Haiti
Ex Post Assessment of Longer-Term Engagement

EXECUTIVE SUMMARY Haiti’s 2006 and 2010 Fund-supported programs started under very different circumstances but shared the main objectives of preserving macroeconomic stability and creating the conditions for a sustained growth takeoff through structural reform. The 2006 program started as Haiti was making progress toward macroeconomic stabilization. Reducing inflation and avoiding fiscal dominance of monetary policy were top priorities. The 2010 program started in the aftermath of a devastating earthquake. It faced the challenges of dealing with a huge inflow of aid and scaling up public investment. Both programs aimed to foster reforms to address long-standing governance and transparency concerns, as well as improving revenue mobilization to increase fiscal space and reduce fiscal vulnerabilities. Growth was to rise to 4-6 percent, an ambitious objective given Haiti’s long history of serious fragility with near zero or negative growth. There were notable successes during the two programs. Despite a series of shocks, inflation remained in single digits and international reserves increased by more than expected, which helped to limit exchange rate volatility in the context of the large aid inflow. Program performance criteria effectively eliminated central bank credit to the government, thus reducing risks of fiscal dominance. Structural reforms also advanced, particularly during the 2006 program which coincided with the HIPC process. Revenue administration improved, with tax revenue rising steadily over the two programs. These achievements were not negligible given Haiti’s history.

Abstract

EXECUTIVE SUMMARY Haiti’s 2006 and 2010 Fund-supported programs started under very different circumstances but shared the main objectives of preserving macroeconomic stability and creating the conditions for a sustained growth takeoff through structural reform. The 2006 program started as Haiti was making progress toward macroeconomic stabilization. Reducing inflation and avoiding fiscal dominance of monetary policy were top priorities. The 2010 program started in the aftermath of a devastating earthquake. It faced the challenges of dealing with a huge inflow of aid and scaling up public investment. Both programs aimed to foster reforms to address long-standing governance and transparency concerns, as well as improving revenue mobilization to increase fiscal space and reduce fiscal vulnerabilities. Growth was to rise to 4-6 percent, an ambitious objective given Haiti’s long history of serious fragility with near zero or negative growth. There were notable successes during the two programs. Despite a series of shocks, inflation remained in single digits and international reserves increased by more than expected, which helped to limit exchange rate volatility in the context of the large aid inflow. Program performance criteria effectively eliminated central bank credit to the government, thus reducing risks of fiscal dominance. Structural reforms also advanced, particularly during the 2006 program which coincided with the HIPC process. Revenue administration improved, with tax revenue rising steadily over the two programs. These achievements were not negligible given Haiti’s history.

Introduction

1. An assessment of the Fund’s engagement with Haiti is timely. Since joining the Fund in 1953, Haiti has participated in 28 Fund-supported programs, covering 42 of the past 60 years. Most recently, Haiti received financial support through a Poverty Reduction and Growth Facility (PRGF) arrangement (2006-2010) and an Extended Credit Facility (ECF) arrangement (2010-14). Under the Fund’s policy on Longer-Term Program Engagement (LTPE), an Ex-Post Assessment (EPA) is required for Haiti before the authorities can request a new Fund arrangement. This is the first EPA report prepared for Haiti.

2. In this context, the EPA aims to provide a frank review of progress during the 2006 and 2010 programs, and a forward-looking assessment that takes into account lessons learned. It begins by providing the background and context for the two programs. It then discusses their overall objectives; how the programs were designed to achieve these objectives; and program performance. It concludes with some lessons, including for Haiti’s future engagement with the Fund.

Table 1.

Haiti: Arrangements with the Fund, 2006-2014 1/

(In thousands of SDRs)

article image
Sources: Mona Database, SPR; FIN.

Under the new architecture for low-income country lending, PRGF arrangements were converted into ECF arrangments on January 7, 2010. Note that the normal cumulative limit for total outstanding concessional credit under PRGF and ECF arrangements is 300 percent of quota, net of scheduled repayments. The Fund approved at the time of the third, fourth, and sixth reviews of the 2006 program access augmentations of 20 percent, 30 percent, and 80 percent of quota, respectively.

Background and Context for The Programs

A. Background

3. Haiti is characterized as a “fragile state” by various institutions.1 As of 2004, Haiti’s real per-capita income was more than 20 percent lower than in 1990 and one of the lowest in the world. This dismal performance, even by the standard of fragile states, reflected political strife, withdrawal of external economic support, violent crime, frequent natural disasters, and weak infrastructure and institutions, as discussed in numerous studies (see, for example, World Bank, 2011, and Acemoglu and Robinson, 2012). Haiti has consistently scored poorly on widely-used indicators of perceptions of corruption, ease of doing business, and global competitiveness. As discussed later on, Haiti’s difficult political context has hampered the approval and implementation of economic and financial structural reforms.

A01ufig01

Real Income per Capita

(1990=100)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

Source: IMF World Economic Outlook database.Note: LAC denotes Latin America and the Carribbean; LIC denotes lower-income countries. Interquartile range denotes middle 50 percent of distribution.

B. Context for the 2006 PRGF/ECF-Supported Program

4. Beginning in 2004, against the backdrop of challenging political circumstances, the Haitian authorities began the implementation of economic policies that reduced fiscal dominance of monetary policy and set the stage for the approval of the PRGF arrangement. A key achievement was the elimination of central bank financing of budget deficits, a key source of inflation in previous years, made possible through government spending restraint and revenue increases. Inflation declined, the currency stabilized, and growth resumed, albeit at a modest pace—an average of 2.0 percent during 2005-06. The Fund supported Haiti with policy advice and financial assistance in the context of a Staff Monitored Program (April-September 2004) and two Emergency Post-Conflict Assistance (EPCA)-supported programs (October 2004-September 2006).2 This set the stage for the Fund’s approval of the PRGF arrangement in November 2006, which sought to consolidate the stability gains achieved.3

C. Context for the 2010 ECF-Supported Program

A01ufig02
Note: Figure re ports worst 10 earthquakes since 1900 based on International Disaster Database (EM-DAT), and three-letter International Organization for Standardization (ISO) country codes.

5. The context for the 2010 program was the aftermath of a devastating earthquake and a large inflow of aid. The January 12, 2010 earthquake caused unprecedented loss of life and destruction of physical capital. Before the earthquake, the political and security situation had been gradually improving and foreign investors’ interest in some areas (including tourism and textile manufacturing) had begun to materialize. After the earthquake, total donor pledges of aid for 2010-12 amounted to US$8.1 billion (37 percent of 2010-12 GDP) (IMF, 2010), of which US$1.2 billion (18 percent of 2009 GDP) corresponded to debt relief, which significantly reduced Haiti’s external indebtedness. The Fund approved a new three-year ECF arrangement on July 21, 2010, along with full relief on Haiti’s outstanding liabilities to the Fund, of about SDR 178 million (equivalent to US$268 million) under the Post Catastrophe Debt Relief Trust (PCDR).4

Program Objectives and Design

A. Objectives

6. While the two programs shared the same overarching objective—promoting macroeconomic stability and achieving faster and sustainable growth to support poverty reduction—their very different starting conditions implied different policy priorities. The 2006 program started as Haiti was making progress toward macroeconomic stabilization, and building on this progress, including by avoiding fiscal dominance of monetary policy, was a key priority. In contrast, the 2010 program started in the aftermath of a devastating earthquake with a huge, but uncertain, expected inflow of aid. Mitigating the associated risk of a loss of competitiveness through exchange rate appreciation and “Dutch disease,” with a decline in private sector activity not directly related to reconstruction, as in the export sector, was a priority, including through addressing longstanding structural weaknesses in electricity supply and beyond.5 The common challenge facing both programs was the need to address weaknesses in governance and improving the business environment.

The 2006 PRGF/ECF-Supported Program

7. The 2006 program aimed at solidifying macroeconomic stability, including by further reducing inflation and maintaining fiscal policy discipline, and providing the conditions necessary for private-sector growth. It thus sought to consolidate the achievements of the 2005-06 SMP and EPCA programs, and to provide a macroeconomic framework for the authorities’ poverty reduction strategy and for needed accompanying international assistance. The broad objectives were as follows.

  • Monetary policy. The program targeted a reduction of inflation from double digits (an about 15 percent in 2005) to single digits (7 percent by 2009) by restraining money growth, ensuring zero central bank financing of the budget, and keeping the policy rate positive in real terms. It also called for a continued steady increase of international reserves.

