.1 5.1 Consumer prices (period average) 2.7 8.3 24.5 8.8 6.9 8.7 8.7 9.4 9.7 9.9 8.9 Exports of goods and nonfactor 67.8 –27.2 34.6 –15.9 20.9 10.1 8.8 9.4 9.7 9.8 11.5 services (US$) Imports of goods and nonfactor 59.1 3.4 26.2 –8.4 5.0 5.8 6.7 8.0 9.0 9.6 7.4 services (US$) Percent of GDP Current account deficit (incl. grants) –14.5 –15.7 –12.3 –13.0 –13.6 –16.3 –14.0 –13.0 –13.6 –12.6 –13.8 Current account deficit (excl. grants) –36.3 –37.4 –32.8 –28.4 –28.8 –27.7 –26.2 –25.1 –23.7 –22.8 –25.7 Revenue (excluding grants) 18.9 18.6 18.5 19.0 19.1 19.2 19.4 19.5 19.7 19.9 19.5 Total expenditure and net lending 38.2 38.5 44.1 50.4 50.8 45.6 44.9 43.8 40.5 39.3 44.1 Overall balance (excl. grants)a –19.3 –19.8 –25.6 –31.4 –31.7 –26.3 –25.5 –24.3 –20.8 –19.4 –24.7 Overall balance (incl. grants)b –1.8 0.5 –0.8 58.5 –8.4 –6.9 –5.2 –4.6 –2.4 –2.1 –4.9 Gross international reservesc 3.3 3.8 6.3 7.1 6.0 6.0 6.0 6.0 5.0 4.0 5.5 Sources: International Monetary Fund, World Bank, and Burundi authorities. a. Commitment basis. b. Includes grants received under the HIPC Initiative. c. in months of imports.
4.9 The development of the main economic indicators is given in table 4.1. This outlook assumes that • after some softening in 2009, the economic growth rate will get stronger; • inflation will remain below 10 percent, owing to lower levels of petroleum prices, improved liquidity management, and limited financing of government activities by the central bank;
64 Some of the medium-term growth projections are obtained through policy simulations performed with the Burundi Macroeconomic Model developed for the preparation of the medium-term expenditure framework. The growth path in this model is determined by macroeconomic and sector policies that affect long-term productive capacity (human capital, infrastructure, and direct investment flows); short- and medium-term fluctuations in aggregate demand (monetary and fiscal policies); and the external environment (oil price and terms of trade). The above endogenous growth model was not estimated econometrically for Burundi. Rather, elasticities estimated through cross-country regressions were used in the model. Other components of the macroeconomic framework are determined by accounting equations and standard financial programming principles. The medium-term budget framework allocates the envelope of resources into sector budgets based on government priorities dictated by exogenous codes.
85 • fiscal policy will remain prudent, with stable government revenue and gradually declining public expenditures; • mainly as a result of the weakening of the world demand following the current global economic downturn, export sector performance will deteriorate slightly in 2009 and improve in 2010 (whereas import growth will slow down in 2009); • the current account deficit remains fairly constant in the following years; and • external financing will continue to be in the form of grants and highly concessional loans.
4.10 An alternative scenario evaluates the impact of accelerated reforms and increased infrastructure spending on growth, poverty, and attainment of the MDGs. As the experiences of other countries have demonstrated (for example, Rwanda and Sierra Leone), especially in a post-conflict context, Burundi could achieve even higher growth if the commitment of the government to resolutely accelerate the execution of its overall reform program remained strong. In this alternative scenario, the authorities are expected to continue and accelerate reforms to improve transparency and governance in the management of public resources and to create a business environment that is adequate for the development of the private sector. In addition, the government is expected to carry out the infrastructure investments proposed under the priority recommendations in the previous chapters. These reforms are assumed to yield the following results (table 4.2): • There will be a substantially higher growth rate than in the baseline scenario, averaging 8.8 percent for the period 2010–15. • There will be a similar development of inflation, compared with the baseline scenario. • Greater import volume and value triggered by the large infrastructure projects will be countered by greater export volume and value resulting from the better business environment, a more diversified economy, and improved infrastructure. Consequently, the current account balance is projected to be lower, on average, than in the baseline scenario. • There is prudent fiscal policy with revenue projections similar to those in the baseline scenario, but higher spending because of increased capital expenditure.
Table 4.2: Alternative Scenario—Projections for the Main Economic Variables, 2009–15
Variable Actual Est. Projected Average 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2010 15 Annual percentage change Real GDP growth 5.1 3.6 4.5 3.5 4.0 4.2 8.4 11.1 12.5 12.9 8.8 Consumer prices (period average) 2.7 8.3 24.5 8.8 6.9 8.7 9.0 9.6 9.9 9.9 9.0 Exports of goods and nonfactor 67.8 –27.2 34.6 –15.9 20.9 9.8 13.7 17.3 19.1 19.5 16.7 services (US$) Imports of goods and nonfactor 59.1 3.4 26.2 –8.4 5.0 5.6 9.3 12.2 14.1 14.9 10.2 services (US$) Percent of GDP Current account deficit (incl. grants) –14.5 –15.7 –12.3 –13.0 –14.1 –15.2 –11.0 –9.7 –11.4 –12.2 –12.3 Current account deficit (excl. grants) –36.3 –37.4 –32.8 –28.4 –28.8 –27.7 –25.8 –24.2 –22.4 –21.2 –25.0 Revenue (excluding grants) 18.9 18.6 18.5 19.0 19.1 19.2 19.4 19.6 19.8 20.1 19.5 Total expenditure and net lending 38.2 38.5 44.1 50.4 50.2 46.7 47.1 45.3 40.1 36.6 44.3 Overall balance (excl. grants)a –19.3 –19.8 –25.6 –31.4 –31.1 –27.4 –27.7 –25.7 –20.2 –16.5 –24.8 Overall balance (incl. grants)b –1.8 0.5 –0.8 58.5 –8.4 –6.9 –5.0 –4.2 –2.1 –1.7 –4.7 Sources: International Monetary Fund, World Bank, and Burundi authorities. a. Commitment basis. b. Includes grants received under the HIPC Initiative.
4.11 A deterioration in the projections of either scenario is possible, as the effects of the global economic slowdown deepen. First, the overall fiscal deficit (including grants) may be exacerbated if donor countries cut back on international aid in the face of global economic instability. Finding the
86 resources to finance this budget deficit would be difficult, although Burundi may be able to use savings that it expects to accrue from reduced debt servicing following the recent debt relief. Second, a potential stronger decline in migrant remittances could further affect household incomes and the external current account. In the past few years, the volume of remittances (which are transferred through informal channels) has increased significantly, although it is difficult to obtain actual figures. Third, a reduction in projected foreign direct investment is possible, notably in the coffee sector (for example, limited response of foreign investors to the bidding invitations for the sale of coffee washing stations).
A2. Medium-Term Outlook for Public Finance 4.12 Under the baseline scenario, shortfalls in government revenues, resulting from lower-than- expected GDP and trade, and increases in government expenditures are likely to result in reduced fiscal space for economic stimulus and social protection. As a result, the budget deficit (before grants), estimated at an average of 17.5 percent of GDP in 2002–08, is expected to widen to about 31.4 percent of GDP in 2009 (table 4.3). This is partly because of a slight decrease in domestic revenue as a result of the global economic crisis coupled with an increase in total expenditure arising partly from the need to implement the recent peace accord with the last rebel group. Because of increased expenditures in light of the elections and the payment of past government commitments to civil servants, this deficit is projected to further increase to 31.7 percent of GDP in 2010. In 2012, it is projected to decline to 25.5 percent, and to decline further to 19.4 percent in 2015 as revenue increases (partly as a result of improved tax collection performance expected from the newly created Burundian Revenue Office). The ability of the government to reduce military and security expenditures following the anticipated progress of the peace process also explains fiscal position improvement. Furthermore, recurrent expenditures will decrease because of lower interest payments following the January 2009 Heavily Indebted Poor Countries Initiative completion point. The debt stock is expected to fall substantially for the same reason. Wage and salary control will also contribute to expenditure discipline.
Table 4.3: Baseline Scenario—Actual, Estimated, and Projected Government Finance, 2002–15 (Percent of GDP)
Actual Estimated Projected Average 2002–08 2009 2010 2011 2012 2013 2014 2015 2010–15 Indicator (average) Domestic revenue 19.6 19.0 19.1 19.2 19.4 19.5 19.7 19.9 19.5 Tax revenue 17.8 17.1 17.1 17.2 17.3 17.4 17.5 17.7 17.4 Nontax revenue 1.9 1.9 2.0 2.0 2.1 2.1 2.2 2.2 2.1 Total expenditure and net lending 36.9 50.4 50.8 45.6 44.9 43.8 40.5 39.3 44.1 Recurrent expenditure 23.1 25.1 25.5 25.3 25.2 24.5 23.6 22.4 24.4 Salaries 9.3 12.4 12.2 12.0 11.8 11.5 11.2 10.8 11.6 Goods and services 7.0 5.5 6.4 6.6 7.1 7.1 7.0 6.6 6.8 Transfers and subsidies 3.7 5.7 6.0 5.9 5.6 5.3 5.0 4.6 5.4 Interest payments 3.1 1.5 0.9 0.8 0.6 0.5 0.4 0.3 0.6 Other expenditure (incl. exceptional) 1.8 10.0 8.3 3.3 2.9 2.5 0.7 1.4 3.2 Capital expenditure 12.2 15.4 17.0 17.0 16.8 16.8 16.2 15.5 16.6 Domestically financed 3.2 3.6 3.5 3.4 4.6 5.5 6.3 6.9 5.0 Foreign financed 9.0 11.8 13.5 13.6 12.2 11.3 10.0 8.6 11.5 Net lending –0.2 –0.1 –0.1 0.0 0.0 0.0 0.0 0.0 0.0 Primary balance –3.4 –12.5 –10.1 –8.6 –9.8 –10.0 –9.8 –9.0 –9.6 Overall balance (commitment basis) Including grants –3.1 58.8 –8.4 –6.9 –5.2 –4.6 –2.4 –2.1 –4.9 Excluding grants –17.2 –31.4 –31.7 –26.3 –25.5 –24.3 –20.8 –19.4 –24.7 Sources: International Monetary Fund and World Bank staff.
4.13 Although financing requirements appear moderate over the entire period (2010–15) in the baseline scenario, Burundi will remain dependent on external assistance, particularly in the form of budget support, to meet projected external financing requirements. Provided Burundi stays on track with its reform agenda, most of the financing requirements will be covered by budget support; savings recorded from Heavily Indebted Poor Countries (HIPC) Initiative debt relief, multilateral debt relief
87 initiatives, bilateral donors of the Paris Club; and additional debt forgiveness by bilateral donors. Aid pledges (mostly grants) are projected to be at least roughly 25 percent of GDP each year, unless some bilateral donors stop providing budget support. Overall, budget support (including special programs) accounts for about 50–60 percent of total assistance to Burundi. 4.14 The high reliance on external assistance is expected to gradually decline because foreign grants and loans are projected to decrease slightly until 2015. The gradual decline of the high reliance on external assistance could be explained by the projected stability of domestic revenue as well as the commitment of the authorities to rationalize public expenditures and thus create more fiscal space on the basis of domestic resources. 4.15 Capital spending is expected to increase modestly under the baseline scenario. Total expenditure (excluding debt service) is expected to increase initially from 36.9 percent of GDP between 2002 and 2008 to 50.8 percent of GDP in 2009 and to 47.1 percent of GDP a year in 2010–12; however, it will decline to 41.2 percent in 2013–15 (table 4.3). Following a significant increase in 2009 to accommodate the implementation of the peace accord with the last former rebel group, total expenditures are expected to rise again in 2010 (mostly because of the need for resources to prepare for elections). Reduced debt services following Burundi debt relief status in the context of the HIPC Initiative also explains the increase in total spending. Capital expenditure increases to an average of 16.6 percent of GDP over the projection period (2010–15), which is consistent with priorities set in the Poverty Reduction Strategy Paper. Economic and social sectors (education, health, and productive infrastructure) are expected to be the main beneficiaries of this increase in capital spending (table 4.4). The budget share of capital expenditure is projected to increase modestly to 40.1 percent, on average, in 2010–15 (from an average of 33.7 percent in 2002–09). Although still below what is required to achieve the MDG targets, the projected trend is encouraging (figure 4.1).
