(2002) Évaluation de l'Assistance Pays Haïti

Report No. 23637 Haiti Country Assistance Evaluation February 12, 2002 Operations Evaluation Department Document of the World Bank Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Acronyms CAE Country Assistance Evaluation CAS Country Assistance Strategy CIDA Canadian International Development Agency CG Consultative Group CLF Caribbean Loan Facility CPIA Country Policy and Institutional Assessment EERC/P Emergency Economic Recovery Credit/Plan ERC Economic Recovery Credit ESW Economic and Sector Work EU European Union FAES Fonds d'Assistance Economique et Sociale (Economic and Social Fund) GAO (United States) Government Accounting Office GEF Global Environment Facility GDP Gross Domestic Product GNP Gross National Product HID Human Development Index IDI Institutional Development Impact IDB Inter-American Development Bank IFC International Finance Corporation IMF International Monetary Fund LIL Learning and Innovation Loan MIGA Multilateral Investment Guarantee Agency MCN Micro Credit National NGO Non-governmental organization OAS Organization of American States OED Operations Evaluation Department PMU Project Management Unit PSIP Public Sector Investment Program PER Public Expenditure Review PRSP Poverty Reduction Strategy Paper SAR Staff Appraisal Report SMP Staff Monitored Program TA Technical Assistance UNDP United Nations Development Programme USAID United States Agency for International Development Director-General, Operations Evaluation: Mr. Robert Picciotto Director, Operations Evaluation Department: Mr. Gregory K. Ingram Manager, OEDCR: Mr. Ruben Lamdany Task Manager: Ms. Alice Galenson The World Bank Washington, D.C. 20433 U.S.A. Office of the Director-General Operations Evaluation February 12, 2002 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Haiti-Country Assistance Evaluation This Country Assistance Evaluation covers the last fifteen years, a tumultuous period in Haiti. Since the overthrow of the Duvalier regime in 1986, the extraordinary instability of earlier years has continued, with thirteen governments and two periods during which most donor activities, including Bank lending, ceased altogether. Haiti's social and economic indicators are comparable to those of the poorest countries in the world. Roughly two-thirds of the population live in poverty, half of adults are illiterate, and health services are inadequate to address high infant and maternal mortality and major public health crises, including TB, HIV/AIDS, and, most recently, polio epidemics. Real per capita GDP fell at a rate of 2 percent p.a. during the 1980s and 2.5 percent p.a. during the 1990s. The continuous low intensity conflict and political crisis, with peaks of violence, led to Haiti's classification by the Bank as a post-conflict country. Poor governance and political instability are the major impediments to sustainable development, along with weak public sector capacity and accountability, followed by low levels of education, badly deteriorated infrastructure, a paucity of arable land relative to the population, and a devastated environment. The Bank, in collaboration with other donors, has tried to tackle these challenges through emergency projects to support economic recovery, stabilization, public sector reform, privatization, and emergency social programs, as well as investments in infrastructure and a few projects in environment, education and health. While the Bank's objectives were consistent with Haiti's major economic problems, their relevance was limited by their failure to give highest priority to resolving the political and governance problems that undermined economic development. Recent economic and sector work, most notably the 1998 poverty report, has been of good quality, but its impact has been limited by the unresolved political situation. New lending has been blocked since 1997 by the absence of a functioning parliament. The few remaining projects under implementation have been cancelled. Projects in Haiti have unusually low outcome ratings, along with very limited institutional development impact and sustainability. Other than for emergency fast-disbursing projects, positive results tend to be found in project components that bypass traditional channels, using NGOs or autonomous agencies to distribute services and funds. Since 1994, the Bank has led aid coordination, which is remarkable for its intensity, including collaborative economic and sector work, but which has a mixed record on the ground, where inadequate division of labor, multiple conditions and complex processes, have overwhelmed weak institutions. The efficacy of the Bank's program has been negligible, and its efficiency, low. Other than the reform of trade policy and macroeconomic management in the late 1980s and early 1990s, lending has had little impact. Emergency projects were rated satisfactory, but had limited objectives and no longer term relevance. Despite four Bank education projects (a fifth was cancelled), as well as sizable contributions from other donors, Haiti has enrollment and literacy rates that are among the lowest in the world. Private schools, many of dismal quality, account for 80 percent of enrollment. The single health project had some success in treating TB, but accomplished little in primary health care or treatment of HIV/AIDS. Agricultural and environmental projects have failed to stem the decline of agriculture and the deterioration of the environment. And despite nine Bank projects, and assistance from other donors, deterioration of infrastructure has reached critical levels. As is the case in many other small and poorly performing countries, the cost per US$1,000 of net commitment for satisfactory/non-risky projects is high: US$33 in Haiti, compared to US$10 for the LCR Region and US$16 Bank-wide. The development impact of Bank assistance to Haiti since 1986 has been severely limited. The critical constraints to development-governance and public sector capacity and accountability-have not diminished, nor have any sectors registered substantial improvements. Based on both its impact and the ratings of its individual components, the outcome of the assistance program is rated unsatisfactory (if not highly so), the institutional development impact, negligible, and the sustainability of the few benefits that have accrued, unlikely. The Bank and other donors erred by offering traditional assistance programs without identifying the fundamental governance and political barriers to development, and by overwhelming the fragile absorptive capacity. Bank performance has, however, improved in recent years through increasingly realistic assessments of Haiti's problems. The international community has established three conditions for re-engagement in Haiti: a resolution to the political crisis, macroeconomic stability, and a commitment by the government to undertake critical sectoral reforms. These conditions have not been met. Even once they are met, the Bank should approach any re-engagement in Haiti with extreme caution. The means of re-engagement should be presented in a Transitional Support Strategy paper, with highest priority on the reform of governance and institutions. An Institutional and Governance Review could be undertaken to analyze the institutional constraints to reform, with particular attention to the political economy factors which have made it difficult in the past to change the status quo. Re-engagement could be initiated through the use of Post-Conflict grants to reestablish a dialogue with the government and with civil society. Any lending should be directed toward piloting activities that could contribute to institution building, possibly through the use of LELs. Finally, any move toward re-engagement should proceed in close coordination with other donors at all levels: agreement on objectives, determination of responsibilities in line with comparative advantage, and establishment of consistent operational procedures. Contents Prefac ................................................................................................................................. i 1. Introduction ............................................................................................................... 