  • Fiscal policy. The fiscal strategy was to keep the budget deficit in the 2-3 percent range, increase fiscal revenues from 10 to 12 percent of GDP to gain room for social and infrastructure spending; improve public financial management; and make public enterprise operations more transparent and efficient. Given debt sustainability concerns—the 2006 debt sustainability analysis assessed Haiti as being at high risk of debt distress—the program also prohibited external borrowing on non-concessional terms. It also prohibited the accumulation of payment arrears on external debt.

  • Growth. The program aimed for growth of 4 percent per year from 2007 onwards. The staff report explained that “The key objective is a sustainable economic growth of 4 percent” (IMF, 2006, p. 1) and clarified that the “growth objective is conservatively set at 4 percent, a rate similar to that achieved during past periods of stability,” which referred to the 1995-2000 period where growth averaged 3.8 percent (p. 7).

The 2010 ECF-Supported Program

8. The 2010 program sought to buttress the authorities’ post-earthquake reconstruction and growth objectives, ensure basic functioning of the treasury and payments system, and deal with a large but uncertain inflow of aid. The broad objectives were as follows.

  • Fiscal policy. The 2010 program was more flexible regarding the fiscal policy stance than the 2006 program had been. It allowed for larger fiscal deficits and planned a significant increase in public investment (from 10.3 percent of GDP in 2009 to 21 percent of GDP by 2011), financed largely by donor contributions and external (including Petrocaribe) concessional financing. This required strengthening spending execution capacity, which was already weak before the earthquake. Another immediate priority was to contend with the destruction of the tax administration building and information systems, the death of many senior officials, and damaged tax records. Longer-term objectives were to further raise central government revenue (from 11.2 percent of GDP in FY2009 to 13.1 percent of GDP by FY2013) through a reform of the tax system and the building of modern and efficient tax and customs administrations.

  • Monetary policy. The program aimed to keep inflation in single digits, reducing it from an estimated 8.5 percent in 2010 to 7.0 by 2013, by restraining monetary growth. It also aimed to further raise Haiti’s international reserves (from 2.9 months of imports in FY2009 to 3.4 months in FY2013) to help the central bank smooth the impact on the exchange rate and the economy of large expected aid inflows. Given the large aid inflow already mentioned, the risk of a sharp real appreciation and “Dutch disease” was seen as significant (IMF, 2010).

  • Growth. The program projected growth of 4.6 percent during 2010-14 and 5-6 percent over the medium to long term.6 The program request staff report recognized that “[t]he authorities’ objectives to sustain an annual rate of growth of 5-6 percent … are ambitious in light of historical performance” (IMF, 2010, p. 4).

B. Program Design

9. The design of the programs’ quantitative and structural reform targets was broadly consistent with their objectives.

Quantitative conditions

10. The two programs were similar in terms of quantitative conditions—performance criteria and indicative targets—designed to promote macroeconomic stability.

11. To contain inflation to single digits, both programs set quantitative limits on central-bank financing of budget deficits, and on monetary aggregates. Quantitative limits (ceilings) were set on net domestic assets of the central bank; net central bank credit to the central government and to the rest of the nonfinancial public sector; net domestic banking sector credit to the nonfinancial public sector; and base money. Both programs also set targets (floors) on net international reserves to promote reserve accumulation from low levels.

12. At the same time, neither program explicitly targeted the overall fiscal deficit. No quantitative targets were set for the overall (or primary) budget deficit. In particular, while limits were set on the portion of the budget deficit financed by the central bank, externally-financed deficit spending (on concessional terms) was not subject to program conditionality.7 As discussed later on, such externally-financed deficit spending—mostly financed by inflows from Venezuela—grew considerably, particularly during the 2010 program.

13. Moreover, reflecting uncertainty regarding the substantial post-earthquake spending needs and revenue availability, the 2010 program was less stringent than the 2006 program regarding some fiscal policy targets. While the ceiling on net domestic credit to the central government was a quantitative performance criterion for the 2006 program, it was subject to a milder form of conditionality (indicative target) for the 2010 program. Providing more flexibility regarding net domestic credit to the central government was also motivated by supporting the development of the domestic banking sector and the issuance of the recently created T-bills as an alternative to central bank financing.

Structural conditionality

14. The two programs were less similar regarding the design of structural conditionality, reflecting differences in the focus of their respective structural reform objectives.

15. The 2006 program placed greater emphasis than the 2010 program on central bank and financial sector policy reforms. About 44 percent of the 2006 program conditions were in this area, compared to only 20 percent for the 2010 program. In terms of specific plans, the 2006 program aimed to strengthen the central bank through the resolution of the troubled commercial bank it had acquired (to address conflict of interest concerns given the central bank’s role as both owner and supervisor); recapitalization of the central bank; the central bank’s withdrawal from non-core activities (divestiture from a telecom company, TELECO); and legislative passage of a new banking law.

A01ufig03

Structural Conditionality by Topic

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

16. Nonetheless, much of the two programs’ conditionality was in the structural fiscal area, with a focus on revenue administration and public financial management measures. Fiscal reforms comprised 45 and 80 percent, respectively, of the 2006 and 2010 program structural conditions. Specific measures included the following (a full list of program measures is included in the Appendix).

  • Revenue administration. Both programs aimed to increase the revenue-to-GDP ratio to provide more room for infrastructure and social spending, and to reduce debt vulnerabilities. For the 2006 program, conditions included expanding the use of the central taxpayer file to include all taxpayers; legislative passage and implementation of the customs code; strengthening customs checkpoints; maintaining strict control over the granting of tax exemptions; and continuing full implementation of a flexible mechanism for setting petroleum product prices at the pump to ensure that they move in line with corresponding prices on the international market. For the 2010 program, planned measures focused on rebuilding a modern and efficient tax administration; planning a comprehensive tax reform; combating fraud; and broadening the tax base by reducing exemptions. (At the same time, the program accommodated a fuel price freeze, which led to costly shortfalls in fuel-related tax receipts.)

  • Public financial management. Both programs aimed to enhance budgetary management, transparency, and governance. Specific measures planned for the 2006 program included setting up an effective debt management unit (DU); and more comprehensively and transparently reporting budget operations, including the reporting of resources from the Petrocaribe agreement with Venezuela and their use to finance public investment and social projects.8 For the 2010 program, specific plans included strengthening the debt unit; submitting a public debt law to parliament that would establish a sound legal and institutional framework; setting up a macro-fiscal policy unit to strengthen budget analysis and forecasting; and establishing a full Treasury Single Account—a key prerequisite for centralized oversight and control over government cash resources—by end-FY2014.

17. The number of structural conditions per review in the two programs was broadly in line with that in other Fund programs with fragile states. The number of structural conditions per review averaged 8, which is in line with the average for other fragile states—9 per review (IMF, 2011b). In terms of front-loading, the number of conditions in the first review of both programs was above the average, at 17 and 11 conditions in the PRGF and ECF-supported programs. Each program also contained a total of about 8 prior actions. This front-loading and the use of prior actions—measures needed before the Fund can approve a program or the completion of a program review—helped address implementation risks flagged by staff.

Technical Assistance

18. Both programs benefited from technical assistance (TA) tailored to their reform agendas, and the level of TA commitment rose significantly during the 2010 program. In terms of composition, TA support had a much greater focus on fiscal policy during the 2010 program than in the 2006 program, reflecting differences in the programs’ reform agendas already mentioned. TA visits covered tax policy, revenue administration, financial sector development, exchange rate policy, legal frameworks, and statistics. After the 2010 earthquake, there was a significant increase in TA commitment, reflecting the increased post-earthquake needs and donor contributions. The increase was far sharper than for other fragile states during the same period. A resident technical advisor was provided during the 2010 program to support public financial management reform. The programs were also supported by technical assistance aimed at strengthening the reliability and accuracy of key program data reported to the Fund, including by implementing the recommendations of safeguards assessments.

A01ufig04

Haiti: Evolution and Composition of IMF Technical Assistance by Department

(In person years of field delivery)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

Source: IMF Institute for Capacity Development Participant and Applicant Tracking System.

