Republic of Madagascar
Ex Post Assessment of Longer-Term Program Engagement
Author: International Monetary Fund
Publication Date:
11 Oct 2005
eISBN:
9781451825367
Language:
English
Keywords:
ISCR; CR; exchange rate; economic growth; reform effort; government ownership; implementation capacity; structural reform agenda; public expenditure; private sector; government service; monetary policy
Download PDF (2.5 MB)

This paper discusses key findings of the Ex Post Assessment of Longer-Term Program Engagement for Madagascar. The paper focuses on performance during the programs supported by the 1989 and 1996 Enhanced Structural Adjustment Facility programs, and the 2001 Poverty Reduction and Growth Facility. Despite nearly continuous involvement by the IMF, other international financial institutions, and bilateral donors, economic progress has been slow. Only the most recent years have witnessed inroads into poverty reduction of some significance. However, the country’s growth base remains narrow, and its institutional framework and governance weak.

Abstract

This paper discusses key findings of the Ex Post Assessment of Longer-Term Program Engagement for Madagascar. The paper focuses on performance during the programs supported by the 1989 and 1996 Enhanced Structural Adjustment Facility programs, and the 2001 Poverty Reduction and Growth Facility. Despite nearly continuous involvement by the IMF, other international financial institutions, and bilateral donors, economic progress has been slow. Only the most recent years have witnessed inroads into poverty reduction of some significance. However, the country’s growth base remains narrow, and its institutional framework and governance weak.

I. Introduction

1. During the 1970s, Madagascar followed an inward-looking development strategy, which resulted in declining growth, as well as large economic and financial imbalances. Real GDP grew at an average annual rate of less than 2 percent and real per capita GDP declined steadily. Administrative controls multiplied in the economy and several sectors were nationalized. Since the early 1980s, several adjustment programs, supported by Fund arrangements and World Bank resources, succeeded in reducing the economic and financial imbalances, including through extensive debt relief, but not in remedying the structural rigidities.

2. In 1988, the authorities adopted an ambitious strategy to move toward an open and market-oriented economy. A 10-month stand-by arrangement with the Fund served as the bridge between the first annual arrangement under a SAF, and arrangements under a new ESAF, starting in 1989.

3. This report presents an ex post assessment of Madagascar’s long-term program engagement with the Fund. Given the important shift in the economic course in 1988–89, those years are taken as the starting point. The report focuses on performance during the programs supported by the 1989 and 1996 ESAFs, and the 2001 PRGF. Under the 1989 ESAF, only the amounts under the first and second annual arrangements were fully disbursed (Table 1). Under the 1996 ESAF, two disbursements were made under the first annual arrangement. No understanding could be reached on a second arrangement. The program was extended by one year. It eventually expired in November 2000, with the completion of just two of the three envisaged annual arrangements. The 2001 PRGF was the only program in the period under review, under which all disbursements were made. Meanwhile, Madagascar reached the HIPC decision point in December 2000 and the completion point in October 2004. Fund and Bank endorsed the full PRSP in 2003.

Table 1.

Madagascar: History of IMF Lending Arrangements Since 1987

(In thousands of SDRs)

article image
Source: IMF, FIN Department.

4. Despite nearly continuous involvement by the IMF, other international financial institutions (IFI), and bilateral donors, economic progress has been slow. Only the most recent years have witnessed inroads into poverty reduction of some significance (Appendix I). External shocks and political crises frequently interrupted stabilization and growth efforts, and stalled reforms. Structural reform has accelerated with the onset of the PRSP process around 2000. The country’s growth base remains narrow, and its institutional framework and governance weak. Madagascar remains one of the poorest countries in the world (Table 2), and today’s real GDP per capita still stands 50 percent below its 1960 level.2

Table 2.

Madagascar: Growth Comparison (1983–2004)

article image
Source: World Economic Outlook, IMF.

5. Poverty remains prevalent in Madagascar and poverty reduction has been slow. Poverty has dramatically increased since the 1960s and, in particular, during the period of inward-looking policy actions in the 1980s. Structural reforms since the mid-1990s are having a beneficial, though still limited, impact on poverty levels (Table 3 and Appendix I). In 2004, an estimated 75 percent of the population still lived in poverty. This percentage is markedly lower for the urban population, which has benefited more from economic growth (approximately 40 percent). The levels of most social indicators and of access to basic utilities remain low. Net primary school enrollment has increased from 48 percent to 70 percent since the mid-1990s. The percentage of electricity and water connections has doubled in the same period, but remains low (respectively, 19 percent and 27 percent). On the other hand, the infant mortality has hardly come down and remains at 86 deaths per 1,000 births, and immunization rates are only at 36 percent of children. About 80 percent of the rural population still has no reliable transport services, and of these, one-third has no road access at all.

Table 3.

Madagascar: Social Indicators—Comparison with Averages for Sub-Saharan Africa Countries

article image
Source: http://devdata.worldbank.org/

II. Economic and Political Developments

6. Economic growth picked up quickly in the wake of the implementation of market-oriented reforms in 1988–90 (Figure 1). For the first time in almost a decade, real per capita GDP growth was positive (Figure 2). However, the episode was short-lived. A combination of policy slippages and adverse weather conditions dampened growth already in 1990, and the political crisis of mid-1991 paralyzed economic and social life entirely.

Figure 1.
Figure 1.

Madagascar: Real GDP Growth and Major Events, 1988–2004

(In percent)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

Figure 2.
Figure 2.