1 Social, Political and Economic Developments........................................................... 1 Challenges to Developm ent ........................................................................................ 3 2. Bank Products andSeryices ..................................................................................... 5 Strategic and Policy Advice........................................................................................ 5 Economic and Sector W ork ........................................................................................ 6 L en d in g ....................................................................................................................... 7 IFC and M IGA .......................................................................................................... 12 Resource M obilization and Aid Coordination.......................................................... 13 3. Development Impact of Bank Assistance.............................................................. 15 Relevance and Efficacy of the Country Assistance Program ................................... 15 Efficiency of the Country Assistance Program ......................................................... 16 R atin g s ...................................................................................................................... 17 4. Attribution............................................................................................................... 17 5. Lessons of Experience and Recom m endations.....................................................19 Box in Text 2.1 W orking with NGOs- A Relatively Successful Experience................................... 13 Table in Text 2.1 OED Evaluation Findings........................................................................................ 7 Annexes A. Statistical Annexes Annex Table 1: Haiti At a Glance........................................................................... 23 Annex Table 2: Key Economic and Social Indicators, 1990-2000........................ 25 Annex Table 3: External Assistance to Haiti.......................................................... 26 Annex Table 4: ESW List for Haiti........................................................................ 27 This report was prepared by Alice Galenson (Task Manager). John Eriksson and Gianni Zanini peer-reviewed the report. Anar Omarova provided research assistance. Silvana Valle provided administrative support. Contents (cont.) Annex Table 5: OED and QAG Ratings for Haiti and Comparator Countries...... 28 Annex Table 6: Cost of Bank Programs for Haiti and Comparator C ountries, FY 91-99............................................................................................... 29 Annex Table 7: List of Approved Projects ............................................................. 30 Annex Table 8: Bank Senior Management for Haiti, 1990-2001........................... 31 Annex Table 9: International Development Goals ................................................. 32 B. List of People Interviewed....................................................................................... 33 C. Management Action Record .................................................................................... 35 D. Report from CODE ................................................................................................. 37 Attachments 1. Government Comments ........................................................................................... 39 2. Government Comments, English version ................................................................ 63 3. OED Response to Government Comments.............................................................. 79 Bibliography .................................................................................................................... 81 1 Preface This Country Assistance Evaluation (CAE) examines Bank assistance to Haiti.' It focuses on the fifteen years since the overthrow of the Duvalier regime in early 1986, but draws on material from earlier experience when relevant to establish patterns and trends. The CAE is a countrywide evaluation that concentrates on the relevance, efficacy, efficiency, sustainability and institutional development impact of the Bank's program of assistance. The CAE is issues-oriented, is geared toward current decision-making, and is selective as to time span and issues covered; the focus is on issues which are important today. Section 1 describes social, political and economic developments in Haiti, as well as the major challenges to development. Section 2 evaluates Bank assistance from the bottom up by assessing the Bank's products and services: strategic and pplicy advice, economic and sector work, lending, and resource mobilization and aid cdordination. Section 3 evaluates the Bank's development impact. Section 4 discusses the attribution of the development results to the Bank, the country and other partners, as well as to exogenous influences. Section 5 presents lessons and recommendations. Good practice in Bank assistance is highlighted in Box 5.1. The CAE is based on Bank strategy and lending documents, economic and sector reports (both formal and informal), and Project/Implementation Completion Reports; project and general country files; IMF reports; and interviews with staff of the Bank, the IMF, IFC, and other donors and experts. Annex B lists those interviewed. IFC and MIGA staff contributed information on their programs in Haiti. In addition, the CAE draws on OED's 1998 report "The World Bank's Experience with Post-conflict Reconstruction," which included a desk study of Haiti. The government provided a constructive and substantive response to an earlier version of the CAE, including both comments on the text of the report and an alternative agenda for the future. The comments have been taken into account in this draft of the report, and the response in its entirety is presented in Attachments 1 (in the original French) and 2 (the English translation). Attachment 3 contains a brief response from OED. The response to the CAE from Management is included in the Management Action Record (Annex C). The CAE was prepared in parallel with a similar evaluation by the Office of Evaluation of the Inter-American Development Bank of their program in Haiti. The CAE benefited from consultations and comments from the IDB task manager, and the two evaluations generally agree on their analysis. The IDB report will be issued shortly. 