19. The TA support provided by the Fund was, in general, closely coordinated with other donors’ interventions. Coordination regarding TA with the various donors, was particularly important (and challenging) during the post-earthquake surge in TA support, and agreement of a coherent matrix of technical assistance consistent with the 2010 program objectives was reached. In the fiscal area, Fund TA focused on tax policy and revenue administration, budget preparation and planning, fiscal reporting, and Treasury management. The Fund also assisted in financial sector issues with the establishment of a Partial Credit Guarantee scheme (PCG Fund) in collaboration with the World Bank, the IDB and the U.S. Treasury; the development of a domestic Treasury bill market; central bank recapitalization; and the design of a framework for insurance regulation. Haiti’s development partners focused mainly on building capacity in complementary fields, such as payroll management, internal and external audits and procurement, anti-corruption and governance (World Bank), debt management (Canada, IDB, U.S. Treasury), the energy sector (World Bank, Canada), and restoration of IT systems (USAID). The European Union supported overall economic governance and public financial management reforms.

Program Performance

20. This section assesses macroeconomic and structural reform performance relative to the programs’ plans. The programs were broadly successful in promoting macroeconomic stability. However, growth, while positive, considerably disappointed relative to projections, and, while structural reform advanced, progress was slower than planned, reflecting a challenging political context and capacity constraints.

A. Macroeconomic Outcomes

21. The programs preserved the positive, if low, growth momentum achieved during the preceding staff-monitored and EPCA programs. GDP growth during the 2006 and 2010 programs (2007-14) averaged 2.2 percent per year, in line with Haiti’s 2.0 percent growth during 2005-06. With population growth of around 1.3 percent, this implied positive, if low, growth in income per capita. Accordingly, the level of GDP followed closely the (2.0 percent per year growth) trend set in 2005-06 until the 2010 earthquake, and recovered back to it by 2014.9

A01ufig05

Real GDP

(level; 2004 = 100)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

22. However, the sustained rise in growth projected by both programs did not materialize, and long-term growth assumptions were significantly marked down at the end of the second program. The 2006 program predicted 4 percent growth every year from 2007 onward, a doubling in growth relative to 2005-06. The 2010 program predicted average growth of 4.6 percent over 2010-14, and 5 percent over the long term. However, actual growth during the 2006 and 2010 programs averaged 2.4 percent and 2.1 percent, respectively. Moreover, staff downgraded the long-term growth assumption from 5 percent to 3 percent at the time of the seventh ECF review (IMF, 2014), noting that even this would constitute a major shift from the past.

23. Why did growth disappoint? Projecting economic growth in a fragile state is difficult, particularly after an earthquake. However, the baseline projections of the 2006 and 2010 programs can be considered too optimistic.10

  • Optimistic baseline. The growth forecast for the 2006 program of 4 percent was derived from the 1995-2000 period of stability, during which growth averaged 3.8 percent (IMF, 2006, p. 7-8). However, this average included the 9.9 percent growth rate of 1995, a bounce-back from the 11.6 percent contraction in 1994 associated with the U.S. military intervention that restored democratic government. Excluding this outlier, growth during 1996-2000 averaged 2.5 percent, close to actual performance during the 2006 program (growth during 2006-09 averaged 2.4 percent). The 2010 program projected even faster growth: 6-8 percent in the near term and 5 percent in the long term.11 The program projections were consistent with the view prevalent in the international donor and policy community at the time that Haiti could quickly shift to robust growth and “build back better” after the earthquake.12 Staff made it clear that the program growth projections were subject to considerable downside risks, virtually all of which subsequently materialized. In contrast, studies based on historical data on catastrophic events suggested the need for cautious growth assumptions regarding the pace of recovery following Haiti’s 2010 earthquake (Cavallo, Powell, and Becerra, 2010, for example).13

  • Lower-than-expected government investment. In the case of the 2010 program, the projections assumed a surge in government investment from 11 percent of GDP in 2009 to 21 percent of GDP in 2011. The actual build up in government investment was much more gradual (to 14 percent of GDP in 2011), reflecting, in part, absorption capacity constraints. The construction sector was not, and is still not, capable of executing all of the projects the various actors are ready to build. Coordination and prioritization problems also contributed to the reconstruction delays, as Laframboise and Loko (2012) explain (p. 25): “The biggest challenge to reconstruction, and absorption of the sizeable foreign financial assistance pledged, is weak capacity and the absence of central coordination … better coordination and prioritization may have resulted in faster reconstruction.” The disbursement of grants associated with public investment also undershot projections, reflecting both the absorptive capacity constraints already mentioned as well as delays in meeting donor disbursement conditions.14 In addition, a different-than-expected composition of aid played a role. In particular, part of the planned public investment increase was subsequently reclassified as humanitarian assistance, which, given its higher import content, could explain the smaller boost to growth.15

  • Weaker than assumed aid effectiveness. The program growth projections assumed a strong degree of aid effectiveness, and achieving this was a challenge, particularly after the 2010 earthquake. The staff report accompanying the request for the 2010 program explains (p. 10) that the “Medium-term projections assume that aid inflows are effectively spent and absorbed.” In practice, problems in donor coordination and aid effectiveness played a role in explaining the shortfall of growth relative to the assumed path (IMF, 2014). A number of donors also faced coordination issues with other donors, significant overruns of costs relative to plan, and disappointing outcomes, as a U.S. Government Accountability Office (2013) study entitled “Haiti Reconstruction: USAID Infrastructure Projects Have Had Mixed Results and Face Sustainability Challenges” illustrates.16 Aid disbursed by some large donors often had to pass multiple layers of sub-contracts and sub-grants, hampering aid efficiency (Ramachandran and Walz, 2012). Much of the post-earthquake aid also circumvented the Haitian public sector. While this reflected donors’ legitimate concerns regarding capacity constraints and governance, it also constitutes a missed opportunity for strengthening the effectiveness and legitimacy of the state—a necessary ingredient for sustainable development.

24. The authorities succeeded in containing inflation to single digits and rapidly built up reserves, particularly after the earthquake, which helped to limit exchange rate volatility in the context of the large inflow of aid. The food and fuel price shocks pushed inflation up in 2008, but prices then fell sharply in FY2009 in line with the drop in commodity prices. Monetary control contributed to containing low inflation. More than 90 percent of the program’s quantitative performance criteria were met in both programs.17 During the 2010 program, reserve accumulation exceeded (admittedly cautious) program targets by a wide margin, and this helped to mitigate exchange rate volatility in the context of significant aid inflows, large remittance inflows and an increased supply of foreign exchange associated with NGO activity in the country (IMF, 2010). Total reserves rose from about US$ 1 billion in December 2009 to (14.4 percent of GDP) to about US$ 2 billion by end-2014 (21.5 percent of GDP). The current level of international reserves provides Haiti a relatively good buffer in the short-term, but preserving external stability in the long term hinges on improving competitiveness and promoting exports.18

25. Nevertheless, the real effective exchange rate appreciated by about 23 percent during 2006-14, on the back of large remittances and external aid inflows, and improving external competitiveness through structural reforms remains critical. Staff analysis accompanying the 2007, 2010, and latest (March 2013) Article IV Consultations concluded that, on balance, the Haitian gourde was not misaligned with fundamentals, and that the economy experienced an equilibrium response to large external aid and private remittance inflows. Since 2010, the gradual real appreciation has consisted of moderate depreciation of the nominal effective exchange rate coupled with inflation above that of key trade partners. With the real exchange rate broadly aligned with the fundamentals (bearing in mind the limitations of the REER as a measure of competitiveness in the context of low income countries), enhancing competitiveness in Haiti is predicated on structural reforms to boost productivity through improving infrastructure and the business environment, as discussed later on.19

A01ufig08

Exchange Rates

(Jan. 2000 - Apr. 2014; index 2010 = 100)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

Sources: INS Database; National Authorities; and IMF staff estimates.

26. Haiti’s twin fiscal and current account deficits both remained below 3 percent of GDP for much of the 2006 program, but widened unexpectedly toward the end of the 2010 program. The 2006 program predicted a fiscal deficit (including grants) of 2.3 percent of GDP, on average, during 2006-09, and the actual deficit averaged 2.2 percent of GDP. During the 2010 program, the fiscal deficit was lower than predicted during 2010-11, owing mainly to the lower-than-expected public investment. However, the deficit unexpectedly increased to 7.1 and 6.3 percent of GDP in FY2013 and FY2014, respectively, compared to the initial (2010) projection of 4.2 and 4.1 percent of GDP, respectively. This deficit overshoot to unsustainable levels mainly reflected transfers to the public electricity company (EDH), which amounted to about 1.5 percent of GDP in FY2013, and the fuel retail price freeze that had a fiscal cost of almost 2 percent point of GDP. Much of the deficit is now financed by external (Petrocaribe) debt. Importantly, the programs’ conditionality did not include overall deficit limits, given the goal of avoiding any impediment to spending aid (in line with conditionality for other economies faced with such circumstances.)20 The counterpart of the rising fiscal deficit was a rising external current account deficit, which was largely financed by concessional (Petrocaribe) flows from Venezuela.

