Countries undertaking programs of adjustment and reform supported by the SAF and ESAF have brought their economies a long way from the doldrums of the early 1980s. Real per capita output growth among nontransition ESAF users has, on average, caught up with that in other developing countries. The social indicators in most countries have improved. Roughly three-quarters of ESAF users have moved closer to external viability. Budget deficits have been trimmed, and instances of very high inflation have been virtually eliminated. Developments in the last one to two years have been even more encouraging: while this may owe something to the favorable global environment, the liberalization and restructuring undertaken over the past decade give grounds for believing that durable gains in economic potential have been achieved in the countries under review.

Countries undertaking programs of adjustment and reform supported by the SAF and ESAF have brought their economies a long way from the doldrums of the early 1980s. Real per capita output growth among nontransition ESAF users has, on average, caught up with that in other developing countries. The social indicators in most countries have improved. Roughly three-quarters of ESAF users have moved closer to external viability. Budget deficits have been trimmed, and instances of very high inflation have been virtually eliminated. Developments in the last one to two years have been even more encouraging: while this may owe something to the favorable global environment, the liberalization and restructuring undertaken over the past decade give grounds for believing that durable gains in economic potential have been achieved in the countries under review.

Notwithstanding these gains, ESAF countries remain among the poorest in the world, and must aim for faster economic growth than other developing countries on a sustained basis if they are to close the enormous gap in living standards.38 This is not in prospect so long as their investment and domestic saving rates continue to fall short of those in the rest of the developing world. Bolder strategies are called for to lift countries onto a higher growth path. This review has confirmed that the achievement of sustained, outward-oriented growth is also critical in accelerating progress toward external viability. Hence, the two principal goals of the ESAF are mutually supporting and can be achieved through a common set of policies, combined with access to financing on appropriate terms and debt relief where necessary.

At the heart of a strengthened approach must be greater efforts to cut government budget deficits, which in 1995 still averaged 7–8 percent of GDP (overall, before grants) in the countries under review. Numerous empirical studies have concluded that public saving is the most—possibly the only—effective policy instrument to influence national saving directly in the near term.39 This study finds that weak fiscal discipline inhibits growth through other channels as well—by contributing to chronic inflation, weak external positions, and stop-go policy implementation. With these interrelationships in mind, proposals for strengthening ESAF-supported programs have been grouped under four broad headings: (1) the key elements of a stronger and reoriented fiscal adjustment effort; (2) a more resolute approach to reducing inflation; (3) a more focused and concerted push on key structural reforms; and (4) steps to reduce the incidence of major policy slippages and encourage more sustained policy implementation. These proposals are discussed in turn.

Growth-Enhancing Fiscal Adjustment

By and large, fiscal adjustment thus far in ESAF-supported programs has been modest, and future programs should aim to bring about more decisive deficit reduction. For many countries, simply meeting targets more or less as ambitious as those set in the past would represent a major advance, while for others the objectives themselves should be more ambitious. The means to achieve the required additional adjustment, however, may also have important implications for economic growth, as well as for the path of adjustment itself. In many countries, the bulk of deficit reduction will have to come from structural reform on the expenditure side, in the civil service and public enterprises in particular. Since such reforms may be costly in the short term (requiring redundancy payments, for instance), it should not always be expected that budget deficits will be reduced in the first program year. A general aim, however, should be to cut deficits and raise national saving rates significantly over the course of a three-year program. As programs are implemented, policies should be adjusted as necessary to ensure this result.

The Tax Burden

There are no established standards for what would constitute an “appropriate” tax burden in ESAF countries. Revenue objectives need to be related to some notion of the adequacy and efficiency of government spending and, more generally, the appropriate size of government. There is evidence, however, that the capacity to raise revenue efficiently tends to rise with the level of economic development. For ESAF countries, revenue-GDP ratios in excess of about 20 percent would be higher than expected, given their level of development.40 Accordingly, some ESAF countries with ratios much above this level should probably aim to reduce the revenue burden over time. Conversely, there are cases—notably those countries in which taxes amounted to less than 10 percent of GDP in 1995—where higher revenues should be feasible, both as part of an additional fiscal adjustment effort and to support a higher level of productive government expenditure in the medium term.

For most ESAF users, however, the share of tax revenues in GDP is already comparable to that in other developing countries. This suggests that these countries should be focusing their tax reform efforts on simplifying and rationalizing tax systems to make revenues more robust and minimize distortions (Box 6), rather than to boost the overall tax take. In the process, resort to short-term ad hoc revenue measures—often made out of budgetary necessity—should be viewed as strictly temporary and not as a substitute for fundamental reform of tax policy and administration.

Strengthening Tax Systems

The review of tax policies in ESAF countries in Abed and others (forthcoming) suggests some important priorities for tax reform in future programs:

  • First, a well-designed, broad-based consumption tax such as the VAT appears to be the surest instrument for achieving a rapid strengthening of the revenue base, in a way that minimizes disincentives to save and invest. It can also lay the groundwork for improving other taxes, such as those on international trade. The tax should have a single rate (or at most two rates) and the fewest possible exemptions.

  • Second, as domestic tax reforms take effect, import tariffs should be reduced to moderate or low average rates, and the dispersion narrowed so as to reduce arbitrary and excessive rates of effective protection. Export duties should be avoided.

  • Third, reforms of taxes on incomes and profits should focus on simplification and base broadening, to allow needed reductions in overly high marginal tax rates. These reforms should usually be complemented by a simplified (generally presumptive) tax regime for small businesses and the informal sector.

  • Fourth, significant gains in revenue performance are attainable from improvements in tax and customs administration, such as streamlined procedures, special attention to large taxpayers, improved audits, and computerization. Such reforms have long gestation lags, however, and need to be started early and sequenced carefully if they are to pay dividends during a program period.

Capital Spending

Deficit reduction achieved through cuts in capital spending would not raise public saving. Whether capital spending in the public sector is too high or too low in ESAF countries is a difficult judgment and one that would need to be made case by case. But it seems clear that repeatedly squeezing capital spending ad hoc when revenues fall short of target, as has been common practice, is not the appropriate way for such important allocative decisions to be made. This problem could be addressed in program design by (1) incorporating in budgets more cautious assumptions regarding the feasible pace of tax reforms (or the response of revenues to reforms); and (2) putting high-priority capital spending in a “core” budget, to be protected against adverse contingencies.

Social Spending

Spending on health and education, though conventionally classified as government consumption, is more in the nature of investment (in human capital). Provided the resources are used effectively, such spending is growth enhancing. Like productive capital spending, therefore, high-priority expenditures on health and education should be protected in a “core” budget.41 To ensure this, considerable improvements are needed in the quality and availability of expenditure data on a functional basis, together with information on the efficiency of such expenditure. This should be a focus of technical assistance and new donor financing. Such efforts would complement World Bank plans to enhance its monitoring of the “outputs” from social spending (school enrollment, literacy rates, access to safe water, and so on) and may lead to a better understanding of the links between input and output indicators.

Closer monitoring is also needed to allow effective evaluation of social safety net measures incollaboration with the World Bank and other institutions. While such measures now feature prominently in many ESAF-supported programs, little followup information has been provided regarding their impact on the targeted population groups.

Other Current Spending

This leaves three areas where progress to date has not been adequate and a clear break from the past is essential to bring about the needed reduction in budget deficits.

  • Civil service wage bills. Attempted reforms in this area have been partial, prone to reversal, and frequently reliant on cuts in wages rather than staff numbers. Thus, the quality and efficiency goals essential for long-term savings have not been met. The design of these reforms lies within the World Bank’s mandate, but conditionality in IMF programs should attach greater weight to the achievement of headcount reductions and reforms of the civil service wage structure that would produce durable—if not immediate—budget savings, while improving the efficiency of the civil service.