Table 4.4: Baseline Scenario—Medium Term Expenditure Framework, 2007–15
Actual Estimated Projected Average Expenditure 2007 2008 2009 2010 2011 2012 2013 2014 2015 2010–15 (percent of total expenditure) Recurrent expenditure (excl. debt service) 63.2 60.6 60.4 59.1 59.0 59.3 58.8 58.9 58.8 59.0 Capital expenditure 36.8 39.4 39.6 40.9 41.0 40.7 41.2 41.1 41.2 41.0 Sectoral allocation (% of total expenditure, excluding debt service) Social sector 28.9 28.8 33.7 34.3 35.2 36.1 36.9 37.8 38.6 36.5 Education 23.2 20.3 23.8 24.0 24.4 24.9 25.3 25.7 26.1 25.1 Health and social development 5.7 8.5 9.9 10.3 10.7 11.2 11.6 12.1 12.5 11.4 Productive infrastructure 6.0 15.8 12.8 13.1 13.1 13.1 13.2 13.2 13.2 13.1 Mining and energy 1.0 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 Transport and telecommunications 0.4 0.4 2.9 3.0 3.0 3.0 3.1 3.0 3.1 3.0 Water, environment, and urbanism 0.3 0.5 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.6 Public works and equipment 4.3 12.7 7.0 7.2 7.2 7.2 7.3 7.3 7.3 7.2 Productive sectors 6.4 5.3 8.8 9.6 10.1 10.5 11.1 11.6 12.1 10.8 Agriculture 5.0 4.1 8.3 9.0 9.5 10.0 10.6 11.1 11.6 10.3 Industry and commerce 1.4 1.2 0.5 0.6 0.6 0.5 0.6 0.6 0.6 0.6 Other sectors 58.7 50.1 44.7 43.0 41.6 40.3 38.8 37.4 36.0 39.5 Sources: World Bank staff and Burundi authorities.
4.16 This increase is justified by the reallocation of nonproductive expenditure toward socioeconomic priority expenditures following the consolidation of peace, the reduction of military expenditure, and continued efforts to rationalize the national budget. The budget share of productive infrastructure increased markedly in recent years, from 6.0 percent in 2007 to 15.8 percent in 2008; the increase is mostly explained by growth in spending on public works and equipment (construction) projects. The share of productive infrastructure is projected to be steady at 13 percent of GDP over the 2010–15 projection period. The budget share of productive sectors (mostly agriculture), which decreased between 2007 and 2008, is estimated to increase to 8.8 percent in 2009, reflecting the authorities’ priorities for rural development. Furthermore, that priority is reflected in the projected budget share of
88 these sectors, which is expected to rise by 2011 above the critical 10 percent level recommended by the New Partnership for Africa’s Development (NEPAD).65
Table 4.5: Alternative Scenario—Actual, Estimated, and Projected Government Finance, 2002–15 (Percent of GDP)
Actual Estimated Projected Average 2002–08 2009 2010 2011 2012 2013 2014 2015 2010–15 Indicator (average) Domestic revenue 19.6 19.0 19.1 19.2 19.4 19.6 19.8 20.1 19.5 Tax revenue 17.8 17.1 17.1 17.2 17.3 17.4 17.6 17.8 17.4 Nontax revenue 1.9 1.9 2.0 2.0 2.1 2.1 2.2 2.3 2.1 Total expenditure and net lending 36.9 50.4 50.2 46.7 47.1 45.3 40.1 36.6 44.3 Recurrent expenditure 23.1 25.1 25.5 25.3 24.3 22.5 20.4 18.1 22.7 Salaries 9.3 12.4 12.2 12.0 11.4 10.6 9.7 8.8 10.8 Goods and services 7.0 5.5 6.4 6.6 6.9 6.5 6.0 5.4 6.3 Transfers and subsidies 3.7 5.7 6.0 5.9 5.4 4.9 4.3 3.7 5.1 Interest payments 3.1 1.5 0.9 0.8 0.6 0.4 0.3 0.2 0.6 Other expenditure (incl. exceptional) 1.8 10.0 8.3 3.3 2.8 2.3 0.6 1.2 3.1 Capital expenditure 12.2 15.4 16.5 18.1 19.9 20.5 19.1 17.3 18.6 Domestically financed 3.2 3.6 3.0 4.5 8.2 10.2 10.5 10.3 7.8 Foreign financed 9.0 11.8 13.5 13.6 11.8 10.3 8.6 7.0 10.8 Net lending –0.2 –0.1 –0.1 –0.1 0.0 0.0 0.0 0.0 0.0 Primary balance –3.4 –12.5 –9.5 –9.7 –12.5 –12.6 –10.8 –8.1 –10.5 Overall balance (commitment basis) Including grants –3.1 58.8 –8.4 –6.9 –5.0 –4.2 –2.1 –1.7 –4.7 Excluding grants –17.2 –31.4 –31.1 –27.4 –27.7 –25.7 –20.2 –16.5 –24.8 Sources: International Monetary Fund and World Bank staff. 4.17 Under the alternative scenario, capital expenditure will be substantially higher, but accelerated reforms will prevent an increase in the overall balance. Increased spending, mainly on infrastructure, will bring capital expenditure to an average of 18.6 percent of GDP in 2010–15 (table 4.5), compared with 16.6 percent of GDP in the baseline scenario. Lower recurrent expenditure resulting from accelerated reform efforts in public finance management and the civil service, however, will offset most of the increase in capital spending. The resulting overall balance is therefore projected to be similar to that under the baseline scenario. The domestic revenue pattern under the baseline scenario is maintained, because it already assumes substantial efforts from tax collection authorities.
Table 4.6: Alternative Scenario—Medium Term Expenditure Framework, 2007–15
Actual Estimated Projected Average Expenditure 2007 2008 2009 2010 2011 2012 2013 2014 2015 2010–15 (percent of total expenditure) Recurrent expenditure (excl. debt service) 63.2 60.6 60.4 59.9 57.5 54.3 51.8 51.2 50.8 54.3 Capital expenditure 36.8 39.4 39.6 40.1 42.5 45.7 48.2 48.8 49.2 45.7 Sectoral allocation (% of total expenditure, excluding debt service) Social sector 28.9 28.8 33.7 33.1 38.0 37.2 37.2 37.9 38.3 36.9 Education 23.2 20.3 23.8 23.1 26.6 25.8 25.6 25.8 26.0 25.5 Health and social development 5.7 8.5 9.9 10.0 11.3 11.4 11.6 12.0 12.4 11.5 Productive infrastructure 6.0 15.8 12.8 21.9 19.9 22.5 23.7 23.4 23.6 22.5 Mining and energy 1.0 2.2 2.2 4.8 1.9 1.9 2.0 2.1 2.1 2.5 Transport and telecommunications 0.4 0.4 2.9 0.8 2.5 2.6 2.8 2.8 2.8 2.4 Water, environment, and urbanism 0.3 0.5 0.6 4.0 0.6 0.6 0.6 0.6 0.6 1.2 Public works and equipment 4.3 12.7 7.0 12.3 14.9 17.4 18.3 17.9 18.1 16.5 Productive sectors 6.4 5.3 8.8 5.3 7.0 7.6 8.3 9.0 9.4 7.7 Agriculture 5.0 4.1 8.3 4.9 6.5 7.1 7.8 8.4 8.9 7.3 Industry and commerce 1.4 1.2 0.5 0.4 0.5 0.5 0.5 0.6 0.5 0.5 Other sectors 58.7 50.1 44.7 39.8 35.1 32.6 30.9 29.7 28.7 32.8 Sources: World Bank staff and Burundi authorities.
65 According to the 2003 Maputo Declaration, the Comprehensive Africa Agricultural Development Program from NEPAD is based on two major principles: (1) pursuit of a 6 percent average annual growth rate at the national level in the agriculture sector, and (2) allocation of 10 percent of the national budget to agriculture.
89 4.18 The budget share of productive infrastructure will be significantly higher than in the baseline scenario, without compromising spending on social sectors. Implementing large infrastructure projects (as recommended in the previous chapters) will yield an average budget share of productive infrastructure of 22.5 percent for 2010–15 (table 4.6), compared with an average of 13.1 percent under the baseline scenario. Spending on social sectors, however, will remain similar, as assumed under the baseline scenario. The share of other sectors, which comprises nonpriority spending, is expected to decrease significantly.
A3. Prospects for Poverty Reduction and MDG Attainment
4.19 Given the assumptions of the Figure 4.1: Selected MDG Trends under the Baseline Scenario, 2006–15 baseline scenario, progress will be made with on social indicators, but it is 7% 120% not likely that the MDGs will be 6% 100% 5% achieved. The macroeconomic 80% framework used for the projections 4% 60% (Burundi Macroeconomic Model) was 3% 40% linked to poverty through growth-poverty 2% elasticities. It is assumed that GDP per 1% 20% capita, inequality/variance of per capita 0% 0% income, social/basic infrastructure, and net enrollment rate are the key factors 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 affecting the incidence of poverty. Using Poverty incidence (RHS) Net primary enrollment (RHS) GDP growth (LHS) a growth-poverty elasticity of –1.25, the Sources: Burundi authorities and World Bank staff. baseline macroeconomic framework Note: LHS = left hand side; RHS = right hand side. indicates that the incidence of poverty would be about 38 percent in 2015—down from 67 percent in 2006, but short of the MDG target of 18 percent (see figure 4.1). The same applies for primary-school enrollment and child mortality. Progress is expected to be made, but not enough to meet the MDGs. 4.20 Under the accelerated reforms scenario that includes increased spending on infrastructure, Figure 4.2: Selected MDG Trends under the Alternative Scenario, 2006–15 growth will be notably higher and the incidence of poverty 14% 140% substantially lower (see figure 4.2). 12% 120% Economic growth will average 9 10% 100% percent over 2010–15, and its broader distribution would reduce the poverty 8% 80% rate by more than 11 percentage 6% 60% points (to 27 percent in 2015, 4% 40% compared with 38 percent in the 2% 20% baseline scenario). This is still 0% 0% insufficient, however, to attain the 2006200720082009201020112012201320142015 MDG target by 2015. It is evident that Poverty incidence (RHS) Net primary enrollment (RHS) successfully implementing structural GDP growth (LHS) reforms (such as privatization and Sources: Burundi authorities and World Bank staff. regulatory reforms) that are required Note: LHS = left hand side; RHS = right hand side. to improve the overall business climate, along with public finance management reforms and the necessary investments in infrastructure, would result in accelerated growth in human capital, infrastructure, and private investment. Consequently, there are clear indications that a higher pace of poverty reduction is well within reach and primarily dependent on the authorities’ commitment, willingness, and their ability to forcefully deepen reforms and swiftly implement policies.
90 B. HOW CAN AID CONTRIBUTE TO INCREASED GROWTH? 4.21 Burundi has not experienced the expected post-conflict growth spike seen in other post-conflict countries (see section A3, chapter 1). This is despite the high levels of aid Burundi has been receiving. The objective of this section, therefore, is to understand why the increased aid has not yet noticeably enhanced growth.