Madagascar: Real GDP Per Capita Growth, 1980–2004

(In percent)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

7. Political and social instability kept the country in its grip until roughly 1996–97. The political turmoil found its origin in a growing popular resentment at the government’s liberal measures—including price and trade liberalization, privatization, civil service reform, and tax and customs reform—combined with a gathering desire for political change. It resulted in massive general strikes and social unrest. Most economic reforms, started during 1988–90, were either suspended or reversed.3 Real GDP growth stalled, inflation soared, internal and external balances widened, and payments arrears accumulated. Real per capita GDP declined by almost 9 percent in 1991, and close to 4 percent in 1992. Work started on a new constitution in 1992, and a new president took office in 1993.

8. The political situation remained unstable and structural reform lost momentum. The rift between the new president, who was elected on a platform of populist measures, and the government, which intended to pursue a drastic reform agenda, continued to dominate political and economic life.

9. In mid-1994, the government prepared a new medium-term policy statement, with support from the IMF and World Bank. This statement provided a framework for economic liberalization and structural reforms in several key areas. However, the political climate did not yet prove right for implementation. The impeachment of the president by parliament in 1997 brought an end to this long period of political instability and opened the way for reform.

10. Since 1997, the record in terms of economic stabilization and growth has been more positive than before, even though this period has not been without crises either. Between 1997 and 2001, economic growth averaged 4½ percent, inflation remained subdued, exports doubled, and international reserves rose sharply, even in the face of adverse exogenous shocks (three cyclones hit Madagascar in early 2000). Structural reform started to take root, laying the foundations for a period of accelerated growth.

11. These positive developments were abruptly interrupted by the 2002 political crisis. Despite strong suspicion of vote-rigging, official results denied victory to the opposition candidate (Mr. Ravalomanana) in the 2001 presidential elections, forcing him into a second round. When he proclaimed himself president in February 2002 after the second round, the country had effectively two administrations because the incumbent (Mr. Ratsiraka) refused to accept defeat. His supporters organized a blockade of Antananarivo, and economic life in the capital city stalled nearly completely. The rest of the country was also affected, but less severely.

12. The crisis was short-lived,4 but its economic and social consequences were devastating, underscoring the persistent fragility of Madagascar’s economy. Real GDP contracted by 13 percent in 2002 and inflation soared to double-digit levels. Per capita real GDP dropped by 15 percent—washing away the gains made in the previous five years—and income poverty increased by as much as 6 percent, affecting 75 percent of the population at the end of 2002. Recovery from this political crisis was quick and economic growth resumed its earlier trend in 2003. However, recovering the losses registered in income poverty and other social indicators is much slower.

III. Assessment of Fund Involvement

A. Objectives of the Programs

13. Given Madagascar’s low level of economic development, Fund involvement sought to achieve and maintain an environment of macroeconomic stability and external viability, conducive to sustained growth. The three programs also supported the government’s structural reform agenda aimed at reducing poverty. To achieve these objectives, the first program focused specifically on domestic and external liberalization. The second program aimed more at a structural transformation of the economy in order to create conditions for sustained economic growth and the reduction of poverty by rebuilding a capable public administration, and addressing corruption and weak governance. The third program supported the government’s objectives—in line with the goals set out in the I-PRSP and, later on, in the PRSP—of strengthening economic growth while improving the distribution of its benefits in order to secure a permanent reduction in poverty.

14. The results of the Fund’s involvement in achieving these objectives have been uneven.5 Fund involvement has contributed to macroeconomic stabilization, strengthening the resilience of the economy and external viability. It has also been fairly successful in bringing about some structural reforms (external sector liberalization, financial sector reform, and exchange rate regime), while in others, notably, tax policy, tax and customs administration, public expenditure management, privatization, and civil service, progress has been limited. This section analyzes progress in meeting the respective program objectives.

B. Areas of Strong Progress

Macroeconomic stabilization

15. Inflation has been brought down, but remains volatile and above the objectives. Inflation was brought down from 23 percent in 1987 to 11 percent in 1990 (Figure 3).6 However, it veered out of control during the subsequent political crisis. Since 1996–97, inflation has, on average, been below 10 percent. The surge during the 2002 political crisis was controlled rapidly. Despite these achievements, inflation remains volatile and above the objectives (to some extent, due to exogenous factors, such as the rise in agricultural and oil prices in 2004).

Figure 3.
Figure 3.

Madagascar: Inflation, 1988–2004

(Annual average, in percent)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

16. Economic growth has picked up since the late 1990s, but this positive track record is still short. Since 1997—with the exception of 2002—growth has been in the 5 percent to 6 percent range, close to program targets (Figure 4), but still below the rates targeted in the PRSP to achieve significant poverty reduction.7 It seems, however, that a basis for sustainable growth has been laid. Recovery from recent shocks (natural shocks in 2000, 2001, and 2004, and the political crisis of 2002) has been faster than before.

Figure 4.
Figure 4.

Madagascar: Real GDP Growth, 1988–2004

(In percent)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

17. Significant progress has been made in achieving the external sector objectives, which focused on improving external viability. In the early 1990s, international reserves were drawn down, bottoming out at SDR 42.7 million by the end of 1994—just over one month of imports (Figure 5). In recent years, they have been buffeted by the weather, commodity price, and political shocks in 1998, 2002, and 2004. At end 2004, Madagascar’s gross reserves position had reached a historically high level of over SDR 300 million. However, this level is still below the set objective, and, given Madagascar’s sensitivity to shocks, it still provides only a limited cushion of around three months of imports.

Figure 5.
Figure 5.