  • Public enterprises. The opportunity costs imposed by public enterprises on the state budget are not limited to the recorded subsidies and transfers. They extend to tax breaks, quasi-fiscal assistance—through directed loans from state banks, for instance—and tax revenues that would accrue if the assets were put to more productive use. Budgetary concerns are thus one of several reasons to intensify public enterprise reform in ESAF-supported programs.

  • Military spending. In line with the global trend, military spending tended to decline during ESAF-supported adjustment and is now estimated to absorb around 2 percent of GDP or less in approximately half of all ESAF countries. In a good number of countries, however, a much higher level of resources continues to be devoted to the military, suggesting the potential to achieve significant economies in this area. Such opportunities should be seized, as and when the social and political climate permits.

Expenditure Management

Finally, improved budgeting and expenditure control systems are needed, both to promote the desired expenditure restructuring and to allow closer adherence to overall fiscal targets. There is a strong advantage to addressing problems in this regard at an early stage. Programs could therefore make greater use of prior actions to encourage progress in a number of areas, including measures to provide appropriate incentives for officials charged with implementing the reforms, and to ensure transparency and accountability in expenditure management.

More Decisive Disinflation

Experience suggests strongly that those countries with inflation stuck in the intermediate range are sacrificing growth. Although the empirical estimates are inevitably imprecise, the present study and others have found that costs appear to mount once inflation exceeds single-digit levels.42 The analysis in this study suggests that the negative relationship (plotted in Figure 20) is significant even after controlling for supply shocks (such as weather and the terms of trade) and other determinants of economic growth; that it is proportional—so that reducing inflation from 30 percent to 10 percent may be just as beneficial as reducing it from 90 percent to 30 percent; and that it applies in ESAF countries just as in other developing countries. In contrast to some earlier findings (particularly those drawn from studies that impose a linear form on the inflation-growth relationship), these results seem to hold even when all cases of high inflation are excluded from the analysis.

Figure 20.
Figure 20.

Association of Real Per Capita GDP Growth with Log Inflation1

(Mean values within groups ranked by inflation)2

Source: IMF staff estimates.1Based on 105 developing countries’ annual data, 1981–95; a total of 1,564 observations (I I missing observations).2The 1,564 observations were ordered according to rate of inflation, then divided into 15 subgroups of 104 observations each (the highest inflation group contains 108 observations).

The nature of this relationship is uncertain and probably complex: growth may benefit from low inflation directly, through improved resource allocation and higher investment, or indirectly as a result of the broad-based reforms that are typically needed to sustain low inflation. Thus, the full effects are likely to accrue only in response to an anti-inflation strategy that is comprehensive and consistent. Nevertheless, the potential gains from policies that achieve and maintain low inflation appear to be substantial, not only for output growth but also for the distribution of income: inflation tends to be a regressive tax, especially in low-income countries, where inflation hedges for the poor are scarce.

Potential Costs of Disinflation

Disinflation is traditionally considered to impose short-run output losses on the economy. This depends on the circumstances, however. There is evidence that costs are lower at higher initial levels of inflation: thus, those with most to gain in the long run—those in or above the higher end of the intermediate inflation range—may also have least to lose in the short run. The degree of inflation inertia that generates the short-run losses is also likely to be less in 11 ESAF countries, with typically large informal sectors, than in more developed economies. Finally, even if there is a short-run cost from the demand squeeze, this may be more than offset by accompanying supply-side effects when disinflation is part of a broader policy package supported by external financing (as is the case in ESAF-supported programs).

What is the evidence? Except in countries with low initial inflation (which have tended not to target disinflation), growth typically rose immediately in SAF/ESAF-supported programs (Figure 21). Looking closer at programs according to the degree of disinflation achieved, the record shows that even countries that halved their inflation rates in the course of a program had on average faster growth than during the preprogram period. This supports the notion that trade-offs between growth and inflation, if they exist, are dominated by other factors. It does not appear, then, that concerns about short-run output effects should deter countries from a bolder effort to reduce inflation; still less do they justify postponing disinflation indefinitely, as has happened in several cases.

Figure 21.
Figure 21.

Real Per Capita GDP Growth in SAF/ESAF Programs1

(Median; percent change over previous year)

Source: Phillips (forthcoming).1Figures in parentheses above bars refer to the number of observations (programs).

Another consequence of disinflation is the need for the public sector to adjust to the loss of seigniorage, either by tapping other revenue sources or through expenditure savings. For countries with initial inflation in the 30–40 percent range (where seigniorage appears to be maximized), reaching 5–10 percent inflation could require the government to find compensating fiscal adjustments of as much as 2 percent to 2½ percent of GDP. This amount is sizable, but should be achievable if—as has been emphasized—low inflation is part of a strategy that would extend to fundamental tax and expenditure reform.

Implications for Program Design

SAF/ESAF-supported programs have in fact consistently targeted single-digit inflation. But the record in achieving this objective has, with few exceptions, been poor, both during and outside programs. The sharp downturn in inflation in 1996 in a number of countries is welcome, but—set against the trend of the previous 10 years—does not dispel concerns about program performance in this area.

Why have programs so often failed to deliver low inflation? One obvious answer would be that financial ceilings—in particular, on credit expansion—have not been respected. Overshooting of targets for net domestic assets (NDA) has indeed been common (Table 7). However, Phillips (forthcoming) finds no systematic link between adherence to program targets for NDA, on the one hand, and the record on meeting monetary growth and inflation targets on the other. Projected money growth was exceeded in 13 of the 14 programs where first-year NDA ceilings were met, reflecting unanticipated balance of payments inflows.43 Inflation overshot in 10 of these programs, and in fact did so even more frequently than when NDA ceilings were exceeded.

Table 7.

Deviations from Inflation Targets, by Net Domestic Assets Performance

(Cases of intermediate and high initial inflation; first program year)

article image

Source: IMF staff estimates.

These findings point to two possible weaknesses in program design. First, programs may have underestimated the extent of fiscal adjustment needed to reach low inflation. The evidence does not show a clear relationship between inflation overshooting and deviations from fiscal targets in the programs under review. But the poor record on inflation calls strongly into question whether the fiscal targets themselves were too loose to achieve the inflation objective, regardless of the stance of credit policy. It is striking that the stronger performers on inflation (countries that achieved average inflation of 10 percent or less during 1993–95) had reduced their primary fiscal deficits to less than one-third of the average level prevailing in other countries (Table 8). These findings add to the case for stronger fundamental fiscal reforms to achieve lower deficits on a sustainable basis.

Table 8.

Financial Policies and Inflation Performance, 1993–951

(Percent per annum, means)

article image

Source: IMF staff estimates.

Excluding countries that did not begin their first SAF/ESAFsupported program until 1994 (Cambodia, Cote d’lvoire, Kyrgyz Republic, Nicaragua, and Vietnam).

Those averaging single-digit inflation during 1993-95 (including Bolivia and Guyana, which averaged 10.1 percent and 10.6 percent inflation, respectively, in this period).

Adjusted to exclude the 1994 outturns for the CFA countries, which were affected by the CFA franc devaluation in that year.

lncluding grants. Excluding Guyana.

Second, the large majority of programs lacked an effective nominal anchor, relying on NDA ceilings alone to enforce monetary discipline. To some degree, this may have reflected a recognition that the likely fiscal stance would be incompatible with a nominal anchor, particularly in light of concerns about competitiveness. However, unless supported by a commitment to a nominal exchange rate peg, or explicit inflation targets, NDA ceilings leave open a wide range of outcomes in which net foreign assets, money supply, and inflation may all exceed projections.44 Only 8 of the 35 SAF/ESAF-supported programs with intermediate or high initial inflation had a nominal anchor, in all but two cases an exchange rate peg. These programs delivered better-than-average inflation performance. By contrast, as many as one in five programs explicitly geared nominal exchange rate policy to a real exchange rate objective, and these had particularly poor inflation performance (Figure 22).45

Figure 22.
Figure 22.