B1. Allocation of Increased Aid and Impact on Growth 4.22 The effect of aid on growth has been widely discussed in the literature, without a consensus. A large body of literature on aid and growth typically has examined the impact of aggregate aid on growth, often using cross-country regressions, and it has found mixed results. Realistically, aid is recipient specific and is given for a multitude of reasons, not solely to promote growth. Often aid provides humanitarian relief following war or natural disasters, or it is given to achieve donors’ political objectives, with growth as a secondary goal. Other times, aid is conditioned on the specific social objectives of the donors, such as promoting education, health, or gender equality. These political or social objectives may or may not fully match the development goals of recipient countries, thus affecting their ability to successfully implement their growth and development strategies. 4.23 It is important to identify the objectives of aid in the analysis of its impact on growth. Clemens, Radelet, and Bhavani (2004) identify three categories of aid and their potential impact on growth. In the first category, aid is destined for immediate consumption in the aftermath of a disaster or war to stabilize the country and facilitate peace building, thus exhibiting little or even negative correlation with growth. There is, however, an important long-term effect on growth because peace and stability have to be established to enable the recovery of the country. A second category of aid provides resources for infrastructure and production sectors, which can affect growth in the short term. A third category of aid focuses resources in sectors—such as health and education—that have long-term impact on growth, but no visible short- or medium-term correlation. Using regression analysis over four years, the authors find a strong, positive, and causal relationship between “short-impact” aid and growth that is two to three times larger than that found in studies using aggregate financial aid. 4.24 The absorptive capacity of aid recipients affects the economic returns to aid and its impact on growth. McGillivray and Feeny (2008) study the well-established U-shaped relationship between aid and growth as it relates to fragile countries’ absorptive capacity. They find that many of the highly fragile states (from a per capita income perspective) are over-aided because they can efficiently absorb only approximately one third of the amount of aid that other countries can absorb. 4.25 The absorptive capacity for aid is systematically different in post-conflict countries. Collier and Hoeffler (2004) find that there is no significant difference in absorptive capacity and no bounce-back of growth in the immediate aftermath of peacemaking (that is, the first three years). After that initial period, absorptive capacity approximately doubles. Aid is especially effective and critical for growth after peace has been fully established, but its effectiveness diminishes about seven years following the end of the conflict. 4.26 Budgetary practices and the low predictability of aid disbursement further affect recipients’ capacity to absorb aid and therefore affect the growth impact of aid.66 Many least-developed countries continue to finance recurrent expenditures with domestic resources while financing investment expenditures through foreign aid. The combination of budgeting investment expenditures based on aid receipts and low predictability of aid affects the composition and effectiveness of spending, and can further aggravate the bias toward government consumption. In fact, as Celasun and Walliser (2008) show, governments tend to cut investment spending in times of aid shortfalls and expand government consumption if additional aid becomes available unexpectedly. This is especially true in low-income
66 The predictability of aid flows is defined as the difference between commitments and actual disbursements in a given year.
91 countries that rely heavily on external assistance to fund their budget and have only limited access to international financial markets to fund budget shortfalls.
Allocation of Aid to Burundi 4.27 The allocation of aid is disaggregated according to its impact on growth, following the approach of Clemens, Radelet, and Bhavani (2004), modified to capture Burundi-specific characteristics.67 The three considered categories are (1) emergency and humanitarian aid, (2) aid with a short-term impact on growth, and (3) aid with a long-term impact on growth. Budget support and debt relief are allocated similar to the composition of the budget. 4.28 The pattern of aid allocation between 2001 and 2007 shows that Figure 4.3: Allocation of Aid by Impact Category, 2001–07 Gross disbursement, current US$ millions there is not necessarily a contradiction between increased aid 160 Humanitarian aid flows and low growth performance 140 if a lagged impact of aid is taken into Short run impact account (figure 4.3). The highest share 120 Long run impact of aid—about 41 percent—has been 100 allocated to emergency and humanitarian aid. In Burundi, it 80 consists mainly of emergency/distress 60 relief and emergency food aid. This type of aid is not directly growth US$ millions (current) 40 enhancing, and might even coincide 20 with a negative growth rate because it is disbursed in times of crisis. It is, 0 however, a necessary condition for 2001 2002 2003 2004 2005 2006 2007 establishing peace, and it has a long- Source: Authors’ calculation based on OECD–DAC data. term stabilizing effect. The Note: Data for 2001 reflects only commitments because no information expenditure pattern observed in on disbursements is available. Burundi is typical for a country getting out of conflict: the humanitarian aid remains high immediately after the end of conflict when a fragile situation prevails. Emergency and humanitarian aid has been declining since 2005, however, with improving security in Burundi. 4.29 Aid allocated to sectors that contribute to growth in the long run is about 37 percent of total aid, and it is the second-largest and increasing category. This is aid allocated fully or partially to education; health; population programs; water supply and sanitation; multisectoral projects; and to planning, development, and research in other areas. Spending in these areas is considered to build human capital and improve service delivery—efforts that are expected to increase the productivity of labor in Burundi in the long run. 4.30 Aid allocation to short-run growth-enhancing sectors68 has been small. Between 2001 and 2007, only about 24 percent of total aid went into these sectors. If disbursements for peace building are
67 This approach is possible by using data from the Creditor Reporting System of the Development Assistance Committee (DAC) of the Organisation for Economic Co-operation and Development (OECD), providing a breakdown of aid disbursement by 194 purpose codes. Each purpose code has been assigned a category that relates to its impact on aid. Data are available on commitments starting in 1995, but the low coverage ratio of the data enables the analysis of disbursement data only from 2002 onward. To ensure consistency with the previous section, however, data for 2001 is included on the commitment basis. This also enables the analysis of the entire period following the official end of the conflict after the Arusha Accords in 2000. 68 These sectors include transport and storage, energy, the financial sector, agriculture, manufacturing, mining, and peace building.
92 excluded, however, the allocation to sectors with a short-run impact is reduced to merely US$290 million or 18 percent of total aid in that period. Considering these results, it is not surprising that the increased amounts of aid have not led to a boost in economic growth rates. 4.31 The effect of the allocation of external resources is compounded by the limited amount of the domestically financed budget that has been spent on infrastructure and production-oriented sectors. This makes a significant positive effect on growth unlikely in the short to medium term. Spending on priority sectors (education, health, infrastructure, and agriculture) has improved, but it is still considered low because domestic resources are relatively limited.
Predictability of Aid 4.32 In the case of Burundi, the low predictability of aid likely has aggravated the spending pattern of the government, exacerbating the bias toward government consumption and away from investment in production-oriented sectors and human capital. This has not Figure 4.4: Disbursement Ratio and Commitment Disbursement Gap, been conducive to promoting growth in the 2001–07 medium term, especially because the 120 300 country already spends a very limited Disbursement ratio (LHS) amount of its domestically financed budget 100 250 on investment. In Burundi, the average aid 80 200 disbursement ratio (that is, the disbursed amount divided by the committed amount) 60 150
was 68 percent between 2002 and 2007, percent 40 100
but varied considerably from one year to US$ millions Commitment another. The average difference between 20 disbursement 50 commitments and disbursements was gap (RHS) US$133 million, peaking at US$286 0 0 million in 2006 when only just over 50 2002 2003 2004 2005 2006 2007 percent of the committed amount was Source: Authors’ calculation based on OECD–DAC data. disbursed. Figure 4.4 shows the large Note: LHS = left hand side; RHS = right hand side. volatility in the disbursement ratio. 4.33 In the next few years, the issues of low predictability and volatility of aid will be especially important, in light of the financial crisis. Aid commitments and disbursements may decline as donors address domestic financial issues. The situation is particularly critical because Burundi is now entering the true post-conflict reconstruction phase. According to Collier and Hoeffler (2004), this is the period when aid is especially effective and crucial for growth. According to their findings, this aid effectiveness diminishes after about seven years following the end of the conflict. It is therefore important that the amount of aid provided by the donor community remains at the current high level for the foreseeable future.
B2. Impact of Increased Aid on Competitiveness 4.34 The fact that Burundi has received increased aid flows (which it has mostly spent and absorbed) raises concerns about currency appreciation and its impact on the competitiveness of the economy and the potential for some Dutch disease effect.69 In fact, the real effective exchange rate appreciated slightly between 2005 and 2008, when the aid surge was observed: by 1.5 percent annually or
69 In the case of Dutch disease, prices of non-tradables increase more than those of tradables, leading to an appreciation of the real effective exchange rate and ultimately to the contraction of the tradables sector.
93 5.2 percent in total.70 The slight deterioration of the terms of trade during the same time period is another indicator for a possible loss of competitiveness. Figure 4.5: Exchange Rate and Price Developments, 2001–08 4.35 This overall slight loss of 120 30 competitiveness, however, has been CPI (12 month percent change) caused primarily by higher inflation REER (2000=100) 110 (RHS) 25 (LHS) compared with that of the trading 100 20 partners, not by appreciation of the nominal exchange rate.71 The 90 15 nominal effective exchange rate 80 10 depreciated by an average of 5.1 percent a year or 20.1 percent overall
Index 70 5
percent between 2005 and 2008. The higher
60 0 inflation, therefore, has more than offset the depreciation of the nominal 50 5 NEER (2000=100) exchange rate. The development of (LHS) these key indicators is shown in figure 40 10 4.5. The high inflation rate has been caused partly by natural events and Jan 08 Jan 07 Jan 06 Jan 05 Jan 03 Jan 04 Jan 02 Jan 01 Sep 07 Sep 08 Sep 06 Sep 05 Sep 04 Sep 02 Sep 03 Sep 01 May 08 May 07 May 05 May 06 May 04 May 03 May 01 May 02 developments in the commodity Source: Authors’ calculation based on International Monetary Fund data. markets that are exogenous for the Note: CPI = consumer price index; LHS = left hand side; RHS = right hand side government of Burundi, and cannot be NEER = nominal effective exchange rate; REER = real effective exchange rate. attributed to the increase in aid flows. For instance, part of the price increases in 2005 was caused by a drought affecting domestic agricultural production, whereas the high inflation in 2008 resulted from the global food and petroleum price crises. 4.36 The relative price development of tradables and nontradables confirms that the slight appreciation of the real effective exchange rate was not caused by the increase in aid. In the case of Dutch disease, prices of nontradables increase more than those of tradables, leading to an appreciation of the real effective exchange rate and, ultimately, to the contraction of the tradables sector. In Burundi, however, the price level of tradables increased by an average of 13 percent between 2005 and 2008, and the price level of nontradables increased by only 11 percent, on average.72 4.37 In addition, goods exports, dominated by coffee and tea, do not seem to have suffered— probably because of their dependence on nonprice factors (such as infrastructure and supply constraints) rather than prices and exchange rates (see also Foster and Killick [2006]). In fact, Burundi’s export performance improved from an average of 6.1 percent of GDP during 2000–04 to 6.4 percent of GDP during 2005–08, indicating that the appreciation of the real effective exchange rate has not been detrimental to exports.73 Dutch disease does not seem to be a significant problem in Burundi so far. This is confirmed by the International Monetary Fund’s assessment of Burundi’s exchange rate, which concluded that the real exchange rate has been broadly in line with economic fundamentals (IMF 2008). Despite the absence of an indication of a negative effect on competitiveness via the real effective exchange rate to this point, the possibility of a Dutch disease effect remains a risk.
70 For a more detailed discussion of absorption and spending of aid in Burundi, see Nielsen and Madani (2010). 71 The real effective exchange rate is calculated as the product of the nominal effective (that is, trade-weighted) exchange rate and the inflation differential between Burundi and its trade partners. 72 Average year-on-year inflation is based on monthly data; authors’ calculation is based on International Monetary Fund data. 73 The share of other, nontraditional exports remained fairly constant.
94 Box 4.1: The Common Performance Assessment Framework in Rwanda The Common Performance Assessment Framework (CPAF) consists of a set of indicators and policy actions selected from the Economic Development and Poverty Reduction Strategy (EDPRS) Results and Policy Matrix and the Joint Governance Assessment. It forms the basis for budget support donors’ joint assessment of the government of Rwanda’s performance in implementing the EDPRS. The CPAF is complemented by the Donor Performance Assessment Framework (DPAF), which consists of a set of indicators and actions drawn from international and national agreements relating to aid and its effective delivery, and forms the basis for government’s assessment of donor performance. The assessments would consider all partners’ performance (including donors giving nonbudget support) at both the aggregate and individual levels. The Process for Elaboration of the CPAF The process of developing the CPAF involved extensive consultations and drew from the work of the existing 15 sector working groups. Each group is chaired by a lead government institution and cochaired by a lead donor. Initial consultations involved the development of sector policy and results matrixes, with technical support and financing from the U.K. Department for International Development (DfID), the World Bank, and other donors. Indicators for the CPAF were agreed on the basis of consultations between government and budget support donors, and were limited to a list of 18 strategic outcomes and 47 indicators. To emphasize the commitment of all budget support donors and the government, a Memorandum of Understanding was signed. In that document, all the parties agreed to use the new reporting mechanisms (CPAF and DPAF) as part of a mutual accountability and harmonization process aligning with the Paris Declaration. Using the CPAF to Report on and Monitor EDPRS Achievements Reporting of performance against individual CPAF indicators is led primarily at the sector level, in advance of the Joint Budget Support Review (JBSR). Sector performance reports are prepared in time for further discussion at the JBSR. The dialogue in the JBSR is based on the joint assessment of progress in EDPRS implementation, including governance indicators and public finance management reform, using the CPAF as a single framework. For additional governance topics not covered in the CPAF, an assessment is to be reviewed on the basis of the Joint Governance Assessment annual review. A forward-looking joint review will be held prior to the Cabinet submission of a draft Finance Law or budget for the following fiscal year. This review enables signatories to discuss proposed budget allocations, forthcoming activities in relation to public financial management reforms, and the revision of the contents of the CPAF. This joint review will be held in March of each budget calendar year. A backward-looking review of performance will be undertaken jointly within the four months that follow the end of the government fiscal year in question. During this review, signatories will discuss the progress and challenges encountered in the implementation of the EDPRS in the previous year, reviewing budget execution and audited government accounts of the year before and value for money. Process for Revising the CPAF Policy actions and targets of the CPAF are revised on an annual basis during the forward-looking JBSR. Under exceptional circumstances, policy actions and indicators may be changed or adjusted as long as they remain consistent with the EDPRS and with ongoing government policies and programs. The revised CPAF is agreed by all signatories and takes the form of an annex to the minutes of the JBSR. Joint reviews provide a forum for dialogue concerning the performance of both the government and its budget support partners, and draw from the DPAF. It is undertaken following the same annual cycle as the assessment of progress in implementing the EDPRS based on the CPAF. Source: Adapted from World Bank (2009b).