27. External debt followed a V-shaped trajectory—declining with debt relief, then rebounding with new Petrocaribe borrowing—and Haiti’s risk of external debt distress remains high. Haiti benefitted from significant debt relief during 2009-11, with external debt declining from 29 percent of GDP in 2008 to 9 percent in 2011. Debt has since rebounded to an estimated 23 percent of GDP in 2014 (without breaching any Fund program targets), and the baseline scenario of the latest Debt Sustainability Analysis (DSA) for Haiti (IMF, 2014) suggests it could reach 49 percent of GDP by 2024. The debt rebound, which contrasts with the more stable path expected earlier, reflects larger-than-expected Petrocaribe borrowing, which currently accounts for 84 percent of total external debt. Debt sustainability will be challenged given Haiti’s narrow export base and low government revenue.21 In particular, the difficult macroeconomic situation in Venezuela casts doubts on Petrocaribe’s sustainability (IMF, 2014). A sudden stop in Petrocaribe financing would cause a severe fiscal and balance-of-payments adjustment, compromising public investment and growth.

A01ufig09

Government External Debt

(Percent of GDP)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

Program response to macroeconomic shocks

28. The programs adapted flexibly to macroeconomic shocks that affected Haiti. The Fund approved at the time of the third, fourth, and sixth reviews of the 2006 program access augmentations of 20 percent, 30 percent, and 80 percent of quota, respectively, as authorities requested more resources to deal with food and fuel price increases, social unrest, hurricanes, and the January 2010 earthquake. Given an unexpectedly large revenue shortfall in 2009, in the context of the global financial crisis, additional fiscal space was provided to avoid cutbacks in priority spending by exceptionally relaxing the constraint on central bank financing.

B. Structural Reforms

29. Structural reforms advanced, but more slowly than expected, reflecting a complicated political situation, as well as capacity, ownership, and institutional constraints.

30. In terms of observance of the programs’ structural conditionality, performance was broadly satisfactory during the 2006 program, but became more challenging during the 2010 program. The share of structural conditions met on schedule declined from 61 percent under the 2006 program to 56 percent under the 2010 program (Appendix). This change in part reflected the post-earthquake environment. It may also have reflected the fact that the 2006 program coincided with the HIPC process (2006 decision point and 2009 completion point). In particular, the HIPC triggers were closely aligned with the structural benchmarks of the 2006 program, and, to reach the completion point, the program had to remain on track. Haiti’s performance nevertheless contrasts favorably with that of other Fund programs in fragile states, where only 33 percent of measures are generally met, whether on time, with delay, or partially (IMF, 2011b).22

31. The difficult political situation, as well as ownership issues, hampered the approval and implementation of reforms, and delayed the completion of program reviews, particularly during the 2010 program. The 2006 PRGF arrangement was extended at the fifth review until end-January 2010 to allow for the completion of the sixth review in the context of a heavy election agenda.23 Performance under the 2010 program was characterized by frequent delays in reviews and implementation, reflecting capacity and infrastructure constraints and at times political complications (Table 2). A slippage in the calendar for Presidential elections delayed the first review. A year-long political stalemate following the November 2010 general elections delayed the second and third reviews, which were combined. Meanwhile, the President’s lack of a majority in parliament hampered the approval of key economic and financial legislation. At the sixth review, the authorities requested an extension of the arrangement by one year to complete key reforms, particularly in the fiscal area. After the seventh review, the authorities requested another extension to provide additional time to implement actions on the fiscal and structural fronts before the eighth review. Program conditionality was also often modified to accommodate implementation challenges, but this did not always ensure the related measures were taken as planned. Ownership challenges played a role in delaying reforms, including, for example, the submission of the debt law to Parliament (Appendix). In turn, laws pending approval in Parliament include the Organic law for the Ministry of Finance (elevating the Treasury to a Directorate General for cash and debt management), and the public debt law.

Table 2.

Calendar of Program Reviews

article image

Progress and challenges in selected reform areas

32. Reforms in the fiscal area advanced, particularly on revenue administration, although many challenges remain. Tax revenue collection steadily improved and central government tax revenue rose from 10.2 percent of GDP in FY2006 to 12.7 percent of GDP in FY2013, despite the complex political and economic context. This achievement, which was supported by the implementation of program measures and substantial technical assistance, was nevertheless less than envisaged (the program had projected a revenue ratio of 13.1 percent by FY2013). Factors that complicated raising the revenue-to-GDP ratio further included a difficult political situation which hindered steps to broaden the tax base and tighten controls over the tax exemptions intended; the surge in fuel import prices and associated cuts in taxes, such as excises and customs duties; and the January 2010 earthquake, which undermined the ability of the tax and customs offices to perform basic functions. Moreover, Haiti’s level of government revenue remains low—in line with other fragile states, but below the Carribbean regional average. Significant initial steps were taken on tax policy, but much remains to be accomplished before a new tax code can be implemented.24

A01ufig10

Central Government Tax Revenue

(Percent of GDP)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

A01ufig11

Central Government Tax Revenue

(Percent of GDP)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

Source: IMF World Economic Outlook and staff estimates.1/ Fragile states.2/ Carribbean.

33. In the public financial management area, the authorities took steps toward establishing a TSA, although progress was significantly slower than assumed, in part reflecting resistance by key stakeholders. The 2010 program had planned the full implementation of a TSA by end-September 2014, which the authorities were unable to achieve given the political and governance environment. Setting up a TSA in such a timeframe is technically feasible, as was the case in Guinea, Mali, and Rwanda, for example. However, in Haiti, resistance by a number of key stakeholders held back timely TSA implementation. For example, the more than 20 line ministries each enjoy a high level of autonomy over the budgetary transfers they receive, and some have resisted TSA implementation, which would come with greater centralized control, as well as greater accountability and transparency. Frequent changes in the Minister of Finance (four changes since 2010) also delayed implementation. More explicit recognition of such challenges earlier on in the program would have led to a more realistic completion timetable. In the event, the related benchmarks were modified and lack of progress toward them played a role in the need to extend the 2010 program prior to the completion of the eighth review. The government’s current action plan aims to ensure the TSA becomes operational by end-September 2015. Making further progress on establishing the TSA will require the cooperation of the main stakeholders, including the central bank, the Treasury, and the line ministries, as well as continued capacity building supported by the resident TA advisor. Other PFM challenges include the timely approval of annual budgets. The budget for FY2013 was not approved until May 2014, more than 6 months after the start of the fiscal year. There was also little progress toward a medium-term budget framework, although a macrofiscal unit with the capacity to undertake budget forecasting and analysis was set up at the Ministry of Economy and Finance in early 2012.

34. Little progress was made toward resolving decade long weaknesses in the electricity sector, and the public electricity company (EDH) remains a fiscal burden. Only 42 percent of EDH’s electricity output is billed and less than a third is paid for. These decade long problems reflect fraudulent connections to the network, billing infrastructure weaknesses, and challenges in ensuring payment by users. A number of local government authorities have unpaid EDH bills, illustrating governance challenges.25 EDH, in turn, has unpaid bills vis-à-vis independent power providers (IPPs), creating a liability for the central government, while the IPPs have unpaid oil bills vis-à-vis the autonomous agency of the Ministry of Finance that manages Petrocaribe resources (BMPAD). EDH remains a major burden for the public finances, receiving fiscal transfers of 2.5 percent of GDP in FY2013, equivalent to 20 percent of government revenue (IMF, 2014). During the 2006 program, the authorities set up a monitoring mechanism for fiscal transfers to EDH, and prepared a strategy to increase electricity supply and EDH collection (together with the World Bank and IDB). During the 2010 program, the authorities signed a MOU with the IDB on the modernization of the energy sector and strengthen EDH’s management team. However, EDH’s financial situation remains unsustainable, and the low quality of electricity service has led many companies to produce their own electricity, often at high cost, which is a constraint on competitiveness and growth.