Inflation Performance With and Without Anchors1

(Sample medians; in percentage points)

Source: Phillips (forthcoming).1Programs with intermediate initial inflation.2Averages of t to t + 2, where t is the first program year.

Programs should not continue to claim single-digit inflation as one of their intended results unless they are designed in such a way that, if the program is fully implemented, this outcome is reasonably assured. The programs under review did not, with hindsight, provide that assurance. Given the strong case for single-digit inflation as an objective, the preferred solution would be the adoption of a significantly bolder approach than has been common hitherto. The goal need not necessarily be rapid disinflation but, particularly when initial inflation is in the intermediate range, it should be expected that it would be brought down to single-digit levels in the course of a three-year program.

The centerpiece in such an approach would be a greatly strengthened fiscal position, underpinned by basic reform of the state finances (including those of the public enterprises). A sufficiently ambitious program of this kind—which includes an inflation goal—should achieve low inflation without formally committing to a nominal exchange rate target or a monetary anchor. But experience suggests that including a formal anchor in such circumstances could add to the program’s credibility and may speed the transition to low inflation. A nominal exchange rate peg or, alternatively, performance criteria for reserve money would serve this purpose. Either one could be adopted on a temporary basis, until such time as low inflation had been established. Each has pros and cons. The exchange rate anchor is more transparent and may be more effective in altering inflation expectations, but it may also be more difficult to exit from smoothly. Money supply targets, on the other hand, do not require the backing of foreign reserves needed for an exchange rate peg and allow the exchange rate to adjust flexibly in response to shifts in the terms of trade; but they can be difficult to manage when the demand for money is highly uncertain (as it often is during disinflation). These factors would need to be weighed case by case.

There can be no doubt that the prerequisites for adopting a formal nominal anchor—staunch adherence to fiscal discipline and minimal indexation—are exacting. There will therefore continue to be situations in which policies sufficiently strong to support a nominal anchor are judged to be beyond reach, at least in the near term. In these cases, programs should acknowledge that their ambitions, not only for inflation but probably also for growth, have to be correspondingly more modest. A more realistic assessment of inflation and growth prospects is needed to provide clearer guidance to national authorities on the additional adjustment needed to place their country on a higher growth trajectory.

Advancing Structural Reforms

Stronger macroeconomic adjustment would deliver a quicker and more pronounced supply response if accompanied by structural measures to encourage private investment and entrepreneurship. Most aspects of structural reform have some bearing on private sector development, and programs should continue to address structural weaknesses across a broad front, setting priorities on a case-by-case basis. Many ESAF countries have achieved a great deal already in areas such as the liberalization of prices and marketing and exchange system reform. Future programs should therefore be expected to devote increased attention to the “second generation” of reforms—those where progress has lagged, or where more can be done to raise countries’ economic potential. These areas would include:

  • Trade liberalization. There is increasing recognition of the important and wide-ranging benefits that result from opening economies to international trade and investment. The countries under review have made progress in this respect, but average tariff rates remain high and should be reduced and streamlined further, together with nontariff barriers.

  • Public enterprise reform. Progress here has been limited, with the result that in most countries public enterprises continue to impede private sector development by preempting resources and hampering competition.

  • Bank restructuring. Persistent weaknesses (operational and financial) in the banking sector are impairing efficient intermediation and thereby constraining business investment.

  • Strengthened property rights. Clear and fair systems of property rights and adherence to the rule of law in business practices are prerequisites for a strong private sector. Many problems in this regard can be traced back to excessive regulation, complex tax and tariff structures, and direct state involvement in commercial affairs. At the same time, governments need to help ensure the development of well-functioning markets to finance private activities. Hence, the most effective contribution the IMF can make is to continue supporting programs that address these root causes, some of the most important of which lie in the first three areas identified above.

The slow pace of reform of public enterprises and banking systems was emphasized in the last ESAF review. The present study has therefore investigated—through detailed examination of selected country cases—the possible reasons for persistently slow progress in these two critical areas. The aim was to identify what could be done in the context of future ESAF-supported programs to help accelerate reform, recognizing that the World Bank retains the lead role in advising on the design and implementation of reforms in both areas. Hence, the focus here is on those aspects of policy—the linkages between structural reform and macroeconomic policies—that fall within the IMF’s area of responsibility. The analysis and findings are set out in Decressin and others (forthcoming).

Public Enterprise Reform

The limited success in hardening the budget constraints of public enterprises in ESAF countries, and especially the disappointing results from performance contracts, have led to increasing emphasis in recent reform programs on privatization and less on state-managed restructuring. Privatization may indeed be the only way that inappropriate financial links between corporations and the government can be severed once and for all. The shift in this direction is therefore welcome, and should be further encouraged.

Nevertheless, because most ESAF users can be expected to retain some enterprises under state ownership, more must be done to ensure that essential discipline can be imposed effectively on these enterprises, and the potential threat they pose to financial stabilization contained. There is no realistic prospect of meeting this goal, however, without the compilation of adequate financial data on public enterprises. Although this issue has long been highlighted by the IMF, little has been done in the context of ESAF-supported programs to press for data improvements in this area. Of the five country case studies examined (involving 18 annual SAF/ESAF-supported programs among them), only two instances could be found of structural conditions being applied to the collection of data on public enterprises.46 Bolivia was the only one of these countries to have compiled financial data on all public enterprises with a reasonable degree of specificity and reliability. In the others, even information on the direct financial flows between public enterprises and the government was frequently unavailable or incomplete, and there were no data on nonbank borrowing (some of which may have been government guaranteed). Also unrecorded were the numerous forms of indirect or noncash support that governments commonly provided to favored public enterprises, such as implicit interest subsidies, loan or exchange rate guarantees, and relief or forgiveness on tax and loan liabilities. In short, the status quo in this area looks much the same as it did at the time of the last review four years ago: public enterprise data are generally adequate in Western Hemisphere countries but still extremely poor in most African countries.

This is an issue on which the IMF should take a more proactive role, in cooperation with the World Bank, by helping to marshall donors’ financial and technical resources to assist countries in what is admittedly a complex and skill-intensive task and by applying conditionality to reinforce the importance of action in this area. The authorities could be encouraged to focus initially on a few key enterprises, expanding coverage as resources permit, but should aim to record not only enterprises’ direct financial transactions with government and banks but also quasi-fiscal support, such as government guarantees for domestic borrowing.

One consequence of the lack of adequate financial data on public enterprises has been that IMF staff were generally not in a position to identify promptly serious emerging financial problems in specific enterprises, and hence to advise the World Bank and the national authorities on the choice of enterprises to be targeted for urgent reform.47 Such problems were most severe in cases where the Bank was supporting reform through sectoral operations, which tended to focus on enterprises that were perceived as major bottlenecks to growth (public utilities and key export producers, for instance), rather than those that imposed the biggest financial drain on the state budget. Once equipped with the necessary information, IMF staff will need to play a more active role in such circumstances, where its central interest in the state finances and monetary conditions in the economy as a whole is at stake. In particular, programs should seek to monitor financial flows between public enterprises, the state budget, and the banking system more comprehensively, so that budget constraints can be hardened.

More generally, conditionality in ESAF-supported programs should continue to reinforce public enterprise reforms planned under the aegis of the World Bank. Particular attention could be paid to measures that promote competition, such as liberalizing the codes governing domestic and foreign direct investment and removing monopoly rights, since these help not only to exert some market discipline over public enterprises but also to create the right environment for privatization.