1. Haiti is an IDA country, but for simplicity this report refers to "the Bank" in discussing the assistance program. 1. Introduction Social, Political and Economic Developments 1.1 Haiti, the poorest country in the western hemisphere, has a GNI per capita of US$480 (2000), and a human development index (HDI) comparable to the poorest countries in the world.' Social indicators are well below those of comparator countries (Annex Tables 1 and 2). Roughly two-thirds of Haiti's 7.8 million people live below the poverty line. Half of adults are illiterate, and less than one quarter of rural children attend primary school. Infant and maternal mortality rates are among the highest in the world, and half the population lacks access to health services. Malaria, tuberculosis and HIV/AIDS are major public health problems. 2 Less than 40 percent of the population has safe drinking water, and sanitary sewer systems are almost non-existent. 1.2 The national level indicators do not reflect the situation in rural areas where two- thirds of the population live, 80 percent of them poor. Poverty and demographic pressure, 3 coupled with inefficient farming practices and vulnerability to natural disasters, combine to make Haiti one of the most environmentally degraded places in the world. Social programs aimed at reducing poverty by improving basic education, health and infrastructure are mainly funded by donors and implemented directly by NGOs; these programs have declined in recent years because of political instability. 1.3 Ever since Haiti gained independence from France in 1804, resources and power have been monopolized by a small elite, which often used force to control the country. Corruption and rent seeking are pervasive, and the growing wealth of the minority contrasts with the growing poverty of the majority. From 1957 to 1971, Haiti was ruled by Francois Duvalier, who closed the economy and conducted a reign of terror. Jean- Claude Duvalier succeeded his father and began to open and modernize the economy, attracting foreign aid; he was overthrown in 1986. After several years of violence and political unrest, Haiti held its first democratic elections in 1990. 1.4 Jean-Bertrand Aristide, elected president with 70 percent of the vote, took office in February 1991, but was ousted by a military coup later that year. A United Nations peacekeeping force intervened and eventually helped President Aristide return in October 1994. Parliamentary and presidential elections were held in 1995, and President Aristide, constitutionally prohibited from running for reelection, was succeeded by Ren6 Pr6val in 1996. A year later the resignation of the prime minister and the subsequent absence of a 1. The UNDP's human development index (HDI) measures a country's achievement in terms of life expectancy, educational attainment, and adjusted real income. Haiti scores 0.44 on a scale of 0 to I (UNDP Human Development Report 2000). 2. The incidence of HIV/AIDS is estimated to be as high as 12 percent for the urban population and 5 percent for rural. 3. Haiti is one of the most densely populated countries of the world with more than 270 inhabitants per square kilometer. It has the highest population per hectare of arable land in the LAC Region. 2 functioning parliament meant Haiti could no longer borrow money, since parliament must approve new debt. Parliamentary elections were held in 2000 but the method used to determine winners of legislative seats was contested by national and international election observers, opposition parties and the international community. Despite the absence of a provisional electoral council to organize elections and a boycott by the opposition, presidential elections proceeded in late 2000. Aristide was elected President and in the spring of 2001 pledged to hold new senate elections and establish a provisional electoral council that would include opposition and civil society members. Neither the opposition nor the government has shown much inclination to deal constructively with the other, and civil society is similarly divided. Political negotiations remain stalled, and with them the resumption of normalized relations with the international community. 1.5 Haiti's economic performance during the last fifteen years has been weak. Inappropriate economic policies and internal political conflict were further aggravated by the military coup in 1991 and the resulting embargoes on most trade and financial transactions and suspension of all but humanitarian external aid for three years. During those years, real GDP fell by close to 25 percent, inflation rose, activity in the textile and export-oriented assembly industries (responsible for over three-quarters of export earnings and significant employment) virtually ceased, the tax collection and expenditure control systems broke down, and maintenance of economic and social infrastructure was all but abandoned (Annex Table 2). Private remittances exceed exports by about one- third, averaging around 7 percent of GDP since 1994. 1.6 Following the return to constitutional rule in 1994, the government adopted an Emergency Economic Recovery Plan (EERP), supported by the international community. Economic performance began to recover, but deteriorated again, particularly on the fiscal side, in 1996 due to changes in the government and the run-up to the presidential election. The new administration took steps to improve economic management, but the political crisis in mid-1997 interrupted once again the implementation of critically needed reforms, the commitment of external aid, and the recovery of investment and output. Since then the IMF has had a series of staff-monitored programs (SMPs) to help the government maintain sound macroeconomic policies and establish a track record that could permit Fund support once the political crisis is resolved. After having declined by about 2 percent p.a. in real terms during the 1980s, per capita GNP fell at an average rate of 2.5 percent p.a. during the 1990s (Annex Table 2). Performance under the SMPs since October 1999 has been disappointing: the fiscal deficit for the first six months of the most recent one was over 3 percent of GDP on an annual basis, a level that is well beyond what Haiti can afford, given the absence of external assistance. 4 4. During the 1990s, public revenue averaged under 7 percent of GDP, while external assistance through 1998 amounted to close to 12 percent of GDP (Annex Tables 2, 3). External assistance has fallen sharply since then. IDA's annual disbursements to Haiti fell from a peak of US$56 million in FY97 to US$4.7 million in FY01, and in the latter year, IDA had a negative flow of transfers of about US$3.5 million. 