Madagascar: Gross International Reserves, 1988–2004

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

18. Madagascar’s external debt situation has improved significantly. Its position could remain sustainable, provided sound macroeconomic policies are maintained and structural reforms continued, according to the HIPC completion point debt sustainability analysis. Since 1981, the country rescheduled its debt to the Paris Club creditors 11 times. With the latest round of debt relief, the NPV of debt to exports was expected to be reduced to a little over 150 percent in 2004 (IMF Country Report No. 04/406, December 1, 2004). Debt service to exports will be below 5 percent and debt service to revenue about 10 percent. Sensitivity analysis indicates that a sustainable debt position depends critically on achieving sustained GDP growth rates of at least 6 percent.

19. The external current account deficit remains wide and its size continues to fluctuate. Since 1996, current account deficits (excluding grants)—have been between 6 percent and 8½ percent of GDP—in all but one year (2001) (Figure 6). These deficits have been largely financed by aid inflows (mainly grants). The trade structure has changed tremendously during this period, with traditional exports fading in relative importance as the exports from the export processing zones (EPZ) and other exports have grown.

Figure 6.
Figure 6.

Madagascar: Current Account Deficit, 1988–2004

(In percent of GDP)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

20. The Malagasy franc has remained broadly competitive, as measured by the REER (Figures 7 and 8). Following a period of loss in competitiveness under a pegged exchange rate, with a widening gap between official and parallel exchange rates, widening external imbalances and trade restrictions driving trade increasingly into informal channels, the authorities decided to adopt a floating exchange rate regime in 1994. This regime has helped in maintaining the currency’s competitiveness.

Figure 7.
Figure 7.

Madagascar: Nominal and Real Effective Exchange Rate, 1988M1–2004M8

(Index, 1990=100)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

Figure 8.
Figure 8.

Madagascar: Real Effective Exchange Rates for Selected Countries (Sep 2002–Feb 2005)

(Index, Sept 2002 = 100)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

Structural reform

External sector liberalization

21. Initially, external trade liberalization was slow, with some reversals. In the early reform years, trade liberalization was considered essential to integrate Madagascar into the global economy. However, the strides taken were limited. Import tariffs were lowered from 80 percent to the 50–60 percent range. Some export taxes were abolished while others, such as on vanilla, remained in existence. The political and economic crisis of 1991 was marked by an intensification of restrictions on current account transactions.

22. Since 1996, trade liberalization was conducted in the context of the Cross-Border Initiative (CBI) and COMESA, and has made major strides. Import duty rates have been lowered, in line with the CBI guidelines. The weighted average rate is currently estimated at 18 percent (Country Report No. 04/403, 12/10/04). Export taxes and the export surrender requirement were fully eliminated in 1997. Since 2004, the country only has four tariff bands, and the two import taxes have been merged. Madagascar accepted the obligations of Article VIII in 1996.

Exchange rate regime and markets

23. The exchange rate mechanism, established in 1994, has operated satisfactorily, despite occasional problems at the level of the market infrastructure. Initially, the operation of the floating rate regime was undermined by the sluggishness of the unified interbank foreign exchange market. One large, state-owned bank, BTM (Bank for Rural Development) dominated the market, but was not very active given its distressed state. Only after BTM’s restructuring in 1996 did the market start to operate more smoothly. The system was also hampered by the fact that residents could transfer foreign exchange deposits among each other, thereby significantly reducing the volumes that went through the market. The introduction of the continuous quotation interbank market in 2004 seems to have improved the efficiency of the system.

Price liberalization

24. Price liberalization started in 1989 and is now almost complete. Many price controls were abolished between 1989 and 1995, and by now, most prices are market-determined. A significant achievement was the liberalization of domestic oil prices in 2004.

Financial sector reform

25. Following an unconvincing start in the early 1990s, financial sector reform gathered speed in the late 1990s. The privatization of two ailing state banks, BTM and BFV (National Bank of Commerce), had been part of the conditionality under Fund-supported programs since the late 1980s. However, outright privatization proved politically difficult and technically complex. Under the 1996 ESAF, intermediate steps, intended to first clean up these banks, were taken. Privatization finally took place between 1996 and 1999. The authorities have also permitted private domestic and foreign banks to operate in the country.

26. The supervisory framework for banking has been improving since the mid-1990s. The establishment of the independent Banking and Financial Supervision Commission in 1996 laid the basis for a modern supervisory framework. The alignment of the prudential supervisory framework with the Basel Core Principles in recent years has contributed to the soundness of the banking system (Table 4). While the banking system is sound, intermediation remains low, mainly because of other impediments to financial services (Table 5).

Table 4.

Madagascar: Financial Soundness Indicators, 1998–2004

(In percent, unless otherwise indicated)

article image
Source: Banking and Financial Supervision Commission, Central Bank of Madagascar (BCM).
Table 5.

Madagascar: Indicators of Financial Deepening Broad Money (M2), 1985–2003

(In percent of GDP)

article image
Source: World Economic Outlook, 2004.
Legal reform

27. The 1996 ESAF had a substantive legal reform component. The most critical achievement was the introduction of legal instruments for long-term land lease by foreigners, as well as procedures to effect a lease within 60 days. This, and other improvements to the legal framework and the work conditions and professional capacity of magistrates have improved the business climate and the rule of law in Madagascar.

C. Areas of Weak Progress

Macroeconomic stabilization

28. Little progress has been made in strengthening public finances. While the authorities in general managed to keep the fiscal deficits close to the program targets, the way they achieved them was often not consistent with the programs (Figure 9). The disappointing tax revenue record (see below) forced the authorities constantly to compress expenditures (and to borrow more from the banking system than programmed), in order to hit the deficit objective. This poor-quality fiscal adjustment constrained the provision of basic public services and economic development in general.

Figure 9.
Figure 9.