A01ufig12

Electricity Billing and Collection

(Percent of EDH electricity output)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

Note: Dashes indicate interpolation through available data.

35. Another fiscal risk relates to the tight regulation of fuel pump prices. In April 2008, in the context of rising commodity prices and riots, the authorities suspended the automatic adjustment of local currency domestic fuel prices to reflect international price movements. They restored price adjustments in October 2008, but froze prices again in March 2011 to avoid social unrest. The price freeze involved cutting fees and taxes collected on fuel products to offset rising imported fuel prices, and, in some cases, outright subsidies to fuel importers. The related fiscal losses reached 1.9 percent of GDP in 2013 at a time of rising international prices. The sixth review of the 2010 program underlined the need for domestic fuel price increases together with a public communication strategy and compensatory measures for the most vulnerable. To this end, the IMF provided technical assistance to reform fuel price subsidies and implement an automatic pricing mechanism. The World Bank provided assistance to design targeted assistance to vulnerable segments of the population. On October 10, 2014, the more than three-year-long fuel price freeze was finally lifted, with prices rising by 6-9 percent, equivalent to closing 22-62 percent of the gap between retail prices and their international levels, depending on the fuel product. Experience from other countries suggests that freezing fuel prices is a costly way of protecting real incomes of the poor, and continuing the transition toward more cost-effective alternatives is important, including through targeted cash transfers to poor families.26

36. Progress on reforms of the central bank and the financial sector was limited. During the 2006 program, the central bank disengaged from non-core activities as planned, including through the sale of its stake in TELECO, but its planned recapitalization did not occur.27 The authorities also started implementing the recommendations of the 2008 Financial Sector Assessment Program (FSAP), including by completing an assessment of options for the recapitalization of key commercial banks. They also submitted to parliament a draft banking law consistent with international standards. During the 2010 program, reforms were shaped by the post-earthquake emergency situation. The program supported the establishment of a PCG Fund and BRH’s decision to restructure bank loans.28 Implementation of the 2008 FSAP recommendations also advanced, including the strengthening of banking supervision, with TA from the IMF and the US Treasury. A new banking law was adopted by parliament in March 2012. The anti-money laundering and counter-terrorist financing (AML/CFT) law was approved and promulgated in November 2013, but a number of deficiencies in the AML/CFT framework remain. The financial situation of banks improved considerably relative to the post-earthquake situation. However, the legal frameworks for microfinance institutions and the insurance sector are still to be adopted by Parliament. More generally, while financial soundness indicators currently reveal no clear weaknesses, Haiti’s financial sector remains small, and business surveys suggest that limited access to finance, especially by small and medium enterprises, is one of the most problematic factors for doing business in Haiti (World Economic Forum, 2013). Bank credit to the private sector as of 2013 was only 18 percent of GDP, which, while comparable to other fragile states, was well below the median for emerging market and developing economies (34 percent of GDP).

37. More generally, there was little visible progress toward strengthening the business climate and competitiveness. Widely used survey-based indicators of ease of doing business, and perceptions of corruption and global competitiveness have their limitations, but the fact that Haiti’s ranking along these dimensions improved little during the two programs underscores the need for additional reforms. The increase in political stability after the low-point of 2004 did come with a rebound in the World Bank’s index of regulatory quality, which measures the ability of the government to formulate and implement sound policies and regulations. However, there has been little further improvement since. Business surveys suggest that limited access to finance, including by small and medium enterprises, inadequate supply of infrastructure, and corruption are the most problematic factors for doing business in Haiti (World Economic Forum, 2013). Haiti has not fared well in attracting FDI, which, at 1.4 percent of GDP in 2013, continues to be among the lowest in the region, and Haiti is highly dependent on remittances and foreign assistance (remittances and official transfers amounted to 27 percent of GDP in 2013) (IMF, 2014). Haiti does not yet fully reap the opportunities provided by the U.S. HOPE/HELP preferential trade access initiative. Textile exports account for about 95 percent of Haiti’s exports, almost entirely to the United States, and taking advantage of the still largely unused quotas would further support export growth (the shares of quotas still unfilled range from 45-95 percent, depending on the type of apparel product).29

A01ufig13

Worldwide Business Climate Indicators

(Haiti’s gap relative to highest-ranked country; -100 to 0)

Citation: IMF Staff Country Reports 2015, 004; 10.5089/9781484305751.002.A001

Sources: Word Bank Worldwide Governance Indicators; and Doing Business; World Economic Forum.Global Competitiveness Report.Note: Figure based on Haiti’s percentile rank within continuous set of countries.

38. Overall, Haiti’s experience highlights the need for realistic expectations for how rapidly structural change and transition from fragility can occur. Assessing the feasible pace of institutional reform was difficult in Haiti’s context, particularly after the 2010 earthquake, but some of the planned reforms set an overly ambitious timetable. To better understand the need for realism regarding the pace of reform in fragile states, it is worth considering the World Bank’s 2011 World Development Report, which suggests (p. 10) that, even for the fastest transforming countries, improving institutional quality from the level of a country like Haiti to that of Ghana takes in the range of 15-30 years. While faster transition is possible, Collier (2007) estimates the probability of a sustained turnaround starting in any year to be just 1.6 percent, implying an average time of 59 years to transition from fragility. Relatedly, Pritchett, Woolcock, and Andrews (2013) find that rates of improvement in measures such as quality of bureaucracy and administrative capacity are typically very low in developing countries, even with substantial donor effort. They conclude (p. 1) that “countries like Haiti or Liberia will take many decades to reach even a moderate capability country like India.” The design, prioritization, and scheduling of structural conditionality over the course of a future Fund program with Haiti needs to take this into account, including by aiming for the adoption of second best reforms.

Strategy For Future Fund Engagement

39. Haiti could benefit from continued program engagement with the Fund to safeguard macroeconomic stability and promote reforms in the context of numerous near-term risks. A relatively stable economic environment is a key ingredient for supporting a sustainable transition from fragility. The macroeconomic stability gains achieved in the decade since 2004 cannot be taken for granted, as Haiti faces significant external and internal risks.

  • On the external front, a further deterioration in the macroeconomic situation of Venezuela (including in relation to decreases in oil prices) could result in a sudden stop in Petrocaribe financing, and slow progress in strengthening governance and transparency could increase donor fatigue and reduce external grants. Commodity price shocks could also be a source of stress, particularly to the energy bill (electricity subsidies and fuel price stabilization costs).

  • On the domestic front, governance problems, and political resistance to improving tax administration, as well as renewed spending pressures (including for Petrocaribe-related spending) could worsen an already unsustainable fiscal outlook and reduce the room to react to shocks, such as natural disasters. Reform momentum could also slow if political tensions resume in the context of long-delayed congressional and municipal elections, and a reduction in the MINUSTAH stabilization force.

In this potentially fragile context, and in the presence of the requisite balance of payments problem, the Fund’s advice under the framework of a program could play a more effective role, if there is sufficient domestic ownership, than under a surveillance-only relationship in preserving macroeconomic stability and ensuring progress on structural reforms. The ECF would remain a suitable Fund engagement instrument.

40. Fund program engagement with Haiti could help address its balance of payments problems and help to catalyze donor assistance, including for technical assistance, budget support, and project grants. Disbursement of donor aid is often conditional on good performance under a Fund-supported program. The role of Fund-supported policies in catalyzing international support is illustrated by the increase in donor grants during the 2006 arrangement. By the same token, early in the 2010 program, disbursement of budget support was lower than planned, partly reflecting slow progress in key structural areas important to donors, including procurement and governance (IMF, 2013c). Future Fund engagement in Haiti should also involve continuing to work with donors to identify “quick wins” that can be embedded into the Fund program, and, if useful, the joint preparation of regular progress reports on capacity building. The continuing presence of the resident TA advisor would assist in the coordination of technical assistance and the transfer knowledge, and build capacity more intensively than with mission-based support.

41. A future Fund-supported program would need to take into account the lessons from the 2006 and 2010 programs. Specific recommendations include:

  • Building a realistic macroeconomic framework. Haiti’s experience highlights the importance of setting more realistic expectations regarding future growth. Building a future program’s macroeconomic framework on a cautious baseline scenario is particularly important given the need to address long-standing fiscal sustainability and vulnerabilities concerns. For example, a baseline projection of 2-3 percent growth, near the average observed during the two programs, would be a cautious starting point. The macro-fiscal unit at the ministry of economy and finance, put together with support from the EU and Fund staff, and operational since 2014, is well placed to contribute to building the macroeconomic framework.