Restructuring Banking Systems

One manifestation of the inefficiencies and financial mismanagement in the public enterprise sector was a substantial accumulation of nonperforming loans in the banking systems of ESAF countries, often amounting to one-third or more of banks’ total portfolios. This problem was related to, and compounded by, structural weaknesses in the banks’ own management, many of which stemmed from too much government intervention in lending decisions and insufficient effective prudential regulation and supervision.

The five case studies examined suggest several reasons why progress in restructuring and reforming banking systems in ESAF countries has continued to falter.48 Lack of political commitment—especially an unwillingness to cede political influence over banks’ operations—was an important factor in countries in which progress has been particularly poor. A scarcity of banking expertise and skilled staff has also been a problem, one that can be remedied only with time, and a continuation of extensive technical assistance from the IMF and other institutions. On three counts, however, the IMF could take more direct action to improve performance in future ESAF-supported programs.

First, the huge fiscal costs held back reform in some instances (Tanzania provides a clear example). This appeared to result from a failure to articulate the complete strategy at the outset, which would have allowed costs to be better anticipated and financing identified.49 There is scope for IMF staff to engage more actively in discussions with the World Bank and national authorities at the planning stage to ensure that all potential fiscal costs are assessed, including some that tend to be overlooked, such as running costs for loan recovery agencies. As well as reducing the likelihood that reforms would run into financial roadblocks once under way, more prior attention to these issues would have the important advantage of concentrating minds on the link between the pace of a planned strategy and its costs, and might in some cases lead to the adoption of a more ambitious (hence, typically, cheaper) approach than would otherwise have been selected. It could also help to ensure greater emphasis on design features—such as an appropriate “exit” (bank closure) strategy—that would serve simultaneously to promote substantive reform and save budgetary resources. In this regard, consideration could be given to sharing the costs of restructuring more broadly among share-holders, creditors, and depositors—among other things, to reduce risks of moral hazard.

Second, piecemeal implementation weakened the impact of reforms, leading in some cases to significant delays and even reversals. Proceeding, for instance, with financial restructuring (to tackle the “stock” problem) without reforming banks’ operations or enforcing the new prudential and regulatory environment (the “flow” problem) tended to result in recurring bank distress.50 Similarly, failure to upgrade the legal and judicial framework, to allow effective enforcement of contracts, resulted in low loan-recovery rates (as in Senegal and Tanzania), adding to the net cost and thereby hampering reform. The IMF could contribute to more comprehensive implementation by refocusing its conditionality in ESAF-supported programs. The case studies suggest that past use of structural performance criteria and benchmarks has not been very effective, at least in part because of its emphasis on approving plans and passing laws rather than on operational achievements. Instead, programs should monitor the adequacy of policies affecting the banking system and the ability of the authorities to enforce best practices.51 Key aspects would include licensing and exit policies, lender-of-last-resort facilities, rules for loan classification and provisioning, capital standards, and the legal authority and capacity of supervisors to implement prudential regulations and impose penalties. Performance in these areas as well as in relation to the broader and longer-term goals of banking reform should be assessed in the context of program reviews.

Third, lack of progress in public enterprise reform undermined bank restructuring efforts. Indeed, the problem of weak loan portfolios is unlikely to be solved on a lasting basis until the financial weaknesses in the public enterprise sector are satisfactorily addressed. If, as proposed above, information is compiled on the balance sheets and financial transactions of public enterprises, the interrelationships would become clearer, and this may help to promote more concerted strategies for the public enterprise and banking sectors.

Sustaining Programs

The preceding sections have examined the policies and outcomes in all ESAF countries and identified specific areas where the design of programs could be improved. This final section approaches program design from a different perspective—that of countries in which significant interruptions occurred in or between SAF/ESAF-supported programs. The concern in such cases goes beyond a desire solely to avoid breaks in IMF arrangements: if deviations from planned policies were sufficiently large to cause the IMF to withhold support, they could also have been harmful to investors’ and market sentiment, and hence to economic performance. The resulting loss of credibility is costly even when policy adjustments are subsequently made to retrieve the ground lost during the interruption. In the analysis of this issue presented in Mecagni (forthcoming), two principal questions were posed: What were the main factors that gave rise to the interruptions? And could programs have been designed or monitored differently to reduce the frequency of interruptions without unduly compromising objectives?

To the extent that changes in design or monitoring could not be expected to eliminate interruptions, an important—and considerably more difficult—question is raised: would greater selectivity in approving arrangements help? This question is posed not so much from the perspective of whether it would limit the use of IMF resources in risky situations; rather, the issue considered is whether greater selectivity would encourage countries to commit themselves more forcefully to appropriate policies. While it is not possible to address this question empirically (the construction of a counterfactual would be too subjective to be useful), the concluding paragraph raises some considerations that bear on the question.

In this study, interruptions refer to delays of longer than six months between (1) a scheduled review and its actual completion, (2) two annual arrangements in a three-year arrangement, or (3) two three-year arrangements. Like any definition of interruptions, this one has an element of arbitrariness. However, upon examination it appeared to capture situations where a clear lapse from the desired pace of adjustment occurred. By this definition, program interruptions have been numerous. Fifty-one significant interruptions of SAF- or ESAF-supported programs have occurred since the inception of the SAF in 1986, affecting 28 of the 36 countries under review. Only one-fourth of all three- or four-year arrangements were completed without significant interruption.52 The high frequency of interruptions was a critical motivation for examining separately these particular experiences during SAF/ESAF-supported programs.

There is a temptation to think of interruptions as the result only of severe policy slippages. In fact, in almost one-third of the 51 episodes examined, the primary cause of the interruption was not the need to correct a significant policy slippage. About one-sixth of all interruptions stemmed from severe political upheavals that called into question the authority of the government to negotiate or provide a credible commitment to a program. In most of these cases, past policy slippages had occurred, but even in these the dominant factor underlying the interruption and its length was the need to resolve a political crisis before concerted attention could be given to the economic program. In another one-sixth of interruptions, past policies had been broadly satisfactory, but there were delays in agreeing on future policies. In most of these episodes, time was needed for the authorities to muster political support for certain measures or for the IMF and the authorities to agree on measures to address the effects of a recent unexpected external shock.

Thus, about two-thirds of the interruptions were strongly affected by serious slippages in past policies that either weakened the government’s credibility or produced protracted disagreements between the IMF and the government on remedial measures. Not surprisingly, the evidence suggests that, on average, these interruptions were associated with significantly worse policies than uninterrupted programs (Figure 23 shows the comparative evidence for fiscal policies: the interrupted programs—those in black—clearly tend to fall in the left, or “worse,” half of the distribution of fiscal outturns).

Figure 23.
Figure 23.

Frequency Distribution of Changes in the Fiscal Balance1

(In percent of GDP, excluding grants)

Source: Mecagni (forthcoming).1Excluding Guyana.2Or programs where interruptions were not related to past fiscal slippages.3Represents number of cases in which the change in fiscal balance was less than –7 percent of GDP.4Represents number of cases in which the change in fiscal balance exceeded 7 percent of GDP.

An examination of the interruptions attributable to policy slippages finds little support for the view that changes in program design could have made a significant difference to the occurrence or length of the interruption. Still, specific modifications are identified that might have helped in a handful of past episodes and that are worth considering for the future. These conclusions were reached by examining five possible weaknesses in program design that might have contributed to interruptions:

  • Targets for macroeconomic policies were too ambitious.

  • The program of structural reform was insufficiently prioritized.

  • The provision of technical assistance was inadequate.

  • IMF staff monitoring and consultation with authorities was too infrequent.

  • Contingency planning was inadequate.

Identifying overambitiousness in financial targets is far from straightforward. What is achievable in one country at one time may not be a relevant benchmark for other countries with different capacities and constraints or even for the same country at another time. Still, a comparison of targets for key macroeconomic policy variables—such as the main fiscal aggregates—in interrupted annual arrangements with the average of those for all annual arrangements provides a crude test of the hypothesis that interrupted programs were overly ambitious. By this standard, most of the annual targets for the programs interrupted due to policy slippages were not more ambitious than the “average” program, and the average targeted change adjusted for initial conditions in these interruption cases was not statistically different from that for all programs (Figure 24).53

Figure 24.
Figure 24.