3 1.7 Haiti has not experienced distinct pre or post-conflict periods, but has suffered continuous low intensity conflict, with peaks of violence. OED's post-conflict report noted that the three years following the coup d'6tat in September 1991, while not a civil war, had the characteristics of a complex emergency, with political, social and economic collapse. Economic mismanagement and the international embargo led to a dramatic decline in living standards. Estimates are that at least 300,000 of the 7 million people were displaced internally (World Bank, OED, 1998). Because of the continuing political crisis, Haiti is still classified as a post-conflict country. Internal Bank incentives favor this classification, because it gives access to grant funding windows, and exempts the country from standard CPIA treatment and corresponding IDA allocations. However, one can question this classification given the recurrent nature of the instability and its deep-seated, historical roots. Challenges to Development 1.8 Poor governance is the greatest impediment to effective development assistance in Haiti.s Since the overthrow of the Duvalier regime in 1986, Haiti has had thirteen governments, and two periods during which Bank lending (and most other donor activities) ceased altogether. The Bank's 1998 Poverty Report noted (more than a decade after the fall of the Duvalier regime) that "Haiti has never had a tradition of governance aimed at providing services to the population or creating an environment conducive to sustainable growth. Instead a small economic elite has supported a 'predatory state' that makes only negligible investments in human resources and basic infrastructure.... [P]ervasive repression through army, police, and paramilitary groups has created deep- seated distrust between civil society and the state.... [T]he absence of a culture of democratic decision making and peaceful consensus building ... has generated tensions... and hampered Haiti's rehabilitation effort." The report underlined poor governance as a major determinant of Haiti's high poverty levels. 1.9 A report on governance and social justice in the Caribbean (Dollar, 2000) argued that Haiti is an "extreme case of a country caught in a vicious circle in which unemployment, inequality, and poor education feed into lawlessness and violence, making it difficult for the economy to grow and create jobs, thus perpetuating the unemployment and inequality." 6 Without fundamental reforms to the political and 5. Public governance can be defined as the way power and authority are exercised in the use, distribution, allocation and management of public resources. Good governance includes limitations on bureaucratic harassment, the rule of law and the control of corruption and crime, the provision of sound regulatory structures for the promotion of a competitive private sector, and strong financial institutions. (Nicholas Stem, "Strategy for Development," Annual Bank Conference on Development Economics, World Bank, Washington, DC, May 1-2, 2001.) Transparency and participation are also important. The World Bank first discussed governance issues in Africa in 1985, and although they were not mainstreamed in the Bank until the 1990s, Haiti is a country where it would have been reasonable to expect attention to governance by the late 1980s and early 1990s. 6. Research has established the fact that good governance and good social outcomes are mutually reinforcing, as are poor governance and poor social outcomes. In this study Haiti scored very low on all of the measures of governance- rule of law, voice and participation, political instability and violence, corruption, and government effectiveness. 4 institutional obstacles to progress, no other reforms, however important in their own right, will work. 1.10 Public sector capacity is also a serious obstacle to aid absorption. Many of the institutions that existed under the Duvalier regime disappeared during the embargo, and new ones have yet to emerge. OED's post-conflict report found a total mismatch between the levels of foreign aid and government capacity. The government was overwhelmed by the diverse, complex procedures of donors, and combined with the governance problems noted above, this resulted in slow or nonexistent implementation, with the exception of emergency projects. The benefits from past investments have not been sustained due to a long neglect of maintenance-itself a major challenge to development. The 1996 CAS called for a major reform of Haiti's public sector, needed to implement reforms in other areas and attain long term growth. But brain drain has decimated the ranks of both professional and technical skills, and a poorly designed civil service reform during 1998-1999 led to the loss of many well qualified people. Decentralization efforts have suffered from the absence of a regulatory framework, political support from the center and financial support for the decentralized institutions. 1.11 Public financial accountability requires a transparent process by which the public and its representatives in Parliament hold the government accountable for its spending. Public Expenditure Reviews a decade apart (PER, World Bank, 1987) found that 17 percent of Treasury outlays went for unidentified recurrent expenditures. A decade later, another PER reported similar findings (Faria and Moreno-Lopez, 1997). "Current accounts," outside the normal ex ante and expost control procedures were used for close to one-fifth of government current expenditures in FY1995/96 (and over 40 percent the year before); they have since risen to about half. These accounts are justified as a way to circumvent cumbersome budgetary processes, but it is virtually impossible to identify their actual use, beneficiaries, or impact. Another issue-highlighted by a government background paper to the PER (Ministry of Economy and Finance, 1997)-was the large amount of externally funded investment not recorded or subject to budgetary process, but executed and monitored directly by donors through NGOs and line ministries. Serious weaknesses in financial management practices, including poor record keeping, submission of withdrawal applications for ineligible expenditures, misuse of funds, missing audits, and non-competitive bidding procedures, have been found in Bank projects. The Bank's second technical assistance (TA) project, intended to help improve the quality of public expenditure information, never became effective. The Country Procurement Assessment Report (draft, 1999) found the virtual absence of a national procurement system and noted that this made misprocurement likely. 1.12 Haiti also suffers from other, more traditional constraints to its development. A multi-donor mission in October 1994 concluded that it was highly unlikely that much economic progress could be achieved without an improvement in educational quality and access; this finding is still valid. Infrastructure has deteriorated badly due to lack of maintenance. A paucity of arable land and widespread use of firewood for cooking and commercial purposes have led to almost complete deforestation. Migration to urban areas has strained water and sanitation facilities. It is difficult to tackle these problems effectively, however, without first addressing the pervasive issue of governance 5 2. Bank Products and Services Strategic and Policy Advice 2.1 Following the overthrow of the Duvalier regime, the Bank and other donors supported the government's Economic Recovery Program, with the objectives of macroeconomic stabilization and improved resource allocation, through the reform of taxes, public expenditure, public enterprises, industrial incentives, trade and agricultural pricing and credit. Lending comprised a US$40 million Economic Recovery Credit (FY87) and projects in transportation, water supply, power and industrial restructuring. The strategy for Bank assistance to Haiti in the early 1990s stressed poverty alleviation, human resource development, infrastructure rehabilitation, and a comprehensive approach to the environment. 