  • Having a clearer fiscal anchor. The worrisome build-up of debt and fiscal vulnerability during the 2010 program implies the need for a thorough stock taking and the setting of explicit targets to anchor fiscal policy. Haiti’s debt risk arises not only from its level, but also the speed of increase. In retrospect, a debt limit should have been considered to constrain the pace of borrowing. Given Haiti’s high risk of debt distress rating, grant financing should be preferred over debt financing. Another concern going forward is on the composition of debt, with 84 percent of all external debt borrowed via PetroCaribe. Future Fund-supported arrangements should specifically address this risk. A clearer fiscal anchor could involve quantitative limits on either the primary fiscal deficit or on the stock of government debt. To illustrate the size of the sizeable fiscal adjustment needed, the latest DSA (IMF, 2014) provides estimates of the debt-stabilizing primary fiscal balance which are some 5 percentage points of GDP below the FY2013 level. The challenge is to balance the need for fiscal consolidation with Haiti’s large social and infrastructure needs, and to determine a feasible pace of adjustment consistent with the social and political context. Guarding against the risk of a sudden stop in Petrocaribe financing by keeping a larger buffer stock of deposits constituted from fuel sales is also warranted.

  • Focusing on fewer key structural reforms with a realistic implementation timetable and well-coordinated TA support. Making progress on the main structural reforms initiated during the 2010 program is a priority. Setting up a TSA is essential for transparency of and centralized oversight and control over government cash resources. Making further progress on revenue mobilization is also important, given the need to increase fiscal space and reduce dependency on potentially volatile external financing. Decade long weaknesses in the energy sector and the associated fiscal risks need to be addressed.30 Improving the business climate and governance remains a key objective, including through strengthening property rights (cadastre), increasing the transparency of the procurement system, and anti-corruption measures. Effective implementation of the AML/CFT framework would support authorities’ efforts to combat corruption and tax evasion, and prevent financial sector abuses. In addition, the enactment of the legal frameworks for microfinance institutions and the insurance sector would contribute to the development of financial services. The challenge is to calibrate the pace of reform to ensure steady progress in priority areas while avoiding unrealistic expectations, given political and capacity constraints, which could lead to premature interruptions in Fund engagement and financial support from donors. Technical assistance should continue to be aligned with the program objectives, the authorities’ capacity, and commitment to implement reform. There is also room for further improving and strengthening TA coordination in the public finance area, including through the use of the TA matrix involving all donors and the authorities.

  • Risks, prior actions, and contingency plans. Given Haiti’s history with economic and political shocks, a future Fund-supported program could benefit from greater reliance on prior actions, especially for missed structural benchmarks. Recognizing that early successes are critical for advancing the reform agenda, a future program would need to strike a balance between giving the authorities the benefit of the doubt in terms of policy commitments and learning from past experience. A number of Haiti’s structural challenges are not new and a clear commitment to reform through prior actions could strengthen policy credibility. At the same time, given the risk of reform fatigue and implementation slippages, and to reduce the need for repeated program extensions, contingency plans for a possible re-evaluation of the modes of engagement, including, for example, a switch to a SMP, are needed.

42. Building a national consensus regarding the risks of the status quo and the need for reform is required for the success of a future Fund-supported program and of Haiti’s transition out of fragility. Reaching a consensus among key stakeholders in the public and private sectors would strengthen ownership and improve implementation of reforms. Strong outreach by Fund mission teams and the resident representative to local civil society, administrators, diverse political and business leaders, and opinion makers could help build support for the reform program and increase understanding of the role of the Fund in the reform process. To achieve stronger ownership, the authorities should ensure any future program is presented to political leaders and the entire Cabinet prior to the signature of the letter of intent (LOI), as well as to Parliament and civil society. In this context, the resident representative should continue playing an important role.

Table 3.

Haiti: Selected Economic and Financial Indicators, 2005-2014

article image
Sources: Ministry of Economy and Finance; Bank of the Republic of Haiti; Fund staff estimates and projections.

Staff assume a range of 3.5-4 percent and a point projection of 3.5 percent for FY2014 growth.

Figures for FY2014 are based on the recently-passed budget and the outcome for the first half of the fiscal year.

Debt ratios differ slightly from those in the DSA given the use of average, instead of end-of-period, exchange rates.

Excludes central bank repurchase operations in FY2013.

In percent of exports of goods and nonfactor services. Includes debt relief.

Includes SDR allocation as both an asset and liability.

Excludes gold; includes transactions related to BRH repurchase operations.

Appendix I. The Views of the Authorities

The authorities welcome the opportunity to comment on the lessons learned from the Fund’s engagement in Haiti over the past decade. The authorities broadly agree with the analysis, the conclusions, and the recommendations of the EPA report regarding the main features of a potential successor arrangement under the ECF. They made the following comments on the programs and the EPA report for clarification and emphasis. The authorities have also expressed interest in a future Fund-supported program, which would enable them to deal with short-term risks associated with balance of payments and fiscal deficits, and to accelerate the implementation of structural reforms conducive to job creation and growth.

Both Fund-supported programs helped to reduce inflation and to end direct financing of the fiscal deficit by the central bank, but this stabilization will be jeopardized without a substantial acceleration in growth. As indicated in the report, growth was below program targets. It will be difficult to maintain social and macroeconomic stability given the country’s fragile situation, unless the next program leads to poverty reduction and to a much faster increase in income and jobs. These challenges require additional quality investments in public infrastructure, social spending, and structural reforms.

International reserves were up significantly during the 2010 program, which helped limit currency appreciation amid large inflows of foreign aid. However, structural reforms need to be sped up in order to enhance competitiveness. Little progress has been made in increasing access to credit and reducing the excessive cost of electricity, which impedes investment and weighs on competitiveness, particularly for small and medium-sized enterprises. More must be done to tackle these problems in an effort to promote growth and ensure the increase and diversification of exports.

As the report points out, the pressing need at the moment is to reduce fiscal vulnerability. However, the pace and composition of the necessary adjustment must be calibrated so as to prevent the dampening of growth or curtailment of the reform and modernization process, which carries a substantial fiscal cost. The authorities are deeply grateful for the assistance they receive from Venezuela, but are aware that they need to diversify their sources of financing and strike the right balance between loan use and debt sustainability, especially as aid inflows decline. In this context, they also recognized the need to reduce subsidies to the energy sector and other current expenditures without altering public infrastructure and social spending, such as cash transfers. Increasing government revenue is also an important measure, particularly by strengthening tax administration and aligning fuel prices at the pump more closely with international reference prices. However, the pace of the adjustment must be gradual in light of the social and political context. The need to develop innovative lending mechanisms was stressed. Fund support is sought to identify mechanisms that would ease the current constraint so that the country may invest in a better future.

The success of a future Fund-supported arrangement depends largely on improved ownership of the reform program by key stakeholders. Decisions regarding the content of the 2006 and 2010 programs were often rushed, with insufficient consultation of key stakeholders beyond the Ministry of Economy and Finance and the central bank. The new program should therefore be presented to political leaders, to the entire cabinet, to Parliament, and to civil society before the letter of intent is signed. To promote ownership of the reforms, it will be especially important to improve communication with those line ministries that are directly involved. Coordination with and between donors must also be improved so that aid is predictable and the structural conditionality of the new program is better aligned with the authorities’ objectives.

Aid-financed spending should be incorporated into the budget more effectively and chosen in closer consultation with the authorities to ensure that it is more reflective of the government’s priorities. The current aid process has shortcomings in a number of respects, a notable one being the authorities’ weak participation in selecting aid-financed projects as well as frequent disbursement delays. The timely provision of more reliable information on aid related expenditures would be essential to strengthen budget execution control and monitoring. Moreover, additional budgetary support is preferable over project grants in order to better integrate international assistance into the national process. A review of aid management mechanisms is deemed vital to strengthening national institutions and government coordination.

The technical assistance played a pivotal role in supporting the reform program, but more resident technical advisors are needed for capacity-building purposes. There is a pressing need to achieve a “critical mass” of qualified staff by improving the conditions of employment and stepping up technical assistance in light of the brain drain from the public sector to the private sector, nongovernment sector, or international institutions. The presence of resident technical advisors is in certain circumstances preferable particularly in the public finance area while short-term technical assistance missions may be more optimal in order to enhance the capacity building process. It is expected that these advisors will in fact facilitate knowledge transfer. At the same time, the authorities and their partners should look into establishing an effective program for retaining qualified executives.