95 C. CONCLUSION 4.38 Although the end of Burundi’s aid dependency is not foreseeable in the near future, improved allocation and efficiency of spending can enhance the impact of aid on growth in the short to medium term. As a result of the conflict in Burundi, security and military spending has been high. As the security situation in the country stabilizes, those funds can be reallocated to production- oriented sectors or social spending. An increase in the fiscal stance by improving revenue mobilization as well as public finance management will also lead to more available resources and more efficient spending. 4.39 A scenario assuming accelerated reforms and increased spending on infrastructure demonstrates that higher GDP growth and increased poverty alleviation are within reach. If the government implements the priority measures identified in the previous chapters—particularly regarding the management of public resources—the improvement of the business environment to foster private sector development and contribute to the economy, as well as the development of the necessary infrastructure, the economic and social prospects can be improved substantially. 4.40 Donor coordination could be strengthened. Donor coordination is important for a number of reasons. First, coordination is needed to ensure that the disbursement of donor funds is in line with the government’s objectives set out in the PRSP and in Vision 2020. Second, it is needed to take capacity constraints into account and avoid the overburdening of authorities with different administrative processes. Third, improved donor coordination can ensure consistency among the programs of different donors. In that regard, a Common Performance Assessment Framework (CPAF) is advisable to align donors’ objectives with those of the government and to agree on a set of measures to be taken. For an example of how the CPAF has been developed and is used in Rwanda, see box 4.1) 4.41 The small amount of domestically financed investment spending has done little to promote economic growth in the past years. This bias toward recurrent spending, however, has been aggravated by the low predictability of aid, which is not conducive to a growth-enhancing composition of public expenditure. Especially in the current economic crisis, firm commitments and timely disbursements by donors are very important to ensure the allocation of funds in growth-enhancing areas.
96 APPENDIX
97 APPENDIX 1: ACTION PLAN
Table A.1: Action Plan Priority Recommendations for the Time Period 2010–15 Area Recommendation Time frame Responsible agency Approximate Cost as share of total (2010–15) cost (US$ expenditure projected million) for 2010 15 1. Closing the infrastructure gap Electricity Develop Burundi’s domestic hydroelectric potential Short to MEN/REGIDESO 80 1.3% (Kaganuzi, Mpanda, Kabu 16) medium term Provide 110 kV lines with necessary substations to all Short to MEN/REGIDESO 60 1.0% provincial capitals medium term Finalize the financial restructuring of REGIDESO Short term MEN/MF tbd Transport Rehabilitate, upgrade, and maintain national highways Short to MPW/MTR/OdR 600 9.8% medium term Rehabilitate provincial and community road networks Short to MPW/MTR/OdR 13 0.2% medium term Build human and institutional capacity in government Short to MTR/OdR 15 0.2% agencies that are responsible for regulation and medium term management of road transport activities Prepare a business and master plan for civil aviation Short term MTR 1 <0.1% 2. Maximizing the benefits of Burundi’s EAC membership Institutional Recruit competent staff in MEACA Short term MEACA 0.2 <0.1% reforms Make institutional arrangements (including capacity Short to MEACA 0.5 <0.1% building and other regional integration issues within the medium term new ministry) Implement the coordination, monitoring, and evaluation Short term MEACA 0.3 <0.1% mechanisms Implement communication/sensitization campaign Short term MEACA 0.5 <0.1% Economic Eliminate remaining NTBs (including equipment at Medium term MTI/CD 1 <0.1% reforms border posts, one stop window at the port, and so forth) Strengthen and operationalize the public private Short to MTI/MEACA 0.3 <0.1% dialogue framework medium term
98 Area Recommendation Time frame Responsible agency Approximate Cost as share of total (2010–15) cost (US$ expenditure projected million) for 2010 15 3. Increasing the contribution of the private sector to growth Financial Provide technical assistance to banks and MFIs to Short to Banks/BRB/develop 1 <0.1% sector develop products that meet the needs of small and medium term ment partners/ medium enterprises administration responsible for private sector Establish units within banks that specialize in lending to Short to Financial Included in small and medium enterprises, and units responsible for medium term institutions/BRB/ previous rural and agricultural customers government technical assistance Revise the regulatory framework limiting the Short to BRB 0.1 <0.1% development of MFIs; and remove the ban on MFIs medium term engaging in leasing and issuing mortgages, subject to prior approval from the central bank Improve capacity of the BRB to supervise MFIs Short to BRB 0.4 <0.1% medium term Regulatory Finalize and adopt the revised mining code Short term MEN 0.6 <0.1% environment Finalize the revision of the tax code, including the Short to MF/MTI 0.35 <0.1% introduction of the simplified impôt synthétique medium term (synthetic tax) Governance Promulgate the revised Privatization Law Short term MGGP tbd Adopt and implement the national policy of good Short to MGGP/MF 30 0.5% governance and its action plan medium term 4. Encouraging emerging growth sectors Food crops Strengthen input distribution systems (technical Short to MA 0.5 <0.1% assistance) medium term Rehabilitate, create, and strengthen local infrastructure Short to MA 10 0.2% for storage, conservation, transformation, and medium term commercialization of agricultural products Organize, structure, and professionalize producers Short to MA 3 <0.1% medium term
99 Area Recommendation Time frame Responsible agency Approximate Cost as share of total (2010–15) cost expenditure projected (US$ millions) for 2010 15 Coffee Strengthen the capacity of the regulatory agency (ARFIC) Short term MA 1.5 <0.1% Launch the second round of washing stations and dry Short term MA/MGGP 0.5 <0.1% mills privatizing Promote microcredit and matching grant programs to Short term MA/MF 30 0.5% help farmers finance inputs or replant trees Strengthen the capacity of the interprofessional Medium term MA 1.5 <0.1% association (INTERCAFE) Improve processing practices Medium term MA 0.5 <0.1% Diversification Develop an action plan to establish facilities to comply Medium term MTI 1.5 <0.1% with international sanitary and phytosanitary standards Improve the availability of cargo space to export fresh Short to MTI/MTR/MA 0.5 <0.1% produce medium term Mining Develop and implement a sectoral strategy that Short to MEN 0.15 <0.1% explicitly accounts for the characteristics and constraints medium term of the different types of mineral resources Become a candidate country of the EITI, be validated as Short to MEN 0.15 <0.1% a compliant country, and use available facilities for medium term assistance with negotiations Revise the institutional framework by creating a unit Short term MEN 0.3 <0.1% within the MEN to deal specifically with artisanal and small scale mining Develop a project to support the formalization of Short to MEN 1 <0.1% artisanal and small scale mining medium term Source: Authors’ compilation. Note: BRB = Banque de la République du Burundi; CD = Customs Directorate; EAC = East African Community; EITI = Extractive Industries Transparency Initiative; MA = Ministry of Agriculture; MEACA = Ministry of East African Community Affairs; MEN = Ministry of Mines and Energy; MF = Ministry of Finance; MFI = microfinance institution; MGGP = Ministry of Good Governance and Privatization, MPlan = Ministry of Planning; MPW = Ministry of Public Works; MTI = Ministry of Trade and Industry; MTR = Ministry of Transport and Telecommunication; NTB = nontariff barrier; OdR = Office des Routes; REGIDESO = Régie de Production et de Distribution d'Eau et d’Electricité; tbd = to be determined. Total expenditure projected for 2010 15 refers to the alternative scenario presented in section A, chapter 4, assuming a constant FBu/US$ exchange rate.
100 APPENDIX 2: EXPORT DIVERSIFICATION
Figure A.1: Product Space Map of Burundi
Cotton (other than linters), not Goat and kid Sheep and Ores and carded or combed skins, raw lamb skins with concentrates of Fish, fresh wool on, raw other nonferrous or chilled, base metals excluding fillets
Tea
Bulbs, tubers, and Leather of rhizomes of other hides flowering or and skins of foliage plants
Cut flowers and foliage
Coffee, whether or not roasted
Gold, nonmonetary
Bovine and equine hides (other than calf), raw
Bones, horns, ivory, hooves, claws, and so Plants, seeds, fruit forth used in perfumery and pharmacy
Classics Disappearances Emerging champions
Source: Authors’ estimates based on calculations by the Economic Policy and Debt Department; map generated using software by Hidalgo et al. (2007), available at http://www.chidalgo.com/productspace/data.htm.
101 Table A.2: Detailed Classification of Burundi’s Exports
Shares (% of total Product name Tech Exports (US$ thousands) exports) code PRODY 1985 89 2005 06 1985 89 2005 06 Classics: RCA 1985 89 = 1; RCA 2005 06 = 1 Coffee, whether or not roasted or freed of caffeine PP 1,936 96,524 51,274 71.02 80.41 Tea PP 1,655 5,186 4,480 3.83 8.52 Cotton (other than linters) PP 1,500 1,288 1,287 0.73 2.75 Gold, non monetary RB 5,716 27,171 1,273 18.67 0.81 Fish, fresh (live/dead) or chilled, excluding fillets PP 4,919 269 196 0.20 0.36 Goat and kid skins, raw (fresh, salted, dried, pickled) PP 1,217 962 126 0.70 0.14 Plants, seeds, fruit used in perfumery, pharmacy PP 3,622 303 92 0.24 0.14 Sheep and lamb skins with wool on, raw (fresh, salted) PP 4,956 222 21 0.15 0.04 Total n.a. 3,190 131,925 58,750 95.53 93.18 Disappearances: RCA 1985 89 = 1; RCA 2005 06 = 0 Bovine and equine hides (other than calf), raw PP 5,653 708 22 0.52 0.01 Bones, horns, ivory, hooves, claws, coral, shells, and so forth PP 4,419 832 0.52 Total n.a. 5,036 1,539 22 1.05 0.01 Emerging Champions: RCA 1985 89 = 0; RCA 2005 06 = 1 Ores and concentrates of other non ferrous base metals RB 1,982 282 384 0.08 0.70 Cut flowers and foliage HV PP 4,015 61 375 0.03 0.69 Bulbs, tubers and rhizomes of flowering or of foliage plants; cuttings, slips, live trees and other plants HV PP 9,062 342 102 0.25 0.17 Leather of other hides or skins LT 2,156 344 53 0.26 0.06 Total n.a. 4,304 1,030 915 0.62 1.62 Marginals: RCA 1985 89 = 0; RCA 2005 06 = 0; average export value > USD 25,000 Sawlogs and veneer logs, of coniferous species RB 8,841 498.12 0.555 Tobacco, not stripped LT 3,317 342.33 289.31 0.124 0.185 Other tools for use in the hand LT 10,548 1.64 216.80 0.001 0.482 Other outer garments of textile fabrics LT 5,408 4.79 177.88 0.002 0.264 Footwear PP 7,765 18.76 91.18 0.002 0.199 Under garments, knitted of cotton LT 4,975 12.21 64.13 0.003 0.087 Bananas, fresh or dried HV PP 5,183 8.77 58.05 0.002 0.065 Other outer garments and clothing, knitted LT 6,020 29.55 30.69 0.004 0.043 Tobacco, wholly or partly stripped RB 1,531 377.47 27.00 0.145 0.017 Wood of non coniferous species, sawn, planed, tongued and so forth LT 3,667 9.78 24.95 0.003 0.041 Tarpaulins, sails, awnings, sunblinds, tents for boats, and so forth LT 4,852 39.85 25.94 0.006 0.045 Sawlogs and veneer logs, of non coniferous species RB 2,287 335.79 18.29 0.095 0.020 Calf skins, raw (fresh, salted, dried, pickled/limed) RB 4,065 406.17 10.56 0.160 0.007 Works of art, collectors pieces and antiques PP 8,542 34.19 8.12 0.024 0.005 Wood of coniferous species, sawn, planed, tongued and so forth RB 11,578 15.23 6.00 0.008 0.004 Shirts, men's, of textile fabrics LT 4,936 66.01 5.28 0.018 0.003 Beans, peas, lentils and other leguminous vegetables HV PP 2,376 4.70 0.003 Other precious and semi precious stones MT 2,846 579.51 0.26 0.323 0.000 Trousers, breeches etc. of textile fabrics LT 4,789 0.14 0.000 Chemical products and preparations, n.e.s. RB 19,098 0.69 0.10 0.000 0.000 Other fresh or chilled vegetables HV PP 5,477 44.88 0.02 0.023 0.000 Animals, live, n.e.s., incl. zoo animals, dogs, cats and so forth RB 3,680 41.28 0.029 Fish, frozen (excluding fillets) PP 6,457 Crustaceans and molluscs, fresh, chilled, frozen and so forth PP 3,369 Fruit, fresh or dried, n.e.s. HV PP 5,187 64.23 0.034 Sheep and lamb skins without the wool PP 2,349 19.46 0.004 Sheep and lamb skin leather LT 2,526 70.22 0.044 Cotton yarn LT 4,262 Cotton fabrics, woven ,unbleached, not mercerized LT 4,128 Portland cement, ciment fondu, slag cement and the like RB 5,109 127.79 0.017 Containers, of glass, used for conveyance or packing LT 6,824 Nickel and nickel alloys, unwrought RB 10,397 4.43 0.001 Aluminum and aluminum alloys, unwrought RB 9,077 Zinc and zinc alloys, unwrought RB 8,334 Total n.a. 5,907 2,615 1,532 1.07 1.98 Source: COMTRADE, SITC 2–4 digits; calculation by authors and the Economic Policy and Debt Department of the World Bank. Note: = no exports; n.a. = not applicable; n.e.s. = not elsewhere specified; RCA = revealed comparative advantage.