Madagascar: Fiscal Deficit, 1988–2004

(In percent of GDP)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

Structural reform

Tax policy and administration

29. The absence of any significant improvement in tax collection is one of the greatest failures during the period under review. Despite the fact that achieving an increase in domestic revenue mobilization has been recognized as a key objective in each program, the outturn for 2004, a tax-to-GDP ratio of 11 percent, is equal to the level attained in 1989 (Figure 10). Madagascar remains among the countries with the lowest tax revenue/GDP ratio (Figure 11).

Figure 10.
Figure 10.

Madagascar: Tax Revenue, 1988–2004

(In percent of GDP)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

Figure 11.
Figure 11.

Madagascar: Tax Revenue in Selected Comparator Countries, 1989–2004

(In percent of GDP)

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

30. The authorities have been unable to mobilize domestic revenue in a magnitude consistent with the expenditure needs of the country. Since 1996, Fund-supported programs have sought to address deficiencies in tax policies, the tax base, and tax and customs administration, but achievements to date are minimal. On the upside, the introduction of VAT ensured that the drop in trade taxes, as a result of external liberalization was, to some extent, offset. On the downside, however, tax policy has been plagued by the complexity of the tax code, frequent resort to ad hoc (tax and customs) exemptions, and policy reversals. The tax base remains narrow and tax revenue does not benefit from the high growth areas of the Malagasy economy (EPZs and mining sector, for instance, are subject to very low tax rates). These problems are compounded by the fact that—despite some achievements—tax and customs administration remains weak and plagued by corruption.8

Budgetary process and public expenditure management (PEM)

31. Slow progress in PEM reform remains a bottleneck in the fight against poverty. Important achievements of the early 1990s were the inclusion of the public investment plan into the budget; and the presentation of central government operations on a cash, commitment, and payment-order basis. An ambitious reform agenda was developed under the 1996 ESAF, but measures were often delayed. Progress was made in strengthening local treasury offices and a new budgetary classification was introduced. The track record of reforms in PEM improved recently, as conditionality in the PRGF on PEM was key to reaching the HIPC completion point. Some progress was made in streamlining budgetary procedures to accelerate the execution of spending in priority areas and in improving tracking of this spending.

Monetary policy reform

32. Conditions for the use of indirect instruments have now been broadly met, but policy effectiveness still needs improvement. Attempts at reforming the monetary policy framework through the adoption of indirect instruments started early on. However, the conditions for indirect instruments to be effective were not met in those early years. The gradual development of a treasury bill market and a money market in the mid-1990s, made the use of open-market operations gradually feasible. Administered interest rates were abolished and credit ceilings removed as these markets developed. Finally, financial sector reform improved the transmission of monetary policy through indirect instruments. However, markets remain shallow and the central bank’s liquidity management capabilities weak, often contributing to inconsistent policy responses.

Privatization

33. Privatization has been slow, but major enterprises have been taken out of public management in the past four to five years. Even though privatization has been a policy objective since the late 1980s, only a handful of public enterprises has been effectively privatized. Progress began following the 1996 approval by parliament of a legal and institutional framework for privatization. Commercial banks were the first strategic companies to be fully privatized. In 1996, legal state monopolies were abolished in key sectors (power, petroleum, telecommunications, and air transport). Progress under the most recent PRGF program was mixed. The telecommunications company (TELMA) and the cotton company (HASYMA) were privatized, while the sugar company (SIRAMA), Air Madagascar, and the electricity and water company (JIRAMA) have been put under private management—but still majority owned by the government.

Civil service reform

34. Progress in implementing civil service reform has been slow and proves to be a bottleneck for progress in other areas. The need for reform was recognized early on. However, the wage bill—which was already not high by sub-Saharan standards (Table 6)—was cut repeatedly in the ensuing years, due to shortfalls in fiscal revenue, and salary reform and other reforms remained hampered by these shortfalls. Ambitious reforms were included in the ESAF in 1997, but implementation was delayed because of political resistance. The relevant statute introducing a mechanism for advancement and promotions based strictly on merit, and including a code of ethics and discipline, was approved by the National Assembly in 2001, but not implemented.

Table 6.

Madagascar: Civil Service Wages and Salaries in Selected African Countries 1998–2003

(In percent of GDP)

article image
Source: World Economic Outlook, 2004.
Private sector development

35. Private sector development has been slow.9 Foreign and domestic investment started to pick up around the turn of the twenty-first century, benefiting from three to four years of macroeconomic stability and taking advantage of the legal provisions, which allowed foreigners to lease land for a period up to 99 years.

36. The contribution of the EPZ regime, created in 1989, to employment and exports has been growing steadily.10 The EPZs’ success is attributable to a combination of factors including, a favorable and stable fiscal regime, the availability of labor with matching skills, relatively low wages and more recently, the relative political and economic stability and Madagascar’s eligibility under AGOA. Nonetheless, worries remain that the spillover effects of the EPZs on growth of other sectors remain limited.

Social and sectoral policies

37. Until the beginning of the PRSP process, social and sectoral policies did not yield significant results. Broadly until the turn of the millennium, education and health policies only reached the urban population and, as such, increased the poverty wedge between urban and rural population.