Specific comments on the ex post assessment report

Program growth projections were not overly optimistic. It was realistic to predict an upturn in activity after the dearth in aid and the economic decline that preceded the 2006 program, and following the earthquake of 2010. The worse-than-expected results can also be attributed to factors beyond the authorities’ control, notably delays in disbursing aid and approving certain laws, and natural disasters. The reform of the fiscal framework and the strengthening of economic governance are necessary to guarantee the smooth implementation of the economic development program supported by the financial program agreed with the IMF.

The potential use of a staff-monitored program (SMP), as suggested in the report with respect to contingency planning, would have a number of drawbacks. An SMP would not give donors and development partners a clear indication of the quality of economic policies and would not serve as a catalyst for the necessary aid from donors.

The report could have emphasized the rapid turnover of Fund staff, which has sometimes led to inconsistent economic policy advice and also complicated program development and monitoring, especially in combination with an accelerated turnover at the ministerial level. Greater efforts to ensure that the IMF country team comprises more experienced executives, particularly in the context of fragile states, as well as better management of the team’s institutional memory, are deserving of further exploration.

The ex post assessment should cover not only performance as far as program implementation is concerned, but also the consistency over time of the proposed economic policy measures and programs. Also, focus should be placed on aligning macroeconomic objectives with policies designed and implemented in the sectors, taking account of economic structures characterized by serious imbalances, in order to achieve growth targets. The coordination necessary for orderly implementation (completion and financing schedules) is equally important, in respect of which macroeconomic policies must be more extensive. For example, what expenditure policy would have greater effectiveness on growth? What steps can be taken to ensure that a policy designed to increase investments does not end up deteriorating the external imbalance more rapidly? A growth diagnostics approach would have to be taken to identify the leverage effects and harness them more effectively. Budgeting needs to be strengthened and the strategic objectives and priorities of all stakeholders in the expenditure chain need to be determined. The macroeconomic objectives must be broken down into targets and sector policies associated with monitoring and assessment indicators.

The structural reforms should be organized according to a somewhat more realistic schedule that takes account of institutional challenges and prevents the authorities’ from having to request extensions. Greater prominence should be given to communication concerning reforms and training for the participants involved.

Other imperatives include enhancing forecasting instruments and economic analysis methods, and sustaining analysis work in support of the framework. This improvement in forecasting tools should be coupled with a strengthening of the statistical system for macroeconomic policy design and monitoring, notably with a view to improving national accounts, consolidating public finance data, and enhancing monitoring of public and private investments.

Appendix II. 2006 Program Structural Reforms

article image
article image
Source: Monitoring of Fund Arrangements (MONA)Note: M=Met; MD=Met with delay; NM=Not met; W=Waved; MOD(Mod)=Modified; PA=Prior action; SB=Structural benchmark; SPC=Structural performance criteria

Appendix III. 2010 Program Structural Reforms

article image
article image
Source: Monitoring of Fund Arrangements (MONA)Note: CAN=Cancelled; M=Met; MD=Met with delay; NM=Not met; W=Waved; Mod=Modified; PA=Prior action; SB=Structural benchmark; SPC=Structural performance criteria

References

  • Acemoglu, Daron, and James Robinson, 2012, Why Nations Fail: The Origins of Power, Prosperity, and Poverty (Crown Publishing Group, New York).

    • Search Google Scholar
    • Export Citation
  • Cavallo, Eduardo, Andrew Powell and Oscar Becerra, 2010, “Estimating the Direct Economic Damages of the Earthquake in Haiti,” The Economic Journal, 120 (August), pp. 298312.

    • Search Google Scholar
    • Export Citation
  • Collier, Paul, 2007, The Bottom Billion: Why the Poorest Countries are Failing and What Can Be Done About It (New York: Oxford University Press).

    • Search Google Scholar
    • Export Citation
  • Ho, Giang, and Paolo Mauro, 2014, “Growth: Now and Forever?IMF Working Paper No. 14/117 (Washington).

  • International Monetary Fund, 2006, “Haiti: Request for a Three-Year Arrangement Under the Poverty Reduction and Growth Facility—Staff Report; Staff Press Release on the Executive Board Discussion; and Statement by the Executive Director for Haiti” (Washington).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2010, “Haiti: 2010 Article IV Consultation and Request for a Three-Year Arrangement Under the Extended Credit Facility - Staff Report; Staff Supplement; Public Information Notice on the Executive Board Discussion; and Statement by the Executive Director for Haiti” (Washington).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2011a, “Haiti: First Review Under the Extended Credit Facility Arrangement—Staff Report; Staff Statement; Press Release on the Executive Board Discussion; and Statement by the Executive Director for Haiti” (Washington).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2011b, “Macroeconomic and Operational Challenges in Countries in Fragile Situations” (Washington).

  • International Monetary Fund, 2012, “Staff Guidance Note on the Fund’s Engagement with Countries in Fragile Situations” (Washington).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2013a, “2013 Low Income Countries Global Risks and Vulnerabilities Report” (Washington).

  • International Monetary Fund, 2013b, “Energy Subsidy Reform: Lessons and Implications” (Washington).

  • International Monetary Fund, 2013c, “Haiti: 2012 Article IV Consultation and Fifth Review Under the Extended Credit Facility” (Washington).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2014, “Haiti: Seventh Review Under the Extended Credit Facility, Requests for Waiver of Nonobservance of Performance Criterion, and Modification of Performance Criteria-Staff Report; Press Release; and Statement by the Executive Director for Haiti” (Washington).

    • Search Google Scholar
    • Export Citation
  • Laframboise, Nicole, and Boileau Loko, 2012, “Natural Disasters: Mitigating Impact, Managing Risks,” IMF Working Paper No. 12/245 (Washington).

    • Search Google Scholar
    • Export Citation
  • Pritchett, Lant, Michael Woolcock, and Matt Andrews, 2013, “Looking Like a State: Techniques of Persistent Failure in State Capability for Implementation,” The Journal of Development Studies, 49:1, pp. 118.

    • Search Google Scholar
    • Export Citation
  • Rajan, Raghuram G., and Arvind Subramanian, 2011, Journal of Development Economics 94 (2011) pp. 106118.

  • Ramachandran, Vijaya, and Julie Walz, 2012, “Haiti: Where Has All the Money Gone?,” Center for Global Development Policy Paper 004, May 2012 (Washington).

    • Search Google Scholar
    • Export Citation
  • U.S. Government Accountability Office, 2013, “Haiti Reconstruction: USAID Infrastructure Projects Have Had Mixed Results and Face Sustainability Challenges,” GAO-13-558 (June 18, 2013).

    • Search Google Scholar
    • Export Citation
  • World Bank, 2011, World Development Report 2011 on Conflict, Security and Development (Washington).

1

There is no universally accepted definition of “fragility,” but countries with fragile situations typically have severe institutional weaknesses that have impaired the pace of economic development over time, as discussed, for example, in the Fund’s 2013 Low Income Countries Global Risk and Vulnerabilities Report IMF (2013a). For the purposes of the analysis in this report, fragile states are defined as the set of 27 economies considered to be fragile states by IMF (2013a) (based on a score of 3.2 or less in the World Bank’s Country Policy and Institutional Assessment).

2

The United Nations Stabilization Mission in Haiti (MINUSTAH) was established in mid-2004 with the objective of restoring security.

3

The authorities’ success at maintaining macroeconomic stability and implementing structural reforms also allowed Haiti to qualify for the cancellation of US$1.2 billion of debt under the Enhanced Heavily Indebted Poor Countries (HIPC)/ Multilateral Debt Relief Initiative (MDRI), which occurred in 2009.

4

In line with an agreement with Fund staff, PCDR resources were allocated to the financing of projects in infrastructure, social housing, the financial sector, and institutional capacity building. About 38 percent of the PCDR resources were used during 2010-14.

5

The staff report accompanying the 2010 program request explained (IMF, 2010, p. 20) that “[t]he competitiveness impact of the real appreciation associated with the foreign inflows should be mainly addressed through structural policies aiming at reducing supply-side bottlenecks, expanding the export base and improving the business environment.” Aid inflows can have unintended adverse effects on a country’s competitiveness through a number of theoretical channels discussed in the literature, including by raising overall wages, causing real exchange rate appreciation, and reducing tradable competitiveness (see Rajan and Subramanian, 2011, for some empirical evidence of such effects at work in low-income countries).