102 Table A.3: Horticulture: Markets, Constraints, and Opportunities for Burundi Market Constraints Opportunities Developed economy Low tolerance for failure to compete on Massive market of millions of consumers markets cost, quality, volume, timing, and flexibility Large ethnic populations with higher demand for (particularly Longer travel distance; requires higher exotic goods EU) standards for quality, safety, and packaging Organic and environmentally friendly products in materials high demand More strict environmental regulations Capacity to customize production of small orders regarding packaging materials in EU (compared with other products) of specialties markets Retail industry organization highly focused on Commoditization of mainstream category management; prefer suppliers who horticulture products, such as mangoes, provide several or all products under the same papayas category (for example, tropicals or leafy greens) High quality standards difficult for Burundi rather than just one product to meet in the short term (for example, Direct access to supermarkets and major EurepGAP regulations) wholesalers Shrinking market windows for countries exporting the same products Emerging/regional Cost driven markets Relatively short travel distances within the country markets Feeder roads to trunk roads in Burundi favor transport of perishable products (surrounding limited to gather/consolidate product Higher tolerance for lower quality, less volume, countries) across different production areas longer timing, and less flexibility Trucking availability also limited, especially Rise of supermarkets in several neighboring reefer containers necessary to reach sizable countries (such as Kenya) represents an incipient markets 2–10 days away (such as Kampala, opportunity to establish long term supply Kigali, Mombasa, Nairobi) contracts Products may compete with local Incipient category management criteria production if no quality differentiation is attained Emerging markets, Increasingly demanding on cost High potential to establish category management the Middle East Sales programs must include supplying supply programs with supermarkets and the countries oriented several products under one category, hospitality business sector toward free trade rather than only one or two products Shorter flying time than to EU markets (Bahrain, Kuwait, Direct transport services lacking Food products face no quantitative restrictions on Oman, Qatar, and Difficult to negotiate pricing tariff or nontariff barriers (except vegetable oils) the United Arab High quality demanded These countries import 90 percent of their food Emirates) Buyers may not be trustworthy, requiring needs very close supervision and intensive One stop redistribution hub (Dubai) to other management of accounts Middle Eastern nations Difficult to find consistent wholesalers More tolerance on quality, food safety, and Neighboring countries (such as Kenya) packaging standards already have started to develop these Payments are upon arrival and acceptance of markets for temperate and tropical shipments products. Afghanistan, with USAID support, High demand for tropical and semitropical has started an aggressive program products exporting specialty vegetables and leafy Prices higher than in European markets greens. Source: Baghdadli, Harborne, and Rajadel 2008. Note: EU = European Union; USAID = U. S. Agency for International Development.
103 Table A.4: Recommendations for the Artisanal and Small Scale Mining Sector Area Recommendations Revision of the Create within MEN a unit within the DGGM that deals specifically with all aspects of artisanal and small scale institutional mining, including framework granting of exploitation permits; administrative and technical supervision; training; management supervision of environmental aspects of mining. Revision of the legal - All regulations should be guided by the mining code, implemented centrally by the DGGM. The and regulatory regulatory framework should be summarized and translated into Kirundi to be accessible to the artisanal framework and small scale miners. - The roles and responsibilities of the different stakeholders need to be clarified—that is, the government, concessionaries and other private operators, provincial and local authorities, mining associations and cooperatives, and the miners. - Given the currently ongoing revision of the mining code, the following critical points should be considered: Revision and/or precision of the definition of artisanal and small scale mining activities, the supervision of artisanal miners, the allocation of artisanal exploitation titles, and the transfer and lease of titles. Mining title and the responsibilities of the titleholder should contain regulations regarding the relationship between the titleholder (concessionary) and the landowner. Fair distribution of revenue should be guaranteed for all stakeholders; new mining legislation should recommend a strategy to develop and promote artisanal and small scale mining that is integrated with other development strategies in the country to ensure the contribution of mining to poverty reduction, the promotion of the role of women in the sector, and the fight against child labor. Environmental aspects should be integrated in the revised legislation. Role of communities should be strengthened. Implementation and The sectoral strategy of the MEN needs to account explicitly for the characteristics and constraints of the rationalization of different types of mineral resources found in Burundi, by considering current knowledge of the subsoil, the strategies to exploit time horizon and realistic forecast of the exploitation of the deposits, the fragility of the economy, and the the mineral lack of opportunities for employment creation in rural areas. potential Formalization of Proposed project to support the formalization of artisanal and small scale mining should include the artisanal and small following activities: scale mining Develop an integrated management plan for artisanal and small scale mining, accounting for technical, organizational, legal, institutional, social, and environmental aspects. Within the DGGM, create and support a unit that deals specifically with the promotion of artisanal and small scale mining and the protection of the environment. Strengthen the capacities of the institutions relating to artisanal and small scale mining. Begin technology transfer with two pilot projects in the provinces of Cibitoke (gold) and Kayanza (coltan), Kirundo (wolfram), and Muyinga (gold and wolfram) through private operators with support of the DGGM. Establish and/or encourage the establishment of organizational structures within the mining community. Integrate ex combatants as well as unemployed youth. Furthermore, the following conditions need to be fulfilled to formalize the sector: improved access to finance; improved access to mining technology and equipment improved investment climate for local and foreign investors; and an institutional framework that manages the sector in a transparent and efficient manner. Source: Midende 2009. Note: DGGM = Directorate General for Geology and Mining; MEN = Ministry of Mines and Energy.
104 APPENDIX 3: REGIONAL INTEGRATION
Table A.5: Burundi’s Post Independence Experience with Regional Integration Agreements Agreement Members Objectives Instruments Results CEPGL: Economic Burundi, Democratic Promote Reducing barriers and Little progress in Community of the Great Republic of Congo, cooperation and improving mobility of liberalizing trade and Lakes, created in 1976 Rwanda economic factors of production, factor mobility: some development joint industrial projects industrial projects through a customs (agriculture, energy, union and sectoral health, and construction) cooperation OBK: Organization for Burundi, Rwanda, Promote projects Joint industrial projects — Development and Tanzania, Uganda for development of Management of the the Kagera River Kagera River Basin, Basin created in 1979 CEEAC: Economic Angola, Burundi, Economic Study of methods of — Community of Central Cameroon, Republic development recovery process African States, of Congo, Gabon, through a customs established in 1983 Equatorial Guinea, union and sectoral Central African cooperation Republic, Democratic Republic of Congo, Rwanda, São Tomé and Principe, Chad COMESA: Common Burundi, Democratic Economic and social Remove barriers to Setting up of FTA (13 of Market of Eastern and Republic of Congo, development trade; PTA Bank for 19 members); trade Southern Africa, created Comoros, Djibouti, through a common Trade and Development; facilitation by in 1994 (formerly PTA, Egypt, Eritrea, market, monetary Program ITF/RIFF; harmonizing customs created in 1981) Ethiopia, Kenya, and financial clearinghouse; Center for procedures and Madagascar, Malawi, cooperation, and Commercial Arbitration, documents across Mauritius, Namibia, coordination of Court of Justice, multiple channels Uganda, Rwanda, macroeconomic Interregional (SYDONIA, DTDR, and so Seychelles, Sudan, policies Coordination Committee, forth); free movement of Swaziland, Zambia, and Insurance and Re people (even partial free Zimbabwe Insurance Regional movement) Company Source: Hugon 2003. Note: DTDR = Déclaration de transit douanier routier; FTA = free trade area; PTA = Preferential Trade Area; SYDONIA = System Douanier Automatisé.
105 Box A.1: Potential Benefits, Costs, and Guiding Principles of Regional Integration Potential Benefits of Regional Integration Increased returns and increased competition—Enlarging markets through integration of small economies (1) promote possibilities to achieve economies of scale and improve efficiencies; and (2) increase competition, leading to lower prices and expanded supply. Trade and location effects—Preferential reductions in tariffs within regional agreements can induce shifts in both demand and supply. Net effects on national income depend on costs of alternative supply and trade policies toward nonmember countries. Investments—Regional cooperation and established agreements can attract more foreign direct investment by (1) enlarging markets (particularly for “lumpy” investments viable only above a certain size) and (2) reducing distortion and lowering marginal cost of production. Coordination and collective bargaining power—Regional integration agreements may enable countries to coordinate negotiating positions in international forums, thus raising visibility and possibly strengthening bargaining power. Management of shared natural resources—Many watersheds, mineral deposits, fisheries, and sensitive natural environments are shared among countries. No country acting alone can ensure sustainable management, which depends on collaborative efforts. Management of “regional commons”—Effective action to combat migratory diseases (such as HIV/AIDS and malaria) and vulnerabilities arising from climate change depends on collaborative efforts among groups of countries. Policy lock in and commitment mechanism—Regional agreements can provide a “commitment mechanism” for countries’ domestic trade and other policy reforms, reducing the likelihood of policy reversals. This mechanism bears upon political as well as economic reforms. Insurance—Integration agreements provide members insurance against exogenous shocks (terms of trade shocks, conflict, protectionism in developed countries, and impacts from climate change). Security—Risks of conflict are perceived to be lowered through integration agreements, as a result of improved intraregional confidence and trust, common defense arrangements, and interdependence in key aspects of countries’ national development. Potential Costs of Regional Integration Trade diversion—Displacement of low cost products from nonmembers by higher cost products from countries within a regional integration arrangement has been a major problem with several regional integration agreements. Welfare gains are realized if trade creation dominates trade diversion, but this outcome cannot be ensured in advance. Revenue loss—Trade integration agreements reduce governments’ tariff revenues, both directly through tariff cuts and indirectly through shifts away from imports from nonmembers that are subject to tariffs. Net costs depend on how much new trade is generated from the integration agreement, but they can be high when trade facilitation is difficult. Indirect costs—These costs arise from the free movement of people and capital across national borders (capital flight and loss of skilled human resources, for example), and extra vigilance is required to prevent cross border crime. Guiding Principles for Effective Integration Open regionalism is critical. Regional and global integration are complements, and need to be pursued in parallel and with equal vigor. Attention must be paid to the neighborhood. Neighborhoods have important implications for the focus and sequencing of strategies toward regional integration, particularly for landlocked countries. Starting small is easier. The size of the group is important. The larger the number of participants, the more complex are the coordination challenges, making failure more likely. Stay focused and proceed incrementally. Experience shows that this is likely to give better results. Develop credible mechanisms to compensate losers. Participants often do not share benefits equally; mechanisms of compensation are critical if more ambitious integration is to take hold. Subsidiarity is important. The subsidiarity principle respects national sovereignty and capacity and avoids overloading scarce regional capacity. Variable geometry is necessary. Variable geometry is a nonuniform method of integration that enables subgroups to move faster than the whole group or to move to a deeper form. It involves heterogeneous strategies in design and implementation of regional arrangements. Sources: World Bank 2008c,d.