38. The most recent PRSP-JSA (Country Report No. 04/403, 12/10/04) notes significant progress in recent years in several critical areas, including rural development, education, health, infrastructure, HIV/AIDS, nutrition, rural water and sanitation, and social protection. Madagascar is, at current trends, likely to meet a number of MDGs by 2015 (gender equality, complete enrollment in primary school, halting HIV/AIDS spread, malaria, and deforestation), but the likelihood of meeting other goals, such as halving extreme poverty and hunger, 100 percent primary school completion, reducing by 2/3 infant and maternal mortality) is slim on current trends.11

D. Compliance with Program Conditionality

39. The record on meeting quantitative performance criteria is uneven. The first ESAF (1988) had a promising start, with all quantitative performance criteria being met in the first two reviews (Appendix III). During some reviews in the last two programs, up to 50 percent of the PCs and indicative targets were not met. Compliance with the floor on tax revenue proved systematically problematic. Typical is that PCs were often just narrowly missed, which seems to be an indication of lack of implementation capacity more than of inconsistent policies.12

40. The record on structural benchmarks is mixed, but has been improving (Appendix IV). During the first program, a few, rather easy, benchmarks were met. The more challenging ones (e.g., privatization and a start with civil service reform) were not implemented. Structural conditionality intensified significantly during the last two programs. During the 1996–2000 ESAF, implementation of several structural conditions was delayed. The track record during the most recent PRGF is better, with progress in privatization, fiscal policy management and, to a lesser extent, in civil service reform.

41. The compliance rate under the latest PRGF is markedly better than under the previous Fund-supported programs. Even though the number of PCs and SBs was higher under the PRGF than under the ESAF, the compliance rate was much higher. Under the ESAF, Madagascar implemented 40 percent of the conditionality, whereas under the PRGF, this share went up to 87.5 percent. This positive development is also clear from a comparison with African programs in general and mainly reflects increased ownership of the program: under the 1996 ESAF, Madagascar was far below the African average (61 percent) for the period 1995–2000, while under the PRGF, Madagascar was far above the average for the continent (65 percent).

IV. Policy Implementation—Experience and Lessons

42. A combination of factors explains the lack of more broad-based success. First of all, political crises and external shocks frequently interrupted the authorities’ reform efforts and the Fund’s support. Secondly, ownership was extremely low until the mid-1990s. In the early years, political crises and lack of ownership were not totally separable. The 1991 political crisis was in part against the government’s economic liberalization plans, and because the government’s efforts were not whole-hearted, they stalled in the face of the protests. In that sense, the political crisis was not a completely exogenous shock. Since the mid-1990s, ownership has become increasingly stronger. However, at that point a third factor became even clearer than before: the limited absorption and implementation capacity of the country. Factors that influenced progress in implementation are reviewed in this section.

A. Program Objectives and Design

43. Macro-stabilization objectives were generally adequate. One of the main achievements of the Fund’s involvement has been to enhance the authorities’ capacity in macroeconomic stabilization. “Erratic policy implementation” or “inconsistent policies” were often mentioned as major problems early on.13 In more recent years, stabilization capacity has certainly improved—as was shown by the quick reaction to bring inflation under control in the wake of the 2002 political crisis.

44. Macroeconomic program design on exchange rate policy was generally adequate, but in monetary policy it was initially over-optimistic. The advice on exchange rate policies was appropriate and has, in general, supported the government’s macroeconomic policy. On monetary policy, the desire to rely on indirect instruments seemed premature (see above) in light of the fact that the markets and their infrastructure, as well as the authorities’ implementation capacity, were for a long time not ready for this transition. This undermined policy effectiveness.

45. The design of fiscal policies was constrained by underperforming tax revenues. Throughout the period, programs allowed for large fiscal deficits (before grants), which were justified, given the country’s development needs. On the other hand, the room for maneuver to shrink the fiscal deficits over time and to make qualitative cuts in expenditure was limited, given the constraint imposed by low tax revenues.

46. Program design with respect to tax policies and tax and customs administration struggled systematically with the effects of the disappointing tax revenue record. Revenue shortfalls compared to program targets mainly stemmed from the fact that the authorities did not implement promised policy measures, or reversed them afterwards—signs of lack of ownership.14 15 Confronted with this underperformance and with the lack of policy consistency, program conditionality focused almost exclusively on weaknesses in tax and customs administration, in the hope that revenue would expand through these measures. Almost no tax policy measures were part of the conditionality since 1997. While strengthening the administration was necessary in its own right, programs seem to have put too much hope in short-term revenue gains from these improvements and tried to move too fast.16 Limited implementation capacity and resistance from vested political interests prevented reform in this area.

47. Fund arrangements exhibited a certain degree of flexibility to accommodate shocks. The floors on net foreign assets of the central bank, tax revenue, and bank credit to the government, were adjusted on several occasions to take into account the impact of natural disasters, and economic and political shocks, which allowed the country to comply with the (revised and adjusted) framework.

48. Implementation of the broad-based structural reform agenda since 1996 was constrained by the limited implementation capacity of the authorities. The reform agenda became broad-based, addressing simultaneously several areas critical to generating sustained growth. The broadening of the reform agenda was, on the one hand, a reflection of the greater attention that IFIs, in general, had started to pay to issues such as corruption and governance.17 On the other hand, this shift in focus was in response to signs of a greater willingness by the authorities to tackle poverty at its roots. However, in response to this change in approach, programs did perhaps not fully appreciate the limited implementation capacity of the authorities as a major stumbling block (as well as lingering political resistance from vested interest groups against some parts of the programs).

B. Ownership and Implementation Capacity

49. Until the mid-to-late 1990s, program implementation was hampered by lack of ownership. Lack of ownership of, and political resistance to, reforms was obvious in the late 1980s and early 1990s. The authorities were willing to take those measures necessary to jumpstart the economy, but there was certainly no consensus around more far-reaching measures such as privatization and civil service reform.