6

This projection included a contraction of 8.5 percent in 2010 and a rebound of 9 percent growth in 2011.

7

Limits were also set on domestic arrears of the central government, external arrears accumulation, and non-concessional external debt contracted or guaranteed by the central government.

8

Under the Petrocaribe agreement, which Haiti ratified in May 2007, Venezuela provides Haiti and other members with concessional lending through oil product sales. (Petrocaribe has a total of 17 members, plus Venezuela.) Upon the delivery of oil products from Venezuela, they are sold at the full purchase price to domestic purchasers, and the proceeds are paid to an autonomous agency of the Haitian Ministry of Finance, the Bureau de Monétisation du Programme d’Aide au Développement (BMPAD). BMPAD then pays part of the proceeds to Venezuela in cash, with the rest financed over long maturities. Specifically, at prices above US$100 per barrel, 40 percent of the bill is paid in cash, and 60 percent is financed at a 1 percent interest rate, and a 25-year maturity. At prices between US$80 and US$100 per barrel, 50 percent of the bill is paid in cash, and 50 percent is financed. At prices between US$50 and US$80 per barrel, 60 percent of the bill is paid in cash, and 40 percent is financed. Meanwhile, the oil proceeds not paid to Venezuela immediately are set aside for public spending. The budget includes information on Petrocaribe-financed spending, including on investment projects but their use does not follow the rules applicable to treasury resources, including on procurement (IMF, 2014).

9

As of 2014, output is estimated to have returned to potential compared with an output gap of -5.3 percent in 2010.

10

Here, it is also worth acknowledging that optimistic growth forecasts are by no means unique to Haiti’s programs. As Ho and Mauro (2014) highlight, Fund growth forecasts tend to more optimistic than warranted by past experience for many countries, particularly those about to enter an IMF-supported program.

11

Note that even after the initial contraction in growth in 2010 was revised from −8.5 percent (IMF, 2010) to −5.1 percent (IMF, 2011b), and a smaller bounce-back could have been expected, near-term forecasts remained in the 6-8 percent range.

12

See the 2011 World Economic Forum report (WEF, 2011) entitled “Private Sector Development in Haiti: Opportunities for Investment, Job Creation and Growth,” which argued (p. 7) that “The Dominican Republic, which is the other half of the island of Hispaniola, has experienced 6 percent GDP growth for the last two decades. With the right public policies and the positive engagement of the Haitian and international private sector, Haiti could experience the same – or even higher – levels of economic growth for the next decade or more.” See also IMF (2010) for a discussion of the March 2010 high-level UN “International Donors’ Conference Towards a New Future for Haiti” conference at which the government presented its reconstruction plan, aiming at “building a better Haiti.”

13

The shortfall of growth relative to projections had a negative effect on nominal government revenue, although the impact on the fiscal deficit was in part offset through lower-than-expected execution of public expenditure.

14

More generally, it is worth noting that while total aid flows during the first three post-earthquake years (2010-12) were huge—amounting to some US$7.5 billion (34 percent of 2010-12 GDP) (IMF, 2014, p. 34)--but below initial projections by about 4 percent of 2010-12 GDP. IMF (2010, p. 3) indicates donor pledges during 2010-12 of US$8.1 billion (37 percent of 2010-12 GDP).

15

A comparison of successive fiscal and balance of payments tables in the 2010 program staff reports indicates that the share of official assistance (net official transfers) spent on capital expenditure during 2010-12, initially projected at 74 percent of all official assistance (IMF, 2010), was subsequently revised down to 48 percent (IMF, 2014).

16

The report highlights the case of the number of houses to be built with U.S. assistance, which dropped from a planned 15,000 to 2,649, while the estimated number of beneficiaries declined from about 80,000 to 14,000, and costs per unit more than doubled. The report suggests that cost increases resulted from inaccurate original estimates and from the Haitian government’s request for larger houses with more improvements, while construction delays occurred due to difficulties of securing land titles and coordinating with other donors.

17

Note that the zero ceiling on the contracting or guaranteeing by the public sector of non-concessional external debt was missed in 2014 in the context of the contracting by the BRH of repurchase operations (repos) for reserve management.

18

Progress in reducing dollarization was limited. The share of foreign currency-denominated bank credit to the private sector declined from 52 percent in 2006 to 43 percent in 2014. However, due to the shallow financial sector, private sector bank deposits remain dominated by deposits in foreign currency (almost 60 percent). Reducing dollarization will require a long track record period of political and economic stability, exchange rate flexibility and greater access to Gourde-denominated financial services.

19

Haiti’s exchange rate system is classified as a crawl-like arrangement, under which the exchange rate is market determined, with the central bank intervening to smooth out excess volatility. However, improvements in the operations of the foreign exchange market are needed. Haiti’s foreign exchange market remains underdeveloped and not integrated and, in spite of Fund-provided TA, efforts to improve its functioning have been limited by the lack of an auction interbank market. As a result, associated structural benchmarks were not met or had to be modified.

20

The 2010 program also included a floor on poverty-reducing expenditures (defined as domestically-financed spending in health, education, and agriculture). While the floor was not always observed, such spending increased from 1.0 percent of GDP in FY2009 to 3.9 percent of GDP in FY2013.

21

Haiti’s exports amount to an estimated 18.2 percent of GDP in 2014. Note that Haiti’s total government debt is estimated at 22.9 percent of GDP as of 2014, of which 20.1 percentage points represents external debt, and that the ratio of total external debt to liquid international reserves has risen from 48 percent in 2010 to an estimated 93 percent in 2014. It is worth acknowledging that Haiti’s programs did include structural benchmarks related to strengthening debt management, but that these did not provide an explicit anchor for ensuring medium-term debt sustainability.

22

Differences in the difficulty of the conditions could, at least theoretically, explain this relatively favorable completion rate. Ideally, a comparison of Fund programs would be assessed on the basis of the difficulty of the conditions. However, such an assessment is highly context-specific and not amenable to a cross-country exercise.

23

On October 30, 2009, the Haitian Senate voted former Prime Minister Pierre-Louis and her cabinet out of office. A new government was inaugurated on November 11, 2009. The extension to May 31, 2010 provided additional time for the completion of the sixth and final review, and for making the final disbursement under the PRGF arrangement.

24

The technical assistance program aimed at the adoption of a new tax code continues, although progress on the project has been limited over the last year.

25

The central government is current on the payment of EDH bills. For an overview of the Haiti’s troubled electricity sector, see the staff report for the seventh review of the ECF program (IMF, 2014).

26

See IMF (2013b). The under-taxation of fuel favors diesel, used essentially by power generation plants and increasingly by wealthier households, who consume more energy.

27

Vulnerabilities also remain in the area of foreign reserve management, and no progress had been made towards to the adoption of the IFRS. Although formal recapitalization of the central bank did not occur as planned, this became less needed after the central bank’s financial position strengthened with increasing interest receipts on government obligations and the rise in international reserves. In addition, the BRH has not yet published its financial statements for the year ended September 30, 2013, and the external audit for that year is pending.

28

All performing loans that became non performing because of the January 12, 2010 earthquake were eligible to be restructured and subject to a provision of 2 percent instead of 50 percent. Moreover, those loans could qualify for the PCG Fund.

29

See Box 1 in IMF (2014) for details. The Haiti Economic Lift Program (HELP) Act, approved by the U.S. Senate on May 6, 2010, allows for unprecedented access for Haitian textile and apparel to the American market until 2020 (it supersedes the Haitian Hemispheric Opportunity through Partnership Encouragement Act, HOPE). There is also space to increase value added, as two-thirds of apparel exports (by volume) were in the categories of cotton t-shirts and underwear, instead of higher value items, such as wool suits.

30

In particular, a future program should aim at having a strategic plan designed with donor assistance to strengthen the financial position of EDH and improve access to and availability of electricity. This will require that the authorities address deep-rooted governance problems associated with billing and collection and the inadequate institutional framework. Contracts with IPPs could also be readjusted. Also, more flexibility in retail fuel prices, building on the October 2014 adjustment, is warranted to avoid additional shortfalls in fuel-related tax revenue.