106 Table A.6: Matrix of Strategic Objectives and Indicative Outputs for the MEACA Strategic objectives Indicative outputs 1. To build up the organizational and Staff are prepared to contribute to the work of the ministry institutional structure of the ministry MEACA is organized in an optimal way to deliver its mandate so that it can successfully play its role Staff have the right skills to plan, prioritize, monitor, and follow up on all of promoting the integration of ministry activities Burundi in the EAC Staff have adequate equipment (computers, communications equipment, and so forth) The necessary information is available where and when required 2. To adopt effective coordination Strategic and effective structure for coordination is in place mechanisms for implementing EAC Procedures for coordination are in place affairs Communication mechanism is operating Focal points have the necessary resources to be effective (information, skills, mandate, resources, and so forth) Appropriate links with relevant government institutions are in place and roles and responsibilities are clear 3. To negotiate with EAC partners the Technical skills and information needed for negotiation are acquired process of Burundi’s full integration Logistics for cooperation are in place (transport, funds, information, delegated authority—all in a timely manner) MEACA is clear about current status of legislation compared with the acquis communautaire Progress against milestones is monitored and follow up action is taken as required There is assurance that negotiations are being carried out in an efficient and effective manner 4. To manage a system of Each stakeholder group has the required information communication involving all stakeholders 5. To coordinate the private sector and Legal and institutional frameworks are made available civil society so they can contribute to Private sector and civil society participate fully the success of the process of There is a competitiveness review of the key sectors of Burundi’s Burundi’s integration in the EAC economy 6. To ensure mechanisms and Information on current status and challenges facing integration is capacities are in place to monitor and available where and when required evaluate the integration process Appropriate action is taken on the basis of monitoring and evaluation data gathered Activities of Burundi and other EAC member states in regional and multilateral organizations are monitored Source: Strategy of the MEACA, draft, March 2009. Note: EAC = East African Community; MEACA = Ministry of East African Community Affairs
107 Table A.7: Policy Matrix for Advancing Regional Integration in Burundi Measure Timeline Responsible institution and technical/financial assistance A. Implementation and harmonization of fiscal instruments Finalize the tax and customs codes ST MF/WB Computerize the tax department ST MF/WB/DFID Introduce the VAT ST MF/PAGE/WB Adopt EAC fiscal year ST MF Implement the BRA ST/MT MF/DFID B. Implementation of private sector; financial/banking and payment instrumentsa Develop a modern payment system ST/MT BRB/IMF Develop other financial market and banking measures MT/LT BRB/IMF Strengthen the public private dialogue ST/MT MTI/CCIB/WB C. Implementation of trade instruments Adopt the CET ST MTI/CD/MF/WB Eliminate remaining tariffs ST/MT MTI Eliminate remaining NTBs MT MTI/CD Disseminate the VAT and the CET ST/MT MTI/WB Participate in developing and adopting a common trade policy MT MTI Implement the investment and export promotion agency MT MEACA D. Institutional arrangements Ensure a successful accession to the CU ST MEACA Recruit competent staff in MEACA ST MEACA Participate in the EAC missions ST/MT MEACA Provide training (English language skills) ST MEACA/DFID Implement the coordination, monitoring, and evaluation mechanisms ST MEACA Implement the communication strategy ST MEACA/DFID Strengthen regional integration skills ST/MT MEACA/DFID Implement other measures of the MEACA institutional matrix (Action Plan, MT MEACA/DFID see table A.6) E. Infrastructure Roads, railway, pipeline, port, airport MT/LT MTR Communications (submarine cable network) ST/MT/LT MTR Energy ST/MT/LT MEN F. Mining Exploitation of mineral resources MT/LT MEN Source: Authors’ compilation. Note: BRA = Burundi Revenue Authority; BRB = Banque de la République du Burundi; CCIB = Chamber of Commerce and Industry of Burundi; CD = Customs Directorate; CET = common external tariff; CU = Customs Union; DFID = Department for International Development; EAC = East African Community; IMF = International Monetary Fund; LT = long term; MEACA = Ministry of East African Community Affairs; MEN = Ministry of Mines and Energy; MF = Ministry of Finance; MT = medium term; MTI = Ministry of Trade and Industry; MTR = Ministry of Transport and Telecommunication; NTB = nontariff barrier; Page = Projet d’Appui à la Gestion Economique; ST = short term; VAT = value added tax; WB = World Bank. a. See also table A.13.
108 APPENDIX 4: INFRASTRUCTURE ACTION PLAN—TABLES AND MAPS
TableA.8: Basic Infrastructure Coverage for the East African Community, 2006
East African Community Other Country Groupings Sub Other low Indicator Burundi Kenya Rwanda Tanzania Uganda Average Saharan income Africa countries Roads Percent of roads paved 10 14 19 9 23 15 Paved road density 31 134 Road density for arable land 13 12 12 9 14 12 137 211 Total road density 48 11 57 9 36 12 Electric power Generation capacity 6 37 326 Electricity coverage 2 13 5 11 8 10 16 41 Electric power consumption 14 138 61 Communications Mainline density 4 8 2 4 4 5 16 78 Mobile density 25 201 33 146 67 126 175 76 Internet density 7 76 11 10 50 39 34 30 Water and sanitation Access to improved water 71 57 65 55 64 60 58 72 Access to improved sanitation 41 42 23 33 33 35 31 51 Source: As cited in AfDB (2009): World Bank 2008c; Africa Development Indicators, 2008/09; and www.infrastructureafrica.org. Note: — = not available. Road density is in kilometers of road per square kilometer of arable land; telephone density is in lines per 1,000 people; generation capacity is in megawatts per 1 million people; electricity, water, and sanitation coverage are in percentage of population.
Table A.9: Service Costs and Difficulties in the East African Community
Sub Saharan Indicator Burundi Kenya Rwanda Tanzania Uganda Africa Supply problems Water failure for firms receiving water (days per year) 12 85 105 Electrical power outages per month (number) 12 14 12 11 Number of days for electrical connection 24 51 18 44 33 Firms with access to private generator (%) 42 71 58 46 29 Fuel prices (US$ per liter) Diesel fuel 1.22 0.98 1.08 0.99 1.01 0.98 Gasoline 1.20 1.12 1.11 1.04 1.17 1.03 Communications costs (US$) Price basket for internet (per month) 52.00 79.20 29.40 93.60 95.80 42.10 Fixed line local call for 3 minutes in peak hours 0.07 0.11 0.08 0.16 0.28 0.14 Cellular local call for 3 minutes in peak hours 0.58 0.64 0.79 0.69 0.67 0.77 International call to US for 3 min. in peak hours 0.07 0.13 0.18 0.14 0.28 0.13 Source: As cited in AfDB (2009): World Bank 2008c, Africa Development Indicators, 2008/09. Note: — = not available.
109 Table A.10: Description of Various Scenarios Considered in the Report
Program or Project Included in Scenario Current strategy Infrastructure Action Plan Nickel Rail Scenario Description of scenario for national With public With private mining extension development investment investment project into Burundi continues component component
A Base Case Yes Yes Yes Yes No Core Infrastructure Action B Yes Yes Yes No No Plan implemented Private funding for Action C Yes Yes No No No Plan not available Only 50% of public funding Implement 50% D Yes No No No available for Action Plan of Action Plan Only 20% of public funding Implement 20% E Yes No No No available for Action Plan of Action Plan Implementation of Base F Yes Yes Yes Yes Yes Case with rail extension Source: AfDB 2009.
110 Figure A.2: Transport Corridors for the East African Community
Source: AfDB 2009.
111 Figure A.3: Railway Network of the East African Community
Source: AfDB 2009.
112 APPENDIX 5: INFRASTRUCTURE ACTION PLAN—MAIN COMPONENTS74
Main components of the infrastructure action plan The focus of the proposed infrastructure action plan is on power, transport and communications. Discussions with the Government of Burundi have confirmed the key objectives for the program.
Power sector The proposed program for the power sector has six key objectives: Through increased investments in domestic and regional generation capacity, ensure that the business community and households have access to reliable power supply 24 hours a day. Establish a national transmission and distribution grid by 2015, with all 15 of the provincial capitals linked to this grid which would supply electricity 24 hours a day at reasonable cost. With the grid in place, increase the electrification rate from the current two percent of households to 25 percent by 2020 and at least 40 percent by 2030. By 2020, 85 percent of urban households would have continuous access to the national distribution network, and by 2030, one-third of all rural households would be linked to the grid. Give priority to further development of domestic energy sources to avoid excessive dependence on imported supplies of electricity. At the present time, about 45 percent of total supply is imported. The design of the Action Program aims to keep this dependence at less than 50 percent and close to current levels until 2024, after which there would be a gradual increase in imported supplies of electricity. Improve demand management and reduce system losses. Ensure that the national electricity utility is built up into an effective and financially sound entity. To meet the projected demand for power, the required generation capacity for Burundi would be about 600 MW by 2030, compared with less than 40 MW at present. The domestic and regional power plants identified and included in the proposed action plan would be sufficient to meet the needs of the country until the mid 2020s. Provided that no further major mining or power intensive industrial projects are launched, the supply deficit would grow to about 1,240 GWh by 2030. The implication is that Burundi would need an additional 200 MW of capacity to meet domestic demand. The working assumption in the proposed program is that this shortfall is met by the import of additional electricity. The key policy question for the longer-term is whether to investigate other domestic hydro power sites within Burundi and develop these in order to keep dependence on imported power to prudent levels, or whether to allow increased dependence on imported power. If all the additional required capacity was domestic, the share of domestic power in total consumption would be 75 percent by 2030. A potential issue with these domestic sites is that the cost of the electricity produced may be significantly higher than that imported from Ethiopia via the EAPP grid.75 The ongoing operational and financial rehabilitation of the national power utility, REGIDESO, is central to its role as a major source of funding for the future power program. With continued prudent financial management the utility would develop into a major corporation by 2030, at which time it would have assets of about USD 1.6 billion and revenues of about USD 300 million a year (both at 2007 constant prices). The improved cash flow in the decade ahead would allow the utility to pass on benefits to consumers with an ongoing reduction in power tariffs from about 2016, which would lower the average tariff to less than nine U.S. cents a kWh. From about 2020 onwards, the utility would be able to fund the
74 This section draws from AfDB (2009). 75 The trade-off between the degree of self-sufficiency in power supply and the cost of power and its effects on the competitiveness of Burundi business is discussed in more detail in AfDB (2009).
113 bulk of the power development of the country from its own resources and from prudent access to commercial sources of debt financing.