2.2 The joint donor mission in 1994 identified an Emergency Economic Recovery Program comprising stabilization, incentives for private sector development, promotion of economic efficiency, short term social programs, and arrangements to channel aid. The President's Report for the credit supporting that program contained a statement of the Bank's emergency assistance strategy for the following 12 -18 months. That strategy focused on finance of critical imports; modification of the ongoing portfolio to address emergency needs and add technical assistance; and policy dialogue, economic and sector work, and donor coordination. Aside from technical assistance projects to strengthen Haiti's limited capacity to absorb the recovery projects, the Bank's strategy did not address governance or institutional issues. (They were, however, addressed by other donors. 8 ) 2.3 A formal Country Assistance Strategy (CAS) in 1996 returned to the overarching objective of poverty alleviation (through safety nets and improved social services), as well as rebuilding and redirecting the public sector to deliver basic goods and services and implementing an economic reform program to enable sustainable growth. This might have been an appropriate strategy for a country where the rule of law is already in place, but Haiti was not such a country. While the Bank's objectives were consistent with Haiti's major economic problems, their relevance continued to be limited by the failure to give highest priority to resolving the political and governance issues that undermined all attempts to promote economic development. The same was true of the government's 1996 Policy Framework Paper, prepared through close collaboration among the Bank, the IIVF and the IDB. The CAS also stressed non lending services to promote a dialogue with 7. President's Report for "Proposed Emergency Economic Recovery Credit," Report No. P-6482-HA, November 30, 1994. 8. The U.S. provided nearly US$100 million during 1995-2000 to help Haiti establish its first civilian-controlled police force and improve aspects of its judicial sector. The IDB planned a set of loans to strengthen the executive, legislative and judicial branches, the superior court of accounts and civil society. The U.N. and CIDA were also involved in these areas. The government notes that reforms were attempted in many other areas as well, including civil service, fiscal affairs, education, health systems, financial sector, and democracy. 6 stakeholders, the importance of working with NGOs, and the need to serve as a focal point for donor coordination; these objectives were appropriate and realistic. A significant gap was the absence of HIV/AIDS. 9 Economic and Sector Work 2.4 The nature of economic and sector work (ESW) for Haiti has changed over the past 15 years. Several voluminous economic reports and an agricultural sector review were produced in the second half of the 1980s and the early 1990s. They covered the macroeconomic and sectoral issues facing Haiti, but while they acknowledged that economic developments in Haiti have continuously been affected by institutional weaknesses and political instability, and that investment in institution building should be an ongoing development process, they devoted little space to these issues. For example, the list in the lengthy 1991 economic report of the most pressing issues to be addressed in the following 18 months did not include institutions or governance. 2.5 Since the mid-1990s, ESW has been more focussed: a public expenditure review carried out jointly with other donors (PER, 1997), a poverty report (1998), a Country Procurement Assessment Report (draft, 1999) and a series of sector reviews (infrastructure, education, health, all in draft). This approach is consistent with the Bank's post-conflict policy which, where continued lending is not possible, calls for a watching brief to position the Bank to support operations when conditions permit. However, some of the work has not been followed up; the relatively costly PER, for example, was never compiled into one report or used to establish priorities.' 0 2.6 The 1998 poverty report identified political instability, poor governance and corruption as key factors behind the dire poverty in Haiti. It recommended a reform agenda of stronger public sector institutions; macroeconomic stability and reduction of distortions to encourage private sector investment; more resources for the provision of social services; and rationalization of donor assistance. Given the limited capacity of the government, the report suggested that donors focus on maintaining public order and economic stability; securing property rights; building a regulatory role for some sectors; designing a framework for improved provision of basic health and education; and maintaining infrastructure. The report also recommended privatizing key infrastructure and entrusting the delivery of education, health, family planning, and water supply and sanitation to NGOs. In principle, these objectives were highly relevant. However, they may have lacked realism, given that privatization had already proved to be contentious in Haiti. The poverty report was widely disseminated and generated the first open debate 9. In fact, as the government points out, there was very little lending for social purposes after 1996 (see below). The government also notes that neither the CAS nor subsequent loan documents explained clearly the means by which poverty would be reduced. In addition, it asserts that the levels of external aid projected in the CAS were not large enough to jump-start growth. 10. The Region points out that the choice was made to concentrate limited resources on the dissemination of the poverty report. 7 about poverty in Haiti. Its impact on policy has been limited, however, by the unresolved political situation and resulting halt to dissemination and lack of follow through. Lending 2.7 Haiti has always been a difficult country in which to lend. Of the 29 projects rated since 1979, only 15 had satisfactory outcomes. Eighteen projects were rated for institutional development impact (IDI) and sustainability: only 3 had substantial IDI and 1 was considered likely to be sustainable (Table 2.1). Completion reports catalogue a history of implementation problems and weak government support (OED, 1998)." 2.8 Lending since 1986 can be divided into two periods: 1986-1991, after which the military coup led to the suspension of lending for three years; and 1994-1997, after which the political situation once more blocked lending. The remaining active projects were eventually placed under suspension due to arrears, and then cancelled as a measure to end ineffective lending and reduce portfolio risks. Table 2.1: OED Evaluation Findings Number of Number of Number of Total Number of Number of Projects w/ Projects w/ Projects w/ Number of Projects Projects satisfactory substantial likely projects Cancelled rated outcome IDI sustainability All projects 37 3 29. 15 3 (of 18) 1 (of 18) Projects approved 1986-91 10 2 7 3 2 0 Projects approved 1994-97 5 1 2 2 1 1 a. Only 18 were rated for IDI and sustainability. b. No projects have been approved since 1997. Source: Annex Table 7. Economic Management, 1986-1991 2.9 During the post-Duvalier period, from 1986 until 1991, the Bank approved 10 credits, of which two have been cancelled and one is not yet rated. Only 3 of the remaining 7 projects had satisfactory outcomes, 2 had substantial II, and none was considered likely to be sustainable.' 