Targeted Change in Fiscal Balance in Interruptions Affected by Policy Slippages1,2

(In percent of GOP, excluding grants)

Source: Mecagni (forthcoming).1The horizontal and vertical lines shown correspond to the average values of the variables in the control group of annual programs not affected by interruptions.2The regression line shows the estimated relation between targeted change in fiscal balance, initial imbalance, and a constant for the sample of all annual programs. The sample excludes seven outlier observations.3Using the initial fiscal balance as estimated at the time the program was formulated, which in some cases differs from the final data.

Another approach to the overambitiousness hypothesis is to consider what the targets for interrupted programs would have been had the outcomes reflecting policy slippages been the original targets. This possibility was considered for the 13 interrupted programs where fiscal targets appeared from Figure 24 to be relatively ambitious (that is, for programs above and to the right of the regression line). In 10 of these cases, the fiscal deficit actually deteriorated, at times by several percentage points of GDP. In 7 cases, the policies implemented resulted in the accumulation of external arrears. In another 4 instances, “lowering the bar” to match the actual outcomes would have implied accepting inflation either rising or stuck in the 20–30 percent range. In sum, it appears that easing targets to match outcomes would in most cases have implied endorsement of deteriorating macroeconomic conditions.

There is also little evidence in favor of the hypothesis that overly ambitious or insufficiently prioritized structural reforms were a key cause of interruptions. First, interruptions were linked far more frequently and directly to slippages in macroeconomic policies than to those in structural policies. While structural policies were often not implemented as expected (interruptions linked to policy slippages followed periods in which, on average, over two-thirds of structural benchmarks and about half of the structural performance criteria had been missed), only in a few instances could the proximate cause of the slippage be seen as structural in nature. Second, taking structural benchmarks and performance criteria as an indication of the prioritizing of structural reforms, programs where past policy slippages were the major influence on interruptions typically had no more such conditions than the average for all programs: programs interrupted due to policy slippages had on average two- three structural performance criteria and six- seven structural benchmarks, in line with the wider sample.54

Closely linked to the question of whether policy programs were overly ambitious is that of the adequacy of technical assistance. All of the countries supported through the ESAF have poor administrative capacity, and more technical assistance, if properly received and utilized, would help realize stronger adjustment. In at least four countries, however, it appears that administrative constraints were so severe as to be a key factor behind the interruption of a program—that is, the pace of reform consistent with ESAF support required greater administrative capacity than existed. In at least two of these four cases (Burkina Faso and Mozambique), efforts to provide technical assistance met with less than complete cooperation from officials. In the Lao People’s Democratic Republic, despite a substantial technical assistance effort (covering fiscal, monetary, and statistical issues), the reform agenda still far exceeded implementation capacity. These cases illustrate the limits to what technical assistance can achieve in the short term, with or without the full cooperation of the authorities. However, Equatorial Guinea provides an example where, at least with the benefit of hindsight, it can be said that the IMF could have been more proactive in pressing the authorities to request and work with technical assistance teams.

Because ESAF arrangements are subject to requirements for relatively infrequent formal monitoring (by comparison with stand-by and extended arrangements), it is reasonable to ask whether more frequent monitoring could have helped forestall emerging slippages. A comparison between ESAF-supported countries and countries receiving IMF support among the Baltic countries, Russia, and other countries of the former Soviet Union is telling. The number of resident representatives, the frequency of staff visits, and the total staff resources per country have all been considerably greater in these countries than in ESAF countries. Moreover, quarterly test dates and formal quarterly (or even more frequent) review requirements have existed in all stand-by and extended arrangements with the Baltic countries, Russia, and other countries of the former Soviet Union, while ESAF arrangements have had only half-yearly test dates and reviews.55 Although causality cannot be inferred, the much lower frequency of interruptions (5 out of 29) in arrangements with the Baltic countries, Russia, and other countries of the former Soviet Union than in SAF or ESAF arrangements (51 out of 68) is striking.

It is reasonable to suppose, therefore, that more intensive program monitoring—through quarterly test dates (and disbursements), more frequent reviews, and perhaps more widespread assignment of resident representatives to ESAF countries—might help in some instances to keep governments’ attention focused on the requirements of adjustment policies and to prompt an early response to emerging problems. This approach could be applied flexibly, for use particularly in cases where a solid track record has not been established or where the IMF and the authorities consider that it would help sustain program implementation.

The potential importance of contingency planning depends on the extent to which unexpected developments, such as the terms of trade or weather-related shocks, significantly influenced the slippage in policies. For the most part, other than the instances of major political upheavals referred to earlier, exogenous shocks were not important causes of interruptions. They were a contributory factor, however, in eight cases. Of these programs, even though the risk of the external disturbance had been recognized ex ante, only four had an in-built contingency mechanism. Those that did not were not necessarily flawed: the implicit assumption was that unless the program were modified by a subsequent review, the effects of any disturbance would have been offset fully by additional policy adjustment. In no case, however, did these programs anticipate how such adjustment would be brought about. In at least two of these cases—Equatorial Guinea and Guinea—remedial measures (primarily, administered price adjustments) were eventually implemented. These measures would appear to have been good candidates for inclusion in a list of contingent commitments, in the sense that they could be implemented readily and were quick acting.

In sum, none of the aspects of program design examined provides a compelling explanation of program interruptions related to policy slippages. A few changes to the design or monitoring of programs may help to improve implementation, particularly a more proactive approach to the provision and coordination of technical assistance, insisting on the demonstration of countries’ cooperation before some programs are approved, and more frequent monitoring and assignment of resident representatives (although the potential benefits would need to be weighed against the costs). Greater use of contingency planning may also be desirable in some circumstances, bearing in mind the risk that it may overburden programs. But in few cases is it obvious that these kinds of modification would have been sufficient to avert or considerably shorten a program interruption.

The implication is that, in fact, most program interruptions have been the result of factors outside the IMF’s control—that is, major political upheavals (of the kind described at the beginning of this section) and flagging commitment. Discontinuities or weaknesses in policy management appear to have been related in roughly a dozen cases to less severe forms of political disruption, including routine elections, transitions to multiparty political systems, and social unrest. These events typically resulted in government overspending and a general distraction of the decision-making authority. However, in more than one-third of the interruptions no unusual event of this kind occurred during the period leading up to the interruption. For most of these, myriad influences were at play in the failure to implement policies as planned: lack of public support for adjustment policies, disagreements within narrower political circles, reluctance of the authorities to confront special-interest groups, poor organization, and governance-related weaknesses.

It is these conditions that raise questions about whether greater selectivity in approving arrangements would help reduce shortfalls in meeting policy commitments that ultimately produce interruptions. Specifically, by withholding IMF support for programs until authorities have demonstrated a reasonable ability to deliver agreed policy changes, could the implementation record be strengthened? Several considerations are relevant. First, policy slippages frequently occur around elections—a time when selectivity issues are particularly sensitive. Greater selectivity would probably require stronger assurances than have been provided in the past of the authorities’ ability to implement policies during election cycles. Second, past interruptions are an imperfect guide to the likelihood of future interruptions. Only about half of the countries that had interruptions had more than one. Third, staff-monitored programs were almost as likely to be followed by an interrupted arrangement, or no arrangement at all, as they were to precede a program completed without interruption. These observations suggest that while greater selectivity may help build countries’ resolve to implement programs as agreed, establishing the criteria for greater selectivity would be far from straightforward. Without greater selectivity, however, interruptions are likely to remain a feature of the ESAF experience, as the IMF continues to assist members at the margins of commitment and in the midst of difficult political transitions to start the adjustment process or push ahead with reforms already begun.