50. The turning point came in the second half of the 1990s. Civil society, social and religious groups converged on the view that structural reform was the country’s only way out of poverty and began to consider the IMF (and other IFIs) as vehicles to achieve this goal. The movement in favor of removing the president in 1996–97 was, among other reasons, inspired by the fact that he was seen as an obstruction to the government’s reform efforts. Once the new president’s, the government’s and the civil society’s interests were aligned, ownership of Fund-supported programs improved significantly. From 1997–98 onward, the structural reform agenda started to move forward. The support of civil society also brought corruption and governance to the forefront of the agenda, although initial results were weak.18

51. Growing ownership became even more evident from the domestic consultative process that preceded the publication of the PRSP, and from the contents of this document. The same sense of ownership was also felt in the negotiations on the most recent PRGF. The goals of the PFRG were fully consistent with the PRSP framework and were fully supported by the authorities. The high degree of compliance with the structural agenda under the PRGF (see above) is a reflection of the high sense of ownership of program and PRSP. The prospect of debt relief at the HIPC completion point was, of course, an additional incentive for the government to engage in the above initiative, but even without that, there has been a significant change in the climate surrounding Fund program negotiations.

52. Cultural factors contributed to the authorities’ cautious attitude toward Fund-supported programs. The authorities—and Malagasy society as a whole—were for a long time cautious with respect to foreign investment in domestic, productive resources. Critical modifications to the legal framework, to facilitate the entry of foreigners and provide them security regarding their investment, were introduced slowly.

53. Program implementation continues to suffer from limited absorption capacity. The civil service in general is weak and ministries lack adequate technical skills. While recognized early on by the Fund and the Bank as impediments to reform, success in addressing them has been limited. Weaknesses in public expenditure management have aggravated the problems. This has a number of consequences. First, a weak civil service makes implementation of specific policies difficult; in addition, the civil service cannot fulfill its role as advisor or institutional memory for the government; hence, the presence of ad hoc or inconsistent policies.

54. Madagascar has been a recipient of extensive technical assistance in the monetary, fiscal, and statistical areas, but implementation has generally been weak. Thus far, the technical assistance input has only been able to make a small dent in Madagascar’s capacity problems.19 This limited success seems attributable to two factors. One is the limited absorption capacity. Several implementing agencies chronically suffer a lack of adequate funding, with repercussions on staffing and equipment. These constraints have impeded the implementation process. Secondly, technical assistance recommendations could have been better aligned with the authorities’ priorities. In those cases, the authorities did not see a need for implementation, but Fund-supported programs were often based on the assumption that the advice would be implemented.

C. Collaboration with World Bank

55. The division of labor between both institutions has generally been satisfactory. The World Bank has focused its advice on sectoral structural reforms, including privatization and private sector development, basic infrastructure, agriculture, education and health, civil service reform, legal reform, and poverty monitoring and evaluation. The Fund has relied to a great extent on the Bank’s technical expertise in these areas. Cooperation was close on the PRSP and the Bank has become more active in some sectors, such as agriculture, since the publication of the PRSP.

56. A few episodes have caused frictions between both institutions. First, the World Bank was reluctant to engage in a wide-ranging civil service reform, which, to some extent, delayed the Fund’s work in that area. This reluctance stemmed from the Bank’s (negative) experience in other parts of Africa with civil service reform. Hence, the Bank preferred to focus instead on pointed interventions to make specific public-sector units more effective. Secondly, during the 2002 political crisis, the Bank kept its presence in Antananarivo, unlike the Fund. In the absence of the Fund, the Bank was asked by the new president to provide advice on a wide range of policy issues. Advice on tax exemptions, inconsistent with the Fund’s views, created tensions between both institutions for a while. Meanwhile, these issues have been resolved and coordination is smooth.

D. Donor Coordination

57. Donor coordination has gone through ups and downs, but generally could have been better. At times, lack of coordination among donors led to competition among them. The 2002 crisis led to confusion among the donors, putting them in two camps. At other times, coordination was optimal. For instance, in 1997 the Fund managed to get all the donors behind its requirement that tax exemptions be revoked before understanding could be reached on a new Fund-supported program.

58. The lack of a solid public expenditure management framework has impaired the effectiveness of donor assistance. The lack of a good tracking system in the government has impaired the smooth channeling of funds to the projects. The PRSP calls for improvements in public expenditure management and for better coordination among donors. Initiatives to that effect have already been taken.

59. Aid flows were not evenly distributed over time. Aid flows dropped significantly during periods of political uncertainty (1992, 1998, and 2002) (Figure 12). Sharp increases were seen in 1997, 1999, and 2003, coinciding with a new program or the resumption of a program. Reductions, interruptions, and volatility in aid flows sometimes interfered with smooth program implementation.20

Figure 12.
Figure 12.

Madagascar: Grants and External Borrowing, 1987–2003

Citation: IMF Staff Country Reports 2005, 363; 10.5089/9781451825367.002.A001

E. Summary: Lessons for the Future

60. The satisfactory outcome in terms of macro stabilization is the result of constant involvement to ensure that the authorities remained focused on stabilization. More generally, as the following examples indicate, maintaining a constant dialogue—even in the absence of a Fund program—has proven beneficial in the long run. On the positive side, the exchange rate regime was established in 1994 outside the program, but with Fund assistance, and turned out to be a success. On the negative side, the absence of the resident representative during the 2002 political crisis damaged the Fund’s reputation and required a great effort to (re)gain the new government’s confidence.

61. Involvement of civil society (social and religious groups) in the policy dialogue has facilitated achieving the program objectives. The start of a dialogue between Fund and civil society in the mid-1990s has helped in convincing important groups in the Malagasy society of the need for structural reform to address the country’s poverty. This, in turn, strengthened government ownership of the reform agenda in the ensuing period.