Transport sector The key objectives of the proposed action plan for the transport sector are to lower the costs of transport for the entire economy and to improve access to local and international markets. The proposed program focuses on road transport and civil aviation, and also provides for further investigation of the possibility of a rail extension from Tanzania into Burundi. The main component of the transport program is the upgrade and expansion of the road network of the country over the next decade. The proposed program would rehabilitate and pave the entire 1,950 km of national roads by 2020, and for those national routes where traffic densities are high it would upgrade roads to enhanced standards that can accommodate the increased traffic. By 2030, the urban road network would be expanded from the current estimated 650 km of roads to about 1,650 km so that all 2.6 million urban dwellers at that time are within 500 meters of a road that can carry vehicular traffic. The other key component is a program that will improve provincial, community and local feeder roads in key agricultural areas to facilitate access to product markets at home and abroad, and to key inputs, such as fertilizer, required for production activities. In conjunction with the rehabilitation of the road network, the program would increase substantially budget allocations for routine maintenance of these facilities. The international airport in Bujumbura would be expanded and modernized. The intent is to ensure, within the next five years, full compliance with the ICAO-mandated standards and procedures for international passenger travel and for freight. This would be done through a program of staff training and investment in infrastructure. As a result of this program, the international airport would obtain an ICAO “Certification of Aerodromes.” These improvements would allow Burundi to attract major international airlines and air freight companies, thus opening up opportunities for development of tourism and air freight of high value export products to European and Middle East markets. The options for the extension of the Tanzanian rail network into Burundi have also been reviewed as part of this study. What emerges is that if the nickel deposits in the Musongati area of Burundi are developed for the export of nickel ore, and if the mine sites have access to the rail network, the four million tons of ore to be carried makes the investment in the rail extension quite attractive. The potential problem is that international investors interested in developing the site may prefer to refine the ore at the mine and ship the metal to export markets. In this case the volume of mine-related freight is substantially smaller – an estimated 50,000 Mt of nickel metal a year and a small quantity of cobalt. Based on the assessments made for the future volumes of international freight in and out of Burundi each year that are unrelated to the mine activities, it would appear that the rail extension is not economically viable unless it carries a large volume of mine freight. If one of the two options were to go ahead in the absence of the mine, it is likely that the rail operator would require substantial public subsidies. The program proposes further evaluation of the costs and benefits of a possible extension of the Tanzanian rail network into Burundi. The proposed program would lower the costs of transport for the entire economy and improve access to local and international markets. The benefits of lower transport costs under the program would be substantial; for example, at current costs for road freight of 13 U.S. cents per ton/km or more, it costs about USD 230 to move a ton of fertilizer from the ports in Kenya and Tanzania to Burundi. A reduction in transport costs to 8 U.S. cents a ton/km, the rate that prevails in these neighboring countries, would lower the freight cost of the fertilizer by almost USD 100 a ton. These types of cost reductions can have a significant impact of the profitability of farming and other business activities.
114 Communications network The proposed communications program aims to improve substantially access to international communications network and to lay the foundations for a national communications grid that will provide communities and business through Burundi with low cost voice and data communications. East Africa is the only heavily populated region of the world that does not have access to the long- established international system of submarine cables that allow for low cost transmission of data and voice communications. The Eastern Africa international submarine cable network is currently being laid off the coast of East Africa with funding from the World Bank and a consortium of private investors. It is expected to become operational in 2010. Burundi would be linked to this low-cost international communications network via fiber optic cables that are currently being laid in Kenya and Rwanda. This cable extension to Bujumbura is expected to be completed by June 2010. With further developments already underway, Burundi would have four separate access routes to this regional communications network and the submarine cable. Against this backdrop, the key elements of the communications program are as follows: A high priority is attached to the immediate development of a national communications grid of fiber optic cable and digital microwave that would be linked to the regional network. This program is already funded by the World Bank and will begin implementation shortly. On completion in 2012, it will have laid a fibre optic network of some 400 km throughout Burundi, along with a digital microwave network that will serve particular communities throughout the country. Launch an ambitious program to expand access to this low cost international communications network for schools, hospitals, universities, the business sector, and local communities throughout the country. Develop a range of applications, including e-government, e-commerce, e-schools, and e-health and complete the ongoing work on the legislative framework for the communications industry and corresponding regulatory framework related to e-security, fraud, privacy, data protection, and intellectual property rights. Promote the entry of additional private suppliers of communications services throughout the country to ensure competition and service quality. Such a national grid would give communities, businesses, and a wide range of institutions throughout the country, access to low cost communications within the country and with the region and rest of the world. Rural as well as urban communities are expected to benefit from this program. According to the World Bank, for example, incomes of agricultural producers can rise by about nine percent through the use of mobile telephones.
Building human and institutional capacities Given the size of the proposed program to be implemented over the next two decades, Burundi will have to make a large investment in building domestic capacities to manage the program and benefit fully from the services it will be able to provide. There are four specific components for the proposed capacity building program: Strengthen capacities of individual ministries for project design and implementation, including for example, arrangements for procurement and site supervision. These programs are of particular importance for the ministries responsible for power and transport. Strengthen and restructure arrangements for oversight and regulation of the power, transport and communications industries, given the substantial structural changes that would occur under the proposed Action Program. Strengthen capacities for regular collection and analysis of survey data for these three industries. In the case of road transport, for example, surveys of transport service providers will provide
115 basic information about the evolving amounts of passenger and freight traffic, the costs of service provision and prices of services offered to the public. Regular surveys of road traffic will be required for assessments of evolving road maintenance and upgrading requirements. Adopt appropriate standards for infrastructure construction and training of skilled workers for these industries. With specific standards for two lane paved national highways, for example, the donor community can then ensure that projects they fund do comply with these standards. Similarly, there is a need for clear standards for accredited institutions that train skilled trades people such as electricians. In developing these standards, close attention should be given to the evolving requirements of the EAC. The proposed program would provide substantial support for the development of these capacities and for a wide range of technical studies that will be required for informed decision making in the early phases of the program. There is considerable urgency associated with the start of these capacity building programs.
Required investment for infrastructure
Development expenditures The core infrastructure action plan calls for development expenditures of USD 4.6 billion (at 2007 constant prices, table A.11) over the next two decades. Successful implementation of this proposed program will essentially close the large infrastructure gap between Burundi and many other developing countries. The power program would involve expenditures of about USD 2 billion, including about USD 465 million of private investment in new domestic generation facilities that would sell power to the national grid. The roads program would require about USD 2.1 billion. The civil aviation program would involve a public-private partnership (PPP) arrangement under which the upgrade and operation of the airport and related services would be handled by one or more private contractors. The total amount of investment required for the aviation sector is estimated at USD 260 million over the next 20 years. The program includes about USD 120 million for the further development of the national communications grid and widespread community access to this grid.
Table A.11: Development expenditures on the core program Table A.12: Routine maintenance expenditures on the core (US$ millions, 2007 constant prices) program (US$ millions, 2007 constant prices) 2010 19 2020 30 Total 2010 19 2020 30 Total Public expenditures Public sector outlays Power sector 813 764 1,577 Power sector 132 409 540 Transport sector Transport sector Roads 1,139 989 2,129 Roads 105 160 265 Ports 13 15 28 Ports 4 6 10 Civil aviation 11 6 16 Civil aviation 3 3 Sub total 1,163 1,009 2,172 Sub total 112 166 277 Communications 48 28 75 Communications 9 19 29 Total 2,024 1,801 3,825 Total 253 594 847 Associated private Associated private investment sector Power sector 458 8 465 Power sector 79 153 232 Civil aviation 190 55 245 Civil aviation 33 83 116 Communications 24 33 57 Communications 4 14 18 Total 672 96 767 Total 115 250 366 Total 2,695 1,896 4,592 Total 368 844 1,212 Source: AfDB 2009. Source: AfDB 2009.
116 The bulk of these expenditures would be capital outlays on infrastructure assets such as road networks, airport facilities, power stations and transmission and distribution lines, and communications networks. About three percent of the outlays (USD 170 million) would be used to meet the cost of the wide-ranging program of capacity building initiatives and various technical studies included in the Action Plan. In the event that the railway extension was to proceed, the estimated cost is about USD 600 million (at 2007 constant prices), not including the cost of rail extensions to mine sites in the Musongati area. It is assumed that these latter costs would be met by the mine operator, in the event that the rail transport option was to be used. The proposal is to use a PPP-type arrangement to fund and operate the rail service if the project was to proceed. A small amount of public investment would be required for various studies and for capacity building within the government for oversight and regulation of the rail services.
Expenditures on routine maintenance The combination of prolonged civil war and limited public financial resources led to deterioration in the basic infrastructure of the county. The program of infrastructure rehabilitation that is ongoing and the proposals to complete this rehabilitation in the decade ahead will restore and upgrade these facilities. The challenge then will be to step up allocations for routine maintenance of this infrastructure, and thereby contain the need for further large outlays for rehabilitation. The proposed action plan for Infrastructure calls for USD 1.2 billion of outlays on routine maintenance over the next two decades (table A.12). The bulk of these expenditures are for the maintenance of the power and roads infrastructure. As the accompanying table indicates, these estimates for maintenance spending include expenditures by the public sector, as well as private sector outlays on power and civil aviation infrastructure that would be operated under some form of public-private partnership (PPP) arrangement. The proposed program represents a major step-up in funding for routine maintenance. In the case of the government component, outlays for maintenance average USD 25 million a year in the decade ahead. This compares with maintenance outlays for transport, power and communications infrastructure in the range of USD 8 million a year at the present time. In the 2020s, maintenance outlays by the public sector would step up to an average of about USD 85 million a year. These maintenance programs will provide important business opportunities for local companies and will create substantial additional employment opportunities. As the subsequent discussion indicates, early action will be needed to develop the requisite skills in the labor force (electricians, for example) for these programs.
- 117 - APPENDIX 6: BUSINESS ENVIRONMENT
Table A.13: Recommendations for Improving Financial Sector Performance Constraints Measures Timeline Stability Some state banks are in difficulty; Privatize two state owned banks 2010 supervision of banks is not up to Improve supervision by introducing chart of accounts 2009 par Introduce risk based supervision of commercial banks 2010 Computerize the BRB 2010–11 Payment system is archaic Introduce electronic mass clearing and real time gross settlement systems 2010 Introduce card system 2011 Pass new law on means of payments 2011 Microfinance presents Strengthen supervision through chart of accounts and 2010 weaknesses training 2009–12 Strengthen institutions through internal restructuring 2009–12 In the insurance sector there is an Establish independent regulatory/supervisory agency 2010 absence of any supervision and Restructure insurance companies to meet prudential norms or close them 2011 there are undercapitalized institutions that do not meet prudential norms Deficits exist in the pension Change the management structure 2011 systems; there are high costs and Introduce parametric reform 2011 low efficiency in delivering Conduct internal restructuring 2011 services Undertake a study of a second pillar 2012 Access Banks are not well prepared to Provide technical assistance to develop new products and new lending 2011 cater to small and medium mechanism, and establish specialized departments enterprises and rural areas Build up the capacity of the BNDE (Banque Nationale de Développement 2011 Économique) to lend in rural areas Microfinance institutions are ill Change regulation to broaden the scope of microfinance institutions and 2010 equipped to deal with small and allow them to develop new products and reach more customers medium enterprises and rural Develop financing organisms as provided for in the law 2012 areas, and are limited in their Establish relationship with the BNDE 2011–12 lending capacity Depth and reach of financial Develop mobile phone banking 2012 sector are very low Develop specialized institutions (leasing, venture capital, factoring) 2012–13 Develop warehousing 2012 Little housing finance is available Develop activities of FPHU (Fonds de promotion de l’habitat urbain)) 2011 Develop new products 2011 Adopt a land code 2010 Decentralize land registries and increase their budget allocation 2011 Authorize microfinance institutions to offer mortgage loans 2010 Provide training in housing finance 2010–13 Insufficient information is Strengthen the accounting and auditing profession 2010–11 available on potential borrowers; Assist firms in improving their financial statements and business plans 2010–12 there is a lack of structured Institute a dialogue among financial institutions, the nonfinancial private 2010– borrowers sector, and government Strenthen rural players thorugh their structuring into integrated value chains 2010–13 Strengthen existing information centers (for example, introduction of unique 2010–11 identifier, computerization, and so forth) Insurance products are limited Develop brokers 2009–10 Grant entry to foreign companies 2011– Diversify products of the insurance companies 2012–13 Source: Authors’ compilation.
- 118 - APPENDIX 7: MAP OF BURUNDI
- 119 - References
AfDB (African Development Bank). 2009. “An Infrastructure Action Plan for Burundi: Accelerating Regional Integration” Tunis, Tunisia.
Baghdadli I., B. Harborne, and T. Rajadel, eds. 2008. Breaking the Cycle: A Strategy for Conflict- sensitive Rural Growth in Burundi. World Bank Working Paper No. 147. World Bank, Washington, DC.
Balassa, B. 1986. “Comparative Advantage in Manufactured Goods: A Reappraisal.” Review of Economics and Statistics 68 (2): 315–19.
Baldwin, R., J. François, and R. Portes. 1997. “The Costs and Benefits of Eastern Enlargement: The Impact on the EU and Central Europe,” Economic Policy 12 (24): 125–76.
Benitez, D., A. Estache, and D.-M. Niyungeko. 2009. “Overcoming Obstacles to Growth in Burundi: The Role of Infrastructure.” Unpublished manuscript, World Bank, Washington, DC.