2 Of 8 projects approved earlier that closed during this period, 6 had unsatisfactory outcomes, all were rated unlikely to be sustainable, and none had substantial IDI. 2.10 The satisfactory post-1986 projects comprised two emergency operations and a TA project. The Economic Recovery Credit (ERC, FY87) supported fiscal and trade 11. The Aggregate Project Performance Index is 5.4, compared to 6.7 for the Bank as a whole. The APPI is a composite measure of the ratings on outcome, institutional development impact and sustainability, and is calculated only for projects which have ratings for all three variables. The score can range from 2 to 10, with higher scores better than lower. (Annex Table 7). 12. During this period, three arrangements with the IMF also failed. 8 reform, which remained relatively sound for some years, and the closing of two loss- making public industrial enterprises. OED's audit of the project concluded, however, that the project should also have addressed structural poverty and the negative social impact of stabilization. The second tranche of the credit was cancelled following the military coup of 1988. Sustainability was rated unlikely, and IDI modest. The outcome of the TA project was rated satisfactory, with substantial IDI, but its chief accomplishments were nullified when the projects they led to were either cancelled or rated unsatisfactory. 2.11 The outcome of the Economic and Social Fund project (FAES, FY91) was rated satisfactory because it largely achieved its physical objectives of providing infrastructure and jobs. IDI was rated substantial, although the expected decentralization of project management did not take place, and participation by grass roots organizations was limited by their lack of capacity to prepare proposals. Sustainability was deemed uncertain, due to poor maintenance and inadequate attention to operating costs. The IDB, which has provided the bulk of the financing for FAES, found in a multi-country study of social funds that temporary jobs were a good expedient in a recession, but did not solve structural poverty problems. The main impact came from delivering services to and improving the living conditions of the poor (Goodman, et. al., 1997). In Haiti, however, the project was concentrated in the three provinces closest to the capital, while the other six provinces, with nearly half the population and the greatest concentration of poverty, received less than a third of the funds. The project was expected to benefit women, but poor gender analysis and the lack of a clear framework for incorporating gender concerns reduced its impact; women were not sufficiently involved in decision-making processes, did not take adequate advantage of training and business oriented services, and benefited only in limited numbers from the employment generated by subprojects. Nonetheless, despite its flaws, FAES did respond to the basic needs of at least some of the population, and might be worth maintaining, especially for short term emergency needs, provided it does not undermine the systemic institutional changes required over the longer term. Economic Management, 1994-1997 2.12 Five projects were approved after lending resumed in 1994; one was cancelled and two have closed. The outcomes of both were rated satisfactory (one highly so), but once more, both were emergency loans with little long term relevance; only the EERC, which provided quick-disbursing balance of payments support, was rated as having had substantial institutional development impact or likely sustainability, and that was for its role in setting the stage for a medium term economic policy reform program, an objective that was subsequently abandoned. A proposed follow-on project (ERC2), never accepted by the government, was seen as too hasty a push for structural adjustment and 13. Oxfam observed, in a letter dated June 8, 1996 commenting on the draft CAS, that FAES sometimes inadvertently undermined community organization by working with groups formed specifically to respond to it, but without roots in the community. OED's study of social funds (World Bank, OED, 2001) concludes that social funds may be justified in an emergency situation, but that they can undermine local government, and, in the longer run, it is essential to deal with the institutional weaknesses of the country rather than to bypass them. 9 privatization. Negative perceptions about the Bank's role in this connection dominated the Bank's image in Haiti to the detriment of other projects (World Bank, OED, 1998).14 The second Technical Assistance credit, intended to support ERC2, was approved in FY97, but was never ratified by Parliament and was cancelled after 18 months. 2.13 The Employment Generation project (FY96) was an emergency social safety net operation, prepared and activated in three months, which generated 320,000 short term jobs. Its outcome was rated satisfactory, but IDI was rated modest, and sustainability unlikely, and the 1996 CAS noted the need to reassess the role of public employment generation projects in a medium term poverty strategy.' 5 The project made a concrete attempt to ensure a more equitable distribution of benefits between men and women; the latter accounted for 21 percent of those employed under the project, compared to the 20 percent target. Human Resources 2.14 The Bank undertook four education projects in Haiti in the 1970s and 1980s. A fifth project, approved in FY91 and signed five days before the overthrow of President Aristide, never became effective and was cancelled in FY93. Preparation of another project was suspended in 1998. The first two projects were rated satisfactory, having achieved most of their physical objectives, but they failed to achieve a consensus for educational reform. The outcomes of Education 3 (FY83-88) and Education and Training 4 (FY85-92) were both rated unsatisfactory, with negligible institutional development impact and unlikely sustainability. Despite these four projects, as well as sizeable contributions from other donors, Haiti has the lowest enrollment and literacy rates in the western hemisphere, and among the lowest in the world. The principle of free and compulsory basic education for all children, enshrined in the constitution, has never been put in practice. Reflecting the long history of government neglect, the country's education system relies heavily on private and religious organizations, which account for 80 percent of enrollment. Regulation of the private sector is weak, and the quality of private education is dismal. School fees can take as much as 15 percent of the annual income for the poorest fifth of the population, and the dropout rate is high. The government spends very little, only 2 percent of GNP, on education, and even that amount is used inefficiently. Primary net enrollment is estimated at 64 percent, but this masks a much worse situation in rural areas, where only 23 percent of children go to school. Net enrollment in secondary school is under 20 percent. 2.15 Haiti's per capita public expenditure on health is the lowest in the LAC region, and health conditions in Haiti are among the worst in the western hemisphere. The emphasis is on curative rather than preventive care, with little focus on reproductive and 14. The post-conflict report referred to this type of problem as the "folly of conventional wisdom"-pushing for the full menu of adjustment reforms to the extent of being counter-productive in post-conflict settings (e.g. Rwanda, Uganda). 