Appendix. Annual Arrangements Under the SAF and ESAF1

1 Countries for which an ESAF arrangement was approved before December 31,1994. The symbol in the ESAF bars indicates when a review was completed.


  • Abed, G., and others, forthcoming, Fiscal Reforms in Low-Income Countries: Experience Under IMF-Supported Programs Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Basle Committee on Banking Supervision, 1997, Core Principles for Effective Supervision, Consultative Paper (Basle).

  • Bredenkamp, H., forthcoming, “Methodological Issues,” in Economic Adjustment and Reform in Low-Income Countries: Studies by the Staff of the IMF, ed. by H. Bredenkamp and S. Schadler Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • ———, and S. Schadler, eds., forthcoming, Economic Adjustment Reform in Low-Income Countries: Studies by the Staff of the IMF Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Coorey, S. and K. Kochhar, forthcoming, “Economic Growth: What Has Been Achieved and How?” in Economic Adjustment and Reform in Low-Income Countries: Studies by the Staff of the IMF, ed. by H. Bredenkamp and S. Schadler Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Decressin, J., L. Dicks-Mireaux, Zia Ebrahimzadeh, and A. Ibrahim,forthcoming, “Moving Ahead with Structural Reform,” in Economic Adjustment and Reform in Low-Income Countries: Studies by the Staff of the IMF, ed. by H. Bredenkamp and S. Schadler Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Dicks-Mireaux, L., J. Le Dem, K. Kochhar, and S. Phillips, forthcoming, “The Policy Record,” in Economic Adjustment and Reform in Low-Income Countries: Studies by the Staff of the IMF, ed. by H. Bredenkamp and S. Schadler Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Goldsbrough, D., S. Coorey, L. Dicks-Mireaux, B. Horvath, K. Kochhar, M. Mecagni, E. Offerdal, and J. Zhou, 1996, Reinvigorating Growth in Developing Countries: Lessons from Adjustment Policies in Eight Countries, IMF Occasional Paper 139 Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Haggard, S., J-D. Lafay, and C. Morrisson, 1995, The Political Feasibility of Adjustment in Developing Countries, Development Center Studies Paris: Organization for Economic Cooperation and Development.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 1996, World Economic Out-look, October 1996: A Survey by the Staff of the International Monetary Fund, World Economic and Financial Surveys Washington.

    • Search Google Scholar
    • Export Citation
  • Jayarajah, C, W. Branson, and B. Sen, 1996, Social Dimensions of Adjustment: World Bank Experience, 1980–93 Washington: World Bank.

  • Mecagni, M., forthcoming, “The Causes of Program Interruptions,” in Economic Adjustment and Reform in Low-Income Countries: Studies by the Staff of the IMF, ed. by H. Bredenkamp and S. Schadler Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Phillips, S., forthcoming, “Inflation: The Case for a More Resolute Approach,” in Economic Adjustment and Reform in Low-Income Countries: Studies by Staff of the IMF, ed. by H. Bredenkamp and S. Schadler Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Schadler, S., F. Rozwadowski, S. Tiwari, and D. Robinson, 1993, Economic Adjustment in Low-Income Countries: Experience Under the Enhanced Structural Adjustment Facility, IMF Occasional Paper 106 Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Schadler, S., A. Bennett, M. Carkovic, L. Dicks-Mireaux, M. Mecagni, J. Morsink, and M. Savastano, 1995a, IMF Conditionality: Experience Under Stand-By and Extended ArrangementsPart I, IMF Occasional Paper 128 Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • ———, 1995b, IMF Conditionality: Experience Under Stand-By and Extended ArrangementsPart II, IMF Occasional Paper 129 Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Tsikata, T, forthcoming, “Progress Toward External Viability,” in Economic Adjustment and Reform in Low-Income Countries: Studies by the Staff of the IMF, ed. by H. Bredenkamp and S. Schadler Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation

Recent Occasional Papers of the International Monetary Fund

156. The ESAF at Ten Years: Economic Adjustment and Reform in Low-Income Countries, by the Staff of the International Monetary Fund. 1997.

155. Fiscal Policy Issues During the Transition in Russia, by Augusto Lopez-Claros and Sergei Alexashenko [forthcoming].

154. Credibility Without Rules? Monetary Frameworks in the Post-Bretton Woods Era, by Carlo Cottarelli and Curzio Giannini. 1997.

153. Pension Regimes and Saving, by G.A. Mackenzie, Philip Gerson, and Alfredo Cuevas. 1997.

152. Hong Kong, China: Growth, Structural Change, and Economic Stability During the Transition, by John Dodsworth and Dubravko Mihaljek. 1997.

151. Currency Board Arrangements: Issues and Experiences, by a staff team led by Tomas J.T. Balino and Charles Enoch. 1997.

150. Kuwait: From Reconstruction to Accumulation for Future Generations, by Nigel Andrew Chalk, Mohamed A. El-Erian, Susan J. Fennell, Alexei P. Kireyev, and John F. Wison. 1997.

149. The Composition of Fiscal Adjustment and Growth: Lessons from Fiscal Reforms in Eight Economies, by G.A. Mackenzie, David W.H. Orsmond, and Philip R. Gerson. 1997.

148. Nigeria: Experience with Structural Adjustment, by Gary Moser, Scott Rogers, and Reinold van Til, with Robin Kibuka and Inutu Lukonga. 1997.

147. Aging Populations and Public Pension Schemes, by Sheetal K. Chand and Albert Jaeger. 1996.

146. Thailand: The Road to Sustained Growth, by Kalpana Kochhar, Louis Dicks-Mireaux, Balazs Horvath, Mauro Mecagni, Erik Offerdal, and Jianping Zhou. 1996.

145. Exchange Rate Movements and Their Impact on Trade and Investment in the APEC Region, by Takatoshi Ito, Peter Isard, Steven Symansky, and Tamim Bayoumi. 1996.

144. National Bank of Poland: The Road to Indirect Instruments, by Piero Ugolini. 1996.

143. Adjustment for Growth: The African Experience, by Michael T. Hadjimichael, Michael Nowak, Robert Sharer, and Amor Tahari. 1996.

142. Quasi-Fiscal Operations of Public Financial Institutions, by G.A. Mackenzie and Peter Stella. 1996.

141. Monetary and Exchange System Reforms in China: An Experiment in Gradualism, by Hassanali Mehran, Marc Quintyn, Tom Nordman, and Bernard Laurens. 1996.

140. Government Reform in New Zealand, by Graham C. Scott. 1996.

139. Reinvigorating Growth in Developing Countries: Lessons from Adjustment Policies in Eight Economies, by David Goldsbrough, Sharmini Coorey, Louis Dicks-Mireaux, Balazs Horvath, Kalpana Kochhar, Mauro Mecagni, Erik Offerdal, and Jianping Zhou. 1996.

138. Aftermath of the CFA Franc Devaluation, by Jean A.P. Clement, with Johannes Mueller, Stephane Cosse, and Jean Le Dem. 1996.

137. The Lao People’s Democratic Republic: Systemic Transformation and Adjustment, edited by Ichiro Otani and Chi Do Pham. 1996.

136. Jordan: Strategy for Adjustment and Growth, edited by Edouard Maciejewski and Ahsan Mansur. 1996.

135. Vietnam: Transition to a Market Economy, by John R. Dodsworth, Erich Spitaller, Michael Braulke, Keon Hyok Lee, Kenneth Miranda, Christian Mulder, Hisanobu Shishido, and Krishna Srinivasan. 1996.

134. India: Economic Reform and Growth, by Ajai Chopra, Charles Collyns, Richard Hemming, and Karen Parker with Woosik Chu and Oliver Fratzscher. 1995.