Boccardo, J., and V. Chandra. 2008. “Why Uganda Needs to Diversify Its Exports,” Background paper for the Uganda Country Economic Memorandum, World Bank, Washington, DC.
Brenton, P., and R. Newfarmer. 2007. “Watching More Than the Discovery Channel: Export Cycles and Diversification in Development” Policy Research Working Paper 4302, World Bank, Washington, DC.
Celasun, O., and J. Walliser. 2008. “Predictability of Aid: Do Fickle Donors Undermine Aid Effectiveness?” Economic Policy 23 (55): 545–94.
Chandra, V., and I. Osorio Rodarte. 2009. “Structural Transformation for Sustainable Growth in Burkina Faso.” Background paper for the Burkina Faso Country Economic Memorandum, World Bank, Washington, DC.
Clay, D., P. DeLucco, and A. Ottaway. 2007. “Integrative Report in Support of the Burundi Coffee Value Chain Diagnostic and Strategy Development.” Economic Management Support Project, Bujumbura, Burundi.
Clemens, M., S. Radelet, and R. Bhavani. 2004. “Counting Chickens When They Hatch: The Short-Term Effect of Aid on Growth.” Working Paper 44, Center for Global Development, Washington, DC. http://www.cgdev.org/content/publications/detail/2744
Collier, P., and A. Hoeffler. 2004. “Aid, Policy and Growth in Post-Conflict Societies.” European Economic Review 48: 1125–45.
Coulibaly, K., N. Duffy, and K. Ezemenari. 2008. “Productivity Growth and Economic Reform: Evidence from Rwanda.” Policy Research Working Paper 4552, World Bank, Washington, DC.
CPM Group. 2009. “The Current Market Environment and Outlook.” Presentation at the World Bank, Washington, DC.
Dijkstra, T. 2001. “Export Diversification in Uganda: Developments in Non-Traditional Agricultural Exports,” Working Paper, 47/2001, African Studies Centre, Leiden, The Netherlands.
- 120 - EAC (East African Community). 2006. “Development Strategy 2006–2010: Deepening and Accelerating Integration,” East African Community Secretariat, Arusha, Tanzania.
FAO (Food and Agriculture Organization). 2003. “Burundi : Profils des pêches et de l’aquaculture par pays.” Washington, DC. http://www.fao.org/fishery/countrysector/FI-CP_BI/fr.
Finger, J., and M. Kreinin. 1979. “A Measure of ‘Export Similarities’ and Its Possible Uses.” Economic Journal 89 (356): 905–12.
Foster, V., and C. Briceño-Garmendia, eds. 2009. Africa’s Infrastructure: A Time for Transformation. Washington, DC: World Bank.
Foster, M., and T. Killick. 2006. “What Would Doubling Aid do for Macroeconomic Management in Africa.” Working Paper 264, Overseas Development Institute, London, UK.
Fujita, M., and J.-F. Thisse. 2002. Economics of Agglomeration: Cities, Industrial Locations, and Regional Growth. Cambridge, UK: Cambridge University Press.
Global Witness. 2009. “Faced with a Gun, What Can You Do? War and the Militarisation of Mining in Eastern Congo.” July. http://www.globalwitness.org/fwag/.
Hausmann, R., J. Hwang, and D. Rodrik. 2007. “What You Export Matters.” Journal of Economic Growth 12 (1): 1–25.
Hausmann, R., and B. Klinger. 2006. “Structural Transformation and Patterns of Comparative Advantage in the Product Space.” Working Paper 128, Center for International Development, Kennedy School of Government, Harvard University, Cambridge, MA.
Hausmann, R., L. Pritchett, and D. Rodrik. 2005. “Growth Accelerations.” Journal of Economic Growth 10 (4): 303–29.
Heritage Foundation. 2009. “2009 Index of Economic Freedom.” Washington, DC. http://www.heritage.org/Index/Country/Burundi.
Hesse, H. 2009. “Export Diversification and Economic Growth.” In Breaking into New Markets: Emerging Lessons for Export Diversification, ed. R. Newfarmer, W. Shaw, and P. Walkenhorst, 55–80. Washington DC: World Bank.
Hidalgo, C., B. Klinger, A.-L. Barabási, and R. Hausmann. 2007. “The Product Space Conditions the Development of Nations.” Science 317 (5837): 482–87.
Hugon, P. 2003. Les économies en développement à l’heure de la régionalisation. Paris: Karthala.
Imbs, J., and R. Wacziarg. 2003. “Stages of Diversification.” American Economic Review 93 (1): 63–86.
IMF (International Monetary Fund). 2008. “Burundi Selected Issues: Assessing External Competitiveness in Burundi.” IMF Country Report No. 08/292, Washington, DC.
———. 2009. “Burundi: Poverty Reduction Strategy Paper—Annual Progress Report—Joint Staff Advisory Note.” IMF Country Report No. 09/91, Washington, DC.
- 121 - Kaufmann, D., A. Kraay, and M. Mastruzzi. 2009. “Governance Matters VIII: Aggregate and Individual Governance Indicators 1996–2008.” Policy Research Working Paper 4978, World Bank, Washington, DC.
Krishna, P. 2003. “Are Regional Trading Partners ‘Natural’?” Journal of Political Economy 111 (1): 202– 31.
Krugman, P. 1991. “The Move toward Free Trade Zones.” Federal Reserve Bank of Kansas City. Economic Review (November/December): 5–24.
———. 1995. Development, Geography, and Economic Theory. Cambridge, MA: MIT Press.
Lall, S. 2000, “The Technological Structure and Performance of Developing Country Manufactured Exports, 1985–98.” Oxford Development Studies 28 (3): 337–69.
Lederman, D., and W. Maloney. 2009. “Trade Structure and Growth.” In Breaking into New Markets: Emerging Lessons for Export Diversification, ed. R. Newfarmer, W. Shaw, and P. Walkenhorst, 39–54. Washington DC: World Bank.
Lederman, D., M. Olarreaga, and L. Payton. 2009. “Export Promotion Agencies: Strategies and Impacts.” In Breaking into New Markets: Emerging Lessons for Export Diversification, ed. R. Newfarmer, W. Shaw, and P. Walkenhorst, 211–22. World Bank, Washington, DC.
May, J. 2009. “Rapid Population Growth in Burundi: Implications for the World Bank.” Unpublished manuscript. World Bank, Washington, DC.
May, J., and A. Kamurase. 2009. “Demographic Growth and Development Prospects in Rwanda: Implications for the World Bank.” World Bank, Washington, DC.
McGillivray, M., and S. Feeny. 2008. “Aid and Growth in Fragile States,” Research Paper No. 2008/03, United Nations University–World Institute for Development Economics Research, Helsinki, Finland.
Midende, G. 2009. “Etude sur les exploitations minières artisanales du Burundi.” Bujumbura, Burundi.
Mo Ibrahim Foundation. 2009. “2009 Ibrahim Index: Burundi.” London, UK. http://www.moibrahimfoundation.org/en/media/get/20091004_burundi.pdf.
Naudé, W. 2007. “Geography and Development in Africa: Overview and Implications for Regional Cooperation.” Discussion Paper No. 2007/03, United Nations University–World Institute for Development Economics Research, Helsinki, Finland.
Ndulu, B., L. Chakraborti, L. Lijane, V. Bamachandran, and J. Wolgin. 2007. Challenges of African Growth: Opportunities, Constraints, and Strategic Directions. World Bank, Washington, DC.
Nielsen, H., and D. Madani. 2010. “Potential Benefits and Risks of Increased Aid Flows to Burundi” Research Working Paper 5180, World Bank, Washington, DC.
Nkurunziza, J., and F. Ngaruko. 2002. “Explaining growth in Burundi: 1960–2000.” Working Paper No. 162, Centre for the Study of African Economies, Department of Economics, Oxford University, Oxford, UK.
- 122 - OTF Group. 2008. “Vision et Stratégie Industrielles et Commerciales du Burundi: Rapport Final.” Economic Management Support Project, Bujumbura, Burundi.
Page, J. 2008. “Rowing Against the Current: The Diversification Challenge in Africa’s Resource-Rich Economies.” Global Working Paper No. 28, Brookings Institution, Washington, DC.
———. 2009. “Africa’s Growth Turnaround: From Fewer Mistakes to Sustained Growth.” Commission for Growth and Development Working Paper No. 54. World Bank, Washington, DC,
Population Reference Bureau. 2009. “2009 World Population Data Sheet.” Population Reference Bureau, Washington, DC. http://www.prb.org/Publications/Datasheets/2009/2009wpds.aspx.
Rauch, J. 1999. “Networks Versus Markets in International Trade” Journal of International Economics 48 (1): 7–35.
Republic of Burundi. 2006. “Poverty Reduction Strategy Paper.” September, Bujumbura.
———. 2007a. “Assistance technique au Service Chargé des Entreprises Publiques (SCEP) en matière de réorme, restructuration et privatisation des Sociétés à Participation Publique.” Projet d’Appui à la Gestion Economique, February, Bujumbura.
———. 2007b. “Planification basée sur les objectifs du millénaire pour le développement : Rapport général d’évaluation des besoins pour l’atteinte des OMD.” Ministry of Development Planning and National Reconstruction and United Nations Development Programme, March, Bujumbura.
———. 2008a. “Etude diagnostique sur la gouvernance et la corruption au Burundi: Résumé Exécutif.” Ministère à la Présidence Chargé de la Bonne Gouvernance, de la Privatisation, de l’Inspection Générale de l’Etat et de l’Administration Locale, May, Bujumbura.
———. 2008b. “Stratégie Agricole Nationale 2008–2015.” Ministère de l’Agriculture et de l’Elevage, July, Bujumbura.
———. 2009. “Stratégie Nationale de Gouvernance et de Lutte Contre la Corruption: Guide des Consultations Sectorielles.” Ministère a la Présidence Chargé de la Bonne Gouvernance et de la Privatisation, in cooperation with the World Bank and the Institut de Développement Economique, Bujumbura.
Rhata, D., S. Mohapatra, and A. Silwal. 2009. “Migration and Remittance Trends 2009.” Migration and Development Brief 11, World Bank, Washington, DC.
Ross, M. 2008. “Oil, Islam, and Women.” American Political Science Review 102 (1): 107–23.
Summers, L. 1991. “The Move toward Free Trade Zones: Comment.” Federal Reserve Bank of Kansas City. Economic Review (November/December).
UNIDO (United Nations Industrial Development Organization). 2009. Industrial Development Report 2009. Breaking In and Moving Up: New Industrial Challenges for the Bottom Billion and the Middle- Income Countries. Vienna, Austria: UNIDO.
United Nations. 2009. “World Population Prospects: The 2008 Revision.” Department of Economic and Social Affairs, Population Division, New York. http://www.un.org/esa/population.
- 123 - von Grebmer, K., H. Fritschel, B. Nestorova, T. Olofinbiyi, R Pandya-Lorch, and Y. Yohannes et al. 2008. “Global Hunger Index: The Challenge of Hunger 2008.” Welthungerhilfe, International Food Policy Research Institute, and Concern Worldwide, Bonn, Germany.
World Bank. 2007. “Population Issues in the 21st Century. The Role of the World Bank.” Health, Nutrition, and Population Discussion Paper, Washington, DC.
———. 2008a. “Burundi: Une évaluation du climat des investissements—Rapport Final.” Report prepared by Etude Economique Conseil Canada, Montreal.
———. 2008b. “Country Assistance Strategy for the Republic of Burundi for the Period FY09–FY12.” Report No. 44193-BI, Washington, DC.
———. 2008c. “Non-Tariff Measures on Goods Trade in the East African Community: Synthesis Report.” Report No. 45708-AFR, Washington, DC.
———. 2008d. “Regional Integration Assistance Strategy.” Final Report, March 11, Washington, DC.
———. 2008e. “Republic of Burundi—Public Expenditure Management and Financial Accountability Review (PEMFAR): Improving Allocative Efficiency and Governance of Public Expenditure and Investing in Public Capital to Accelerate Growth and Reduce Poverty.” Report No. 42160-BI, Washington, DC.
———. 2009a. Doing Business 2010: Burundi. World Bank, Washington, DC.
———. 2009b. “Republic of Rwanda: Fifth Poverty Reduction Support Grant, Program Document.” Report No. 46455-RW, World Bank, Washington, DC.
- 124 -