15. Oxfam asserted that this project actually had a negative impact on longer term development, because its short term jobs drew people away from productive activities, such as harvesting (Oxfam letter, June 8, 1996). 10 child health care, despite high rates of infant and maternal mortality. Access to services is problematic for half of the population. Donor-supported immunization programs have not been effective, as evidenced by recent outbreaks of polio and measles. The first health project, approved in FY90, and suspended between 1991 and 1995, although unsuccessful on the whole, had one satisfactory component: it strengthened the TB program, with very effective use of NGOs. The project empowered people at the local level to make decisions, which is unusual in Haiti. Contracts with NGOs were well- specified and easy to monitor. The AIDS component was less successful; national coordination is weak, with no strategic view, and contracts with NGOs were not clearly specified.1 6 The least successful aspects of the project concerned institutional strengthening (little progress with cost recovery), the primary health care system (investments were supply driven and not utilized), and small components in reproductive health and essential drugs. Sustainability is questionable, particularly in view of the lack of institutional development, although the TB program has continued with funding from other donors. Disbursements were suspended in late 2000 following the failure of the government to submit a financial audit report, and the project closed as scheduled on March 31, 2001 with US$1.34 million undisbursed. The Bank is currently working under a PHRD grant to help produce a strategic plan for the battle against HIV/AIDS. Rural Development and Environment 2.16 Haiti's agricultural sector has been declining for many years, the result of neglected rural infrastructure, weak research and extension, poorly defined land tenure, limited access to credit, and under-investment in human capital. These factors, along with high population growth, have contributed to shrinking parcel sizes; complex, informal tenure systems that discourage long term investment; and cultivation of marginal lands with steep slopes, encroaching on forests, destroying watersheds, and aggravating the tenuous environmental situation. The share of imported food is rising. A series of agricultural and forestry/environmental projects has failed to stem the decline. 2.17 Of the four projects that have been rated, only one-Post-Hurricane Agricultural Rehabilitation (FY81 )-was rated satisfactory, although the completion report noted that its findings reflected impressions, not quantitative results, and concluded that "Though project benefits may not have been optimal, the project lessened the harmful effects of Hurricane Allen on agriculture." A small forestry project (FY82) failed to develop viable approaches for large-scale reforestation programs and had no impact on the deteriorating forest and wood situation. The outcomes of two rural development projects were rated unsatisfactory, with minimal increase in production, costs higher than projected and a negative economic rate of return; underestimated institutional risk; defaults on audits; and inadequate cost recovery and maintenance. The follow on project for forestry and environmental protection focused on the development of appropriate technology, greater 16. The government states that the prevalence of HIV/AIDS has fallen. Bank staff, however, say that both the data and their interpretation are still open to question. UNAIDS is working in Haiti to clarify the issue. 11 participation by NGOs, and small pilots, but it was approved in FY92, shortly before Bank activities were suspended, and was eventually cancelled. 2.18 The Forest and Parks TA project (FY97, the last approved for Haiti), was a key initiative in environmental management and a pilot for rural development. The project tested innovative ways to deal with governance, including devolving responsibility for national parks to local management committees and contracting agricultural extension and technical support to NGOs. In addition, the project funneled a large component through FAES for use by local governments in buffer zones around the national parks. Unfortunately, the security situation has precluded supervision of the project, so the effectiveness of these components is not known. Moreover, in addition to the problems of financial management and uncertain security common to all projects in Haiti, this project was undermined by a failure to reduce the demand for fuelwood; household energy is a major issue that has yet to be adequately addressed. The project was suspended by a new Minister, who appointed a commission to evaluate it, and by the Bank as part of the portfolio-wide suspension. The remaining funds were cancelled in June 2001. 2.19 The Biodiversity Enabling Activities Project, funded by a small GEF grant, sponsored workshops to raise conservation awareness in Haiti. However, disbursements are tied up by problems with arrears and the failure to submit a financial audit report, and action on the main product, a national biodiversity strategy and action plan, has halted. Infrastructure 2.20 Deterioration of infrastructure is at critical levels in Haiti; this problem has been particularly acute since maintenance fell off in the late 1980s. The Port-au-Prince Water Supply project (FY89-99) achieved much of its physical target, but few of its institutional development objectives. Unaccounted for water exceeds 55 percent of the total; much is consumed through illegal connections or is not metered, and severe rationing continues. OED noted that the project should have been conditioned on private management. Of the four evaluated power projects, all contributed to the expansion of generating capacity, but none were able to stem losses, which amounted to 55 percent in 1998. All had procurement problems. Institutional objectives were never achieved; privatization was suggested in the mid-1980s, but although Power 5 (FY89) was restructured in 1996 to permit privatization of Electricit6 de Haiti, government support has wavered. The outcome of a port project (FY81) was rated unsatisfactory, and the international port of Port au Prince remains the most expensive in the Caribbean. The National Port Administration was set for privatization, but progress has been halted. An Industrial Restructuring and Development project (FY90) was revised in 1997 to support the privatization of state owned utilities, but ultimately failed. Technical work for preparing major state enterprises (electricity, telephone, water, port and airport) for privatization was completed, but the lack of political will at the highest level and the weakness of the judicial framework blocked any progress. 