133. Policy Experiences and Issues in the Baltics, Russia, and Other Countries of the Former Soviet Union, edited by Daniel A. Citrin and Ashok K. Lahiri. 1995.

132. Financial Fragilities in Latin America: The 1980s and 1990s, by Liliana Rojas-Suarez and Steven R. Weisbrod. 1995.

131. Capital Account Convertibility: Review of Experience and Implications for IMF Policies, by staff teams headed by Peter J. Quirk and Owen Evans. 1995.

130. Challenges to the Swedish Welfare State, by Desmond Lachman, Adam Bennett, John H. Green, Robert Hagemann, and Ramana Ramaswamy. 1995.

129. IMF Conditionality: Experience Under Stand-By and Extended Arrangements. Part II: Background Papers. Susan Schadler, Editor, with Adam Bennett, Maria Carkovic, Louis Dicks-Mireaux, Mauro Mecagni, James HJ. Morsink, and Miguel A. Savastano. 1995.

128. IMF Condi tionality: Experience Under Stand-By and Extended Arrangements. Part I: Key Issues and Findings, by Susan Schadler, Adam Bennett, Maria Carkovic, Louis Dicks-Mireaux, Mauro Mecagni, James H.J. Morsink, and Miguel A. Savastano. 1995.

127. Road Maps of the Transition: The Baltics, the Czech Republic, Hungary, and Russia, by Biswajit Banerjee, Vincent Koen, Thomas Krueger, Mark S. Lutz, Michael Marrese, and Tapio O. Saavalainen. 1995.

126. The Adoption of Indirect Instruments of Monetary Policy, by a staff team headed by William E. Alexander, Tomas J.T. Balino, and Charles Enoch. 1995.

125. United Germany: The First Five Years—Performance and Policy Issues, by Robert Corker, Robert A. Feldman, Karl Habermeier, Hari Vittas, and Tessa van der Willigen. 1995.

124. Saving Behavior and the Asset Price “Bubble” in Japan: Analytical Studies, edited by Ulrich Baumgartner and Guy Meredith. 1995.

123. Comprehensive Tax Reform: The Colombian Experience, edited by Parthasarathi Shome. 1995.

122. Capital Flows in the APEC Region, edited by Mohsin S. Khan and Carmen M. Reinhart. 1995.

121. Uganda: Adjustment with Growth, 1987–94, by Robert L. Sharer, Hema R. De Zoysa, and Calvin A. McDonald. 1995.

120. Economic Dislocation and Recovery in Lebanon, by Sena Eken, Paul Cashin, S. Nuri Erbas, Jose Martelino, and Adnan Mazarei. 1995.

119. Singapore: A Case Study in Rapid Development, edited by Kenneth Bercuson with a staff team comprising Robert G. Carling, Aasim M. Husain, Thomas Rumbaugh, and Rachel van Elkan. 1995.

118. Sub-Saharan Africa: Growth, Savings, and Investment, by Michael T. Hadjimichael, Dhaneshwar Ghura, Martin Mühleisen, Roger Nord, and E. Murat Uc, er. 1995.

117. Resilience and Growth Through Sustained Adjustment: The Moroccan Experience, by Saleh M. Nsouli, Sena Eken, Klaus Enders, Van-Can Thai, Jorg Decressin, and Filippo Cartiglia, with Janet Bungay. 1995.

116. Improving the International Monetary System: Constraints and Possibilities, by Michael Mussa, Morris Goldstein, Peter B. Clark, Donald J. Mathieson, and Tamim Bayoumi. 1994.

115. Exchange Rates and Economic Fundamentals: A Framework for Analysis, by Peter B. Clark, Leonardo Bartolini, Tamim Bayoumi, and Steven Symansky. 1994.

114. Economic Reform in China: A New Phase, by Wanda Tseng, Hoe Ee Khor, Kalpana Kochhar, Dubravko Mihaljek, and David Burton. 1994.

113. Poland: The Path to a Market Economy, by Liam P. Ebrill, Ajai Chopra, Charalambos Christofides, Paul Mylonas, Inci Otker, and Gerd Schwartz. 1994.

112. The Behavior of Non-Oil Commodity Prices, by Eduardo Borensztein, Mohsin S. Khan, Carmen M. Reinhart, and Peter Wickham. 1994.

111. The Russian Federation in Transition: External Developments, by Benedicte Vibe Christensen. 1994.

110. Limiting Central Bank Credit to the Government: Theory and Practice, by Carlo Cottarelli. 1993.

109. The Path to Convertibility and Growth: The Tunisian Experience, by Saleh M. Nsouli, Sena Eken, Paul Duran, Gerwin Bell, and Zuhtü Yücelik. 1993.

Note: For information on the title and availability of Occasional Papers not listed, please consult the IMF Publications Catalog or contact IMF Publication Services.


This review covers only countries that began ESAF-supported programs before December 31, 1994, on the grounds that the experience of those starting later than this date would be too recent to evaluate fairly. All SAF and ESAF multiyear arrangements ap-proved before December 31, 1994, are covered.


Bredenkamp and Schadler (forthcoming) and Abed and others (forthcoming).


Schadler and others (1993).


Such action was taken—first, with the introduction by the Paris Club of highly concessional terms in December 1994 (the so-called Naples terms), and subsequently by agreement on the debt initiative for the heavily indebted poor countries sponsored by the World Bank and the IMF.


In this study, the terms “ESAF countries” and “ESAF users” are used interchangeably to refer (solely) to the countries under review.


Bredenkamp (forthcoming) discusses these methodological is-sues and refers to some key papers on the subject.


See Schadler and others (1993) for a fuller description of the state of countries’ economies prior to embarking on SAF/ESAF-supported adjustment programs.


The CFA franc zone countries achieved significant savings in the government wage bill by containing nominal wage growth following their 1994 devaluation.


Unless otherwise specified, the observations in this section are based on a restricted sample of 28 (out of 36) countries and 47 (out of 68) SAF/ESAF-supported programs, because of the absence of three-year-ahead fiscal targets for some programs.


These findings are taken from Abed and others (forthcoming), who examine revenue and expenditure policies in SAF/ESAF-supported programs in more detail, based on one-year-ahead or within-year targets. An analysis on three-year-ahead targets, to be consistent with the rest of this section, was not possible at this level of disaggregation, since medium-term program targets were either not set or not reported in the necessary detail. Three-year-ahead targets were generally more ambitious than nearer-term targets.


The tendency to sacrifice capital before current spending may arise from the perception that cuts in public investment are politically less costly. For evidence, see Haggard and others (1995).


This assortment of outcomes—with some programs recording expenditure overruns and others a combination of revenue and expenditure shortfalls—explains why, in the aggregate (see Figure 3), expenditures appear roughly in line with targets on average, while revenues are lower than targeted. Dicks-Mireaux and others (forthcoming) provide detailed analysis and statistical testing of these findings.


These averages conceal some notable successes in raising capital spending, however. In Bolivia, Ghana, Mozambique, Nicaragua, and Uganda, the share of capital spending in GDP has increased by at least 4 percentage points since the pre-SAF/ESAF period.


References in this section to “low” inflation mean an annual rate below 10 percent, “intermediate” means 10-40 percent, and “high” means above 40 percent.


Bolivia and Nicaragua had largely defeated their near-hyper-inflations in earlier programs supported by stand-by arrangements with the IMF, but they consolidated this progress in their subsequent SAF- and ESAF-supported programs.


Countries in the CFA franc zone had foreign exchange systems that were free of restrictions on current transactions throughout the period under review.


Among ESAF transition countries, the Krygyz Republic and Mongolia have also accepted the obligations of Article VIII.


The indices of trade reform do not, however, capture use of nontariff import charges (see below).


In this context, “moderately restrictive” means a tariff structure with average tariff rates of 20-25 percent (maximum 30—40 percent) and few exemptions.