62. Weak governance and corruption hindered progress in key areas. Various international governance and corruption watchdogs put Madagascar typically between 20 and 50 on a scale from 0 (worst) to 100 (best).21 These sources also agree on improvements in the past three–four years. However, certain parts of the administration—in particular tax and customs administration—are still widely perceived as corrupt.

63. The lack of progress in tax policy and administration begs the question whether a stricter approach, with greater emphasis on tax policy, could have achieved better results. During the last two programs, waivers were typically given when tax revenue did not meet the floor.22 On some occasions, this was justified (impact of cyclones), but on other occasions, the shortfalls were due to policy reversals or to non-implementation of structural measures. The reviews justified the waivers because, in some cases, the authorities had taken additional measures in the run-up to the program review or had committed during the review to strengthening tax and customs administration. Although increasing tax revenue was considered by all programs as an essential policy objective, it was apparently never considered important enough to trigger the noncompletion of a review, perhaps out of fear that the growing momentum in structural reform would have been lost.

64. Faced with this situation, program conditionality could have incorporated more tax policy measures. The combination of failing to meet the quantitative performance criterion on tax revenue, and the PCs or SBs on tax policy could then have been used to take a much harder line during the respective program reviews.

65. A tougher approach once worked. In one instance (1997), the Fund took a harder line by setting the removal of specific ad hoc customs exemptions as a condition for agreeing on a Fund program.23 Coordinated donor support helped in bringing about the desired result. Therefore, the more general question remains whether the Fund should have taken a harder line, given that low tax revenue mortgaged several other aspects of the macroeconomic framework and of the structural reform agenda.

66. More insistence with respect to civil service reform in the mid-1990s could have been beneficial in the medium-to-long term. When it became obvious in the mid-1990s that weaknesses in the civil service would be a bottleneck for reforms in other areas, the Fund should perhaps have insisted more with the World Bank to expedite its work and move to an all-encompassing reform. The inability of the civil service to attract sufficient numbers of skilled staff has impaired the country’s capacity to mobilize revenue and deliver essential public services—justice, public security, education, and health—and ultimately, assist in laying the foundations for poverty reduction. The need to decompress the salary scale and to introduce performance contracts were cited as crucial objectives as early as 1995, but partial implementation only began after 2000.

67. Initially, the structural reform agenda could perhaps have benefited from a more parsimonious approach. When the government began to demonstrate stronger ownership in the late 1990s, the reform agenda—and the program conditionality—was broadened significantly under the ESAF.24 However, the country’s limited absorptive capacity would probably have warranted a more parsimonious approach. While selectivity in terms of areas of reform was not a real option, in light of the country’s wide-ranging needs, a more selective approach in terms of conditionality could have achieved better ownership and outcomes at that time.

V. Policy Challenges for the Medium Term

68. Madagascar’s overarching economic challenge is to achieve higher economic growth and to reduce poverty significantly. Broadly consistent with the MDGs, the government’s goal, as stated in the PRSP (June 2003), is to reduce the poverty rate by half in 10 years by working along three axes: improved governance, growth that benefits the poor, and improved social service delivery. The country’s capacity to address these issues remains constrained by low domestic savings, external viability, limited availability of skilled workers, and a weak public administration network. Hence, in order to achieve the broader goals, it will be necessary to continue reforms in most of the areas discussed in this report.

69. Durable success in reducing poverty and achieving sustained economic development requires a further acceleration of growth. Madagascar’s economic growth rate is still not in the range required to achieve the goals stated in the PRSP and the MDGs. Future programs should contain ambitious measures to bring about this acceleration, including by fostering private sector-led growth—among others, through further privatization—further diversifying the economy, rebuilding the infrastructure, enhancing productivity in agriculture, and capacity building. Special attention needs to go to the post-AGOA period (the AGOA III “third-party” fabrics provision is set to expire in 2007), to ensure that the EPZs contribute more to economic development and are better linked to the domestic economy.

70. Further capacity building is the key to success in meeting these challenges. Enhancing the authorities’ absorption and implementation capacity is to be achieved through further technical assistance in the monetary, financial, fiscal, and statistical fields. Capacity building needs to be accompanied by further civil service reform to ensure capacity retention. To overcome the problems of the past, coordination with the authorities and among donors should be strengthened and the authorities should devote sufficient resources to ensure more effective absorption. Technical assistance should be aligned with the policy priorities. The authorities are currently finalizing a time-bound technical assistance plan in which they define and prioritize their needs. When finalized, it will be taken to the donors in order to arrive at a coordinated delivery process. This current initiative is an excellent starting point for addressing this challenge.

A. Macroeconomic Challenges

71. A key challenge for sustained growth is to maintain a stable macroeconomic environment and reduce external vulnerabilities. To that end, further enhancements in policy formulation and implementation are essential, both in the fiscal and monetary areas. The 2002 crisis demonstrated, on the positive side, that macroeconomic stability can be regained more quickly than before with appropriate policies. However, on the downside, it also demonstrated the importance of a stable environment to reach the country’s long-term goals of poverty reduction. Even though the crisis only lasted a few months, it led to a significant increase in income poverty—thereby washing away many of the gains made in previous years—and in unemployment.

72. Making the fiscal position sustainable in the medium term should rank high on the policy agenda. Given the pitfalls of the past, tax policy reform should be a priority. The authorities need to establish a tax code that is simple, stable, transparent, and eliminates loopholes. An important first step would be to eliminate any remaining ad hoc exemptions. Work should be geared toward broadening the tax base. The tax base is relatively narrow, excluding or only lightly taxing some important sectors (such as the EPZs and the mining sector). This issue needs to be handled with care and changes should be implemented over time. A balance needs to be found between maintaining the country’s competitive edge and increasing domestic tax revenue.