2.21 Eight road projects have been approved since 1956, the bulk of the funds for the National Road, RN1. Nonetheless, it still takes 7 -8 hours to cover the 250 km. between Port au Prince and Cap Haitien, and 55 km. are nearly impassable. The EU, IDB and 12 bilateral donors have also committed major amounts for roads, with similar results. Transport 7 (FY87-96) ignored experience from previous projects, particularly with respect to institutional risk; rehabilitated roads deteriorated due to lack of maintenance and overloaded vehicles. Since the late 1980s, Haiti has had no regular maintenance program. The more relevant Road Maintenance and Rehabilitation project (FY95) was suspended for a time because of misprocurement and diversion of funds to road works not included in the project. Other problems include failure to submit audits, cost overruns due to continuing deterioration of roads, and long delays in clearing imported equipment through customs. Institutional reforms, including the reorganization of the ministry and the establishment of a road fund for maintenance, have not been implemented. The project was cancelled in June 2001, and sustainability of its achievements is unlikely. Project Management Units 2.22 PMUs are used in all projects in Haiti, since it has proven impossible to implement projects without them. The Road Maintenance and Rehabilitation project was managed under contract to a foreign firm brought in after the discovery of misprocurement; this is one model for dealing with weak governance. Other projects have generally used local staff in their PMUs, arguing that this does provide some capacity building. The unusually stable PMU in the health project contributed to its partial success. Although the use of PMUs brings into question the sustainability of project benefits and illustrates the great need for institution building in the country, the nearly constant state of crisis and recurring instability in Haiti have blocked any longer term strategy to reduce dependence on PMUs. IFC and MIGA 2.23 Haiti has been a member of IFC since 1956; since January 2000 an investment officer in the Dominican Republic has also covered Haiti. Over the last fifteen years, IFC's total investment in Haiti amounted to US$ 0.6 million. The difficult investment climate in Haiti has limited private investment and opportunities for IFC assistance. Under such constraints, IFC focused on developing opportunities with the help of technical assistance. In 1994, IFC began advising the government on privatization. Strategic reviews were undertaken of nine public sector companies ranging from banks to infrastructure and manufacturing companies, but only two transactions went forward. 2.24 In FY98, IFC utilized Trust Funds to perform a pre-feasibility study on the practicability of establishing a commercially oriented micro-finance lending institution. Based on the positive outcome of this study, IFC brought together domestic and foreign parties with a foreign technical partner to form Micro Credit National (MCN). The US$0.4 million equity investment in MCN (FY98) has been disbursed. While no formal review has yet been done, available indicators show that MCN will surpass development impact and financial expectations. It also had a positive demonstration impact, as evidenced by the recent announcement by one of Haiti's larger commercial banks of plans to commence micro-credit operations. Portfolio performance has been satisfactory. [... middle sections omitted for long document ...] 79 Attachment 3 OED Response to Government Comments OED appreciates the government's constructive and substantive comments on the CAE. They were taken into account in the current version. The term "failed state," to which the borrower took exception, was eliminated from the text: it was meant to convey the inadequacy of Haiti's post-conflict classification and the deep-seated, historical roots of the current instability. The government accurately pointed out that efforts were made to reform institutions after 1994. The CAE acknowledges this effort, but notes that the donors resumed traditional types of assistance at levels that the country could not absorb effectively given governance and institutional constraints. The current text now includes a suggestion that the Bank should carry out an Institutional and Governance Review to help in the reform process. With respect to the table in Annex A, referred to by the government (page 71), as is standard OED practice, this table has been removed from the final version of the report. Finally, the CAE retains the suggestion that should lending resume, it should be on a small scale. This is not intended to protect the Bank from financial risk. Rather, it aims to avoid a repetition of the past unsatisfactory record. 81 Bibliography Dayton, Julia. 1998. "World Bank HIV/AIDS Interventions: Ex-ante and Ex-post Evaluation." World Bank Discussion Paper No. 389. June. Dollar, David. 2000. "Governance and Social Justice in Caribbean States." Development Research Group, The World Bank. May. Discussion draft. Elliot, James (USAID), Gilles Gavraux (CIDA) and Naveen Sarna (World Bank). 1997. "Public Sector Investment Program." Haiti Public Expenditure Review Note No. 2. Draft, March. Energy Sector Management Assistance Program. 1991. "Haiti: Household Energy Strategy." UNDP/World Bank, Report No. 143/91. December. Faria, Angelo (IMF) and Paul Moreno-Lopez (IDB). 1997. "Aggregate Public Finances and Current Expenditures." Haiti Public Expenditure Review Note No. 1. Draft, March. Goodman, Margaret, Samuel Morley, Gabriel Siri and Elaine Zuckerman. 1997. "Social Investment Funds in Latin America: Past Performance and Future Role." Inter-American Development Bank, March. Government of Haiti, Ministry of Economy and Finance. 1997. "Project Implementation Problems and Public Expenditures: Managing Development Assistance in Haiti." January 26. International Monetary Fund. 1999. "Haiti: Staff Report for the Article IV Consultation." IMF Staff country Report No. 99/105. September. Michailof, Serge, Markus Kostner, and Xavier Devictor. "Post-Conflict Recovery in Africa: An Agenda for the Africa Region." Africa Region Working Paper Series. Forthcoming. Wallen, Georgia A. G. 2000. "An Inquiry into IDA Experience in Poor Performer Countries from 1994-2000." Draft, prepared for OED IDA Review. August. World Bank reports: 1985. "Haiti: Policy Proposals for Growth." Report No. 5601-HA. June 10. 1987. "Haiti: Public Expenditure Review." Publication No. 6113. January. 1988. "Economic Recovery in Haiti: Performance, Issues and Prospects." Report No. 7469-HA. December 23. 1990. "Haiti: Updating Economic Note." Report No. 8396-HA. April 17. 1991. "Haiti: Agricultural Sector Review." Report No. 9357-HA. April 30. 82 1991. "Haiti: Restoration of Growth and Development." Report No. 9523-HA. May 20. 1991. Operations Evaluation Department. "Program Performance Audit Report: Haiti-Economic Recovery Program." Report No. 10080. November 15. 1996. "Country Assistance Strategy." Report No. 15945. August 13. 1998. "Haiti: The Challenges of Poverty Reduction" (two volumes). Report No. 17242-HA. August. 1998. Operations Evaluation Department. "The World Bank's Experience with Post-conflict Reconstruction." (Five volumes). Report No. 17769. May 4. Volume I: Synthesis Report and Volume V: Desk Reviews of Cambodia, Eritrea, Haiti, Lebanon, Rwanda, and Sri Lanka. 1998. Wider Caribbean Financial Sector Review: Increasing Competitiveness and Financial Resource Management for Economic Growth." Report No. 17556-LAC. May 26. 2000. "Cost and Financing in Basic Education in Haiti." Draft sector note. 2000. "Infrastructure Sector Review." Draft. May 2. 2001. Operations Evaluation Department. "Social Funds: A Review of World Bank Experience." October.