All five of the countries examined in the case studies reported in Decressin and others (forthcoming) provided some form of assistance—usually severance benefits, but sometimes also retraining support—for employees laid off during restructuring of public enterprises.


The criteria are the quality and acceptance of the tools of law and order, the quality of public administration and extent of government corruption, and contract and expropriation risks to foreign investors.


Countries in the CFA franc zone had negligible premiums throughout, owing to the absence of significant exchange restrictions in CFA franc regime.


The average population in Asian ESAF countries is 6 to 15 times larger than the average population in African or Western Hemisphere ESAF countries.


The turnaround is greater still when expressed in terms of median growth rates, which rose from –1.9 percent a year in 1981-85 to 0.6 percent a year in 1991-95. The difference reflects a more asymmetric distribution of growth rates in the recent period.


Uganda’s real per capita GDP growth averaged 3.7 percent a year between 1986 and 1995, compared with an average of 0.9 percent for all developing countries. All four Asian ESAF users also had growth well in excess of the developing country average over this period. Although full time-series data are not available for the transition economies, those in Indochina (Cambodia, the Lao People’s Democratic Republic, and Vietnam) recorded average real per capita growth of over 4 percent during 1991-95.


Formally, in a growth equation estimated on pooled data for a large sample of developing countries, the hypothesis of stable parameters between the ESAF sample and other developing countries could not be rejected by the data. Also, a dummy variable distinguishing ESAF from other countries was not significant. The policy and other variables used in this analysis are listed in the notes to Figure 16.


Although the latter is proxied in the equation for growth by life expectancy, it can be interpreted as encompassing general standards of health and education. Empirical studies of growth commonly find an important influence from primary and secondary education levels.


This increase is less than the debt buildup experienced by all low-income countries over the same period (170 percent), but is similar to the average for developing countries as a whole.


The internal transfer burden is sometimes alternatively referred to, in countries where the bulk of external debt is public, as the “fiscal” burden, and measured in relation to fiscal revenues. The debt-service-to-revenue ratio may, however, understate the debt burden when government revenues are unsustainably high as a share of GDP; conversely, when the fiscal system is in disarray and revenues are “too low,” it can misidentify as a debt problem what is more properly viewed as a problem of fiscal administration. Using the ratio of debt service to GDP is a way of avoiding this potential difficulty.


Note that the coverage of the HIPC group in Figure 16 excludes transition economies and Tanzania, owing to problems of data availability and reliability.


Countries that may have weakened in this respect are Malawi, Mauritania, Senegal, and Togo; for others, the direction of change is unclear.


Failure to make progress does not necessarily mean that these countries now face an unsustainable debt situation: for example, Kenya (a HIPC) and Zimbabwe (a non-HIPC) have had relatively moderate debt burdens throughout the period since their SAF/ESAF-supported adjustment efforts began. More generally, debt sustainability must be assessed on a forward-looking basis.


Two pieces of evidence can be offered on this point: first, since beginning SAF/ESAF-supported adjustment, the proportion of years in which annual programs were completed was much less (about 35 percent) for the “no” progress group than for the other groups combined (about 55 percent); and second, the “no” progress group achieved significantly less fiscal adjustment than the “clear” progress group over the same period (see Figure 19).


Tsikata (forthcoming) reports additional evidence in support of this observation. It is shown there that progress toward external viability is not associated with import compression: as noted above, those making most progress toward this goal had, on average, the highest annual rate of import growth.


Such targets were found to be quite stringent, often disallowing new nonconcessional borrowing altogether, though their coverage was limited (inter alia due to the exclusion of long maturities).


An “uninterrupted program” is defined as one where the supporting arrangement ran its full course and where delays (if any) in completing reviews did not exceed six months (see the section on “Sustaining Programs”).


These 14 cases were those that had, at some point, experienced program interruptions of at least a year.


As is well established in the literature, GDP growth is highly correlated with improved living standards and reductions in poverty (see Jayarajah and others (1996)).


For a survey of this evidence, see Goldsbrough and others (1996). The main factor driving private saving appears to be the growth of income, implying that strategies to raise growth will in-directly stimulate private saving.


See Abed and others (forthcoming). The “level of development” is broadly equated here with per capita income levels.


The reference to prioritization here is important: it would not necessarily be desirable to include, for instance, spending on university education or specialized curative health services in the core budget.


See Phillips (forthcoming) for a detailed review of the evidence on these points and on the other empirical findings and analytical issues referred to in this section. A similar conclusion was reached in “The Rise and Fall of Inflation—Lessons from the Postwar Experience” (see International Monetary Fund (1996), especially pages 116–22).


The data are not adequate to identify the precise source of these inflows. As indicated in Table 7, they appear not to reflect unanticipated official borrowing. The bulk is recorded in the miscellaneous (residual) category “net private capital and balance of payments errors,” the nature of which is highly uncertain.


In principle, inflation control could be achieved with commitments to NDA ceilings and a pure floating regime for the exchange rate (no discretionary intervention in the foreign exchange market), since the latter insulates the economy from accommodating external flows. Few governments would commit to a pure float, however, and there is no example of such a regime among the countries under review. Chile provides an example of the successful use of preannounced inflation targets as a nominal anchor.


The shortcomings of anchorless programs were examined at length in Schadler and others (1995a and 1995b).


The two instances were Bolivia (1988 ESAF) and Senegal (1995 ESAF). The other countries examined were Ghana, Mongolia, and Zimbabwe; see Decressin and others (forthcoming).


In Ghana, earlier recognition of financial weaknesses in the state petroleum company (GNPC) might have brought this enterprise into the reform program before it put at risk the stabilization objectives of the IMF staff-monitored and ESAF-supported programs of the early 1990s.


The five countries examined in this part of the study were Bolivia, Ghana, the Lao People’s Democratic Republic, Senegal, and Tanzania.


By contrast, the comparative success of banking reforms in Ghana and Senegal appears to owe much to their comprehensive initial planning.


This occurred in Bolivia, the Lao People’s Democratic Re public, and, to some extent, in Tanzania.


Best practices could be drawn from Basle Committee’s Core Principles for Effective Supervision; see Basle Committee on Banking Supervision (1997).


Excluded from these calculations were SAF arrangements that were replaced before completion with ESAF arrangements, arrangements that had an original duration of less than three years, and three-year arrangements that have not yet run their course.


This result was obtained by regressing the targeted change in the fiscal balance (excluding grants) on the previous year’s fiscal balance and a constant, for all annual SAF/ESAF-supported programs for which data are available (excluding some extreme outliers). A dummy variable for years in which a program was interrupted because of policy slippages was found to be statistically insignificant.


For these purposes, reflecting data availability, the wider sample was taken to be all ESAF-supported programs approved since 1991.


In fact, annual ESAF arrangements have no test date or disbursement at end year, implying that during the second half of each program year there is no direct penalty for poor policy performance—a penalty applies only indirectly, through the prospect of more difficult negotiations and perhaps harsher corrective measures to secure approval of the subsequent annual arrangement.

Economic Adjustment and Reform in Low-Income Countries
  • View in gallery

    Association of Real Per Capita GDP Growth with Log Inflation1

    (Mean values within groups ranked by inflation)2

  • View in gallery

    Real Per Capita GDP Growth in SAF/ESAF Programs1

    (Median; percent change over previous year)

  • View in gallery

    Inflation Performance With and Without Anchors1

    (Sample medians; in percentage points)

  • View in gallery

    Frequency Distribution of Changes in the Fiscal Balance1

    (In percent of GDP, excluding grants)

  • View in gallery

    Targeted Change in Fiscal Balance in Interruptions Affected by Policy Slippages1,2

    (In percent of GOP, excluding grants)

  • View in gallery
  • View in gallery