73. Addressing the low level of domestic savings is a third challenge. The authorities are to be commended for building a sound banking system. Further work is needed to enhance financial intermediation and access to credit, among other things through the development of other intermediation channels such as microfinance, to reach the rural population, and by removing other impediments to financial services.25 Financial deepening will help boast domestic savings, could contribute to poverty reduction, and thus make the country less vulnerable to external shocks.

B. Structural Reforms

74. Madagascar’s recent track record in reform is promising and the current momentum should be maintained. Moreover, there is good correspondence between the outstanding elements of the structural reform agenda and the priority areas identified in the PRSP.26 Within this broad agenda, the Fund should focus on the areas discussed below, while working in close collaboration with the World Bank, who is actively engaged in several other structural reform areas.

  • Tax and customs administration. An action plan to rehabilitate the tax and customs administration has been formulated recently with FAD’s technical assistance, and its implementation began under the just-completed PRGF. Under a new program, the needs in the areas of management improvement, simplification of procedures, and improved IT should be a priority. Gains in efficiency and effectiveness of customs procedures should lead to better duty collection, which would to some extent offset the effects of tariff reductions. Finally, tax evasion and fraud should be tackled decisively.

  • Public expenditure management. Strengthening the links between approved budgets and budget outturns, and between the PRSP and the annual budgets is needed to protect priority spending, safeguard spending discipline, prevent the accumulation of expenditure arrears, and ensure the efficient delivery of public services. In order to achieve this, it is necessary to (i) strengthen the budget preparation process; (ii) strengthen the link between the treasury and the budget; (iii) strengthen internal and external controls; and (iv) develop an expenditure tracking system.

  • Trade liberalization. Notwithstanding the progress made, given Madagascar’s dependency on external trade for economic growth, further liberalization would be welcome. The recommendation of the Diagnostic Trade Integration Study should be further implemented. A recent paper on regional trade initiatives in Africa argues that countries should engage more in nondiscriminatory liberalization (i.e., beyond the regional initiatives) and take additional measures to facilitate trade.27 28 As such, the focus should be on continuing to simplify the tariff structure; improving export services; strengthening the domestic supply response to export opportunities; developing a strategy to maximize the benefits from joining SADC; preparing measures to adjust to the new market conditions in the textile sector in the near term; and fostering cross-border sector cooperation.

  • Corruption and governance. An anti-corruption strategy has been developed. An independent anti-corruption bureau was established at the end of 2004. In the coming years, the government is urged to implement anti-corruption reforms, focusing on government services. Corruption is perceived to be still pervasive in the police force, medical centers, land-titling agencies, and customs and lower courts (Country Report No. 04/403, 12/10/04). Progress in this area is essential to improve the business climate. In that regard, special attention should be given to making the judicial system more efficient and effective (the World Bank is providing technical assistance) and recent improvements in customs should be consolidated.

VI. Future Fund Relations

A. Rationale for a Successor Program

75. To assist in achieving Madagascar’s poverty reduction goals, there is a strong case for continued Fund engagement, preferably through a successor-PRGF arrangement. A PRGF would be justified on the grounds that Madagascar’s economy remains vulnerable, domestically as well as externally. In addition, it is expected that the country will continue to have significant balance-of- payments needs in the medium term, given the very large investment needs and the low domestic saving rate. Continued Fund support would also facilitate the mobilization of donor support, in the form of grants and technical assistance. Donors themselves are keen on having further Fund involvement. Given the country’s vulnerable position, external borrowing should be measured and remain at highly concessional rates. External assistance should preferably be in the form of grants. The structural reform agenda should build on the momentum achieved during the recently completed PRGF, and continue in the areas listed above. Fund involvement should assist in building up a diversified economy. Actions and benchmarks should be specific and selective.

76. Risks associated with this strategy mainly stem from the political side. There is nowadays a broad consensus in Madagascar about the need for, and direction of, reforms. However, this stronger ownership, experienced during the most recent years, needs to be consolidated. Democracy is maturing, although political upheaval remains a risk. In this context, broader-based growth will help distribute the benefits of the reforms and contribute to political and economic stability. To minimize the risks of political resistance to Fund program conditionality, careful coordination and prioritization, in consultation with the authorities, will be necessary. Finally, Madagascar is in very good standing with respect to repaying the Fund, so this is not considered among the high-risk factors.

B. Exit Strategy

77. Fund involvement with Madagascar should be reevaluated after this successor program. If the country continues on the current path of economic growth, stability, and structural reform, and lasting impact is made on poverty reduction, Madagascar could then perhaps move toward a low-access PRGF arrangement. If no progress is made, the Fund’s position should be re-evaluated, in particular, with respect to the credibility of its involvement.

APPENDIX I

Madagascar: Millennium Development Goals

article image
article image
article image
article image
Sources: World Bank; and Fund staff estimates.

Targets 12–15 and indicators 33–44 are excluded because they can not be measured on a country specific basis. These are related to official development assistance, market access, and the HIPC initiative.

APPENDIX II

Madagascar: Ethiopia, and Tanzania—A Comparison

article image
article image
1/ Madagascar, Ethiopia, and Tanzania emerged from decades of economic stagnation under a policy of state controls and planning in 1995. Since then, Tanzania has made greater progress in a number of areas than Ethiopia and Madagascar.

APPENDIX III

Madagascar: Quantitative Performance Criteria and Benchmarks Under Fund-Supported Programs, 1988–2004

article image
article image

Revised targets.

At the time of the 4th review several of the December 2003 PCs were predicted to have been missed, hence it was decided to assess compliance with the program based on March 2004 targets

APPENDIX IV

Structural Conditionality Under Fund Program 1988–2004

article image
article image
article image