I Introduction and Summary

Benedict Clements, Liam Ebrill, Sanjeev Gupta, Anthony Pellechio, Jerald Schiff, George Abed, Ronald McMorran, and Marijn Verhoeven
Published Date:
March 1998
  • ShareShare
Show Summary Details

Since 1986 the International Monetary Fund has supported the adjustment programs of its low-income members with loans on highly concessional terms through two arrangements, the Structural Adjustment Facility (SAF) and the Enhanced Structural Adjustment Facility (ESAF) (Box 1). These facilities rest on two premises: first, that macroeconomic stabilization and structural reform of economic systems and institutions complement each other and second, that both are needed for economic growth with external viability.

At the heart of structural adjustment undertaken by developing countries lies the fiscal component of the program. Sound fiscal policy fosters macroeconomic stability. For one thing, sound fiscal policies work against inflation, which is important given the relatively limited financial intermediation in most SAF/ESAF countries and the associated potential for budget deficits to be monetized. Moreover, well-designed revenue and expenditure reforms may directly promote growth as well as enhance the economy’s supply-side response.1 More efficient revenue mobilization, for example, can fund needed public goods and services and help cut fiscal imbalances as well as promote investment and growth by reducing the adverse allocational effects of the tax system. On the expenditure side, more cost-effective public programs free resources for better uses, and changes in the composition of public spending can reorient public resources away from current consumption and toward growth-promoting investment in physical infrastructure and in social and human capital.

This paper briefly reviews aggregate fiscal developments, particularly those concerning the fiscal deficit and its financing. It focuses, however, on developments in the levels and composition of revenue and expenditures, and on reforms in revenue and expenditure policies and administration in the context of SAF/ESAF-supported programs. Thus, although much of the analysis takes place at a relatively disaggregated level, the policies described were generally guided by broad objectives of fiscal sustainability and other macroeconomic considerations. Finally, to keep the analysis manageable, the treatment of “revenue” and “expenditures” does not specifically address quasi-fiscal operations that, in some cases, may be significant.

As regards the methodology, we analyze revenue and expenditures on two levels. On one level, the analysis presents an aggregated overview of the experience of all 36 SAF/ESAF countries in the sample during 1985–95 (Box 2).2 On another level, we examine the experiences of individual countries (summarized in Boxes 310). In the aggregated analysis, we review the program targets for the overall levels of revenue and expenditures and their individual components (generally as a percentage of GDP) using widely accepted best practices in tax and expenditure policy. We then compare the targets with outcomes before and after program initiation. Differences between targets and actuals are then analyzed to shed light on the reasons for the “success” or “failure” to achieve prescribed targets. We also review the longer-term impact of the adjustment efforts of SAF/ESAF countries by comparing revenue and expenditure outcomes in the period just before program initiation to those in the latest year (1994 or 1995) for which data were available.

The targets used for analysis are the latest revised annual targets for programs agreed to by the IMF and the countries’ authorities, although estimates based on the original, unrevised annual targets are also reported. Most programs also established targets for the next three years at the time of initial approval. These targets generally cover macroeconomic indicators such as the overall deficit, and sometimes revenue or expenditures, although rarely their components. Thus, the choice of annual targets reflects both the absence of the necessary details in the original three-year-ahead targets, and a concern that any detailed three-year projections that did exist were highly tentative. However, for countries for which data are available, revenue and expenditure outcomes are also compared to the initial three-year targets.

Box 1.SAF/ESAF: A Concessional Facility to Assist Poorer Countries

The IMF’s Executive Board established the Enhanced Structural Adjustment Facility (ESAF) in 1987 to better address the macroeconomic and structural problems faced by low-income countries. It offers loans with lower interest rates and for longer terms than the typical IMF market-related arrangements. The principal objectives are to promote balance of payments viability and foster sustainable long-term growth. Although the objectives and features of the ESAF are similar to those of its predecessor, the Structural Adjustment Facility (SAF), set up in 1986, the ESAF was expected to be more ambitious with regard to macroeconomic policy and structural reform measures. The IMF no longer makes disbursements under the SAF.

ESAF loans are disbursed semiannually (as against quarterly for regular IMF stand-by arrangements), initially upon approval of an annual arrangement and subsequently based on the observance of performance criteria and after completion of a mid-term review. ESAF loans are repaid in 10 equal semiannual installments, beginning 5½ years and ending 10 years after the date of each disbursement. The interest on ESAF loans is 0.5 percent a year. By contrast, charges for stand-by arrangements are linked to the IMF’s Special Drawing Right (SDR) market-determined interest rate, and repayments are made within 3¼ to 5 years of each drawing. A three-year access under the ESAF is up to 190 percent of a member’s quota. The access limits of stand-by arrangements are 100 percent of quota annually and 300 percent cumulatively.

An eligible member seeking to use ESAF resources develops, with the assistance of the IMF and the World Bank, a policy framework paper (PFP) for a three-year adjustment program. The PFP, updated annually, sets out the authorities’ macroeconomic and structural policy objectives and the measures that they intend to adopt during the three years. The PFP also lays out the associated external financing needs of the program, a process that is meant to catalyze and help coordinate financial and technical assistance from donors in support of the adjustment program.

Performance during the first year of an ESAF arrangement is usually monitored by quarterly quantitative and structural benchmarks that reflect the program’s key elements. Quantitative benchmarks typically include monetary, fiscal, international reserves, and external debt variables, whereas structural benchmarks typically cover major institutional reforms, such as in the areas of public enterprise, the financial sector, structural fiscal policy, or tax and expenditure management. For the midterm review, the semiannual quantitative bench-marks usually serve as performance criteria, although key structural reforms may also be included. Currently, 79 countries are eligible for ESAF loans; at end-April 1997, 35 ESAF arrangements were in effect, with cumulative commitments totaling SDR 8.8 billion (about $12 billion)—of which SDR 7.2 billion (about $8 billion) had been disbursed.1

1 The SDR, an international reserve asset created by the IMF and allocated to its members since 1970, is the IMF’s unit of account. Its value is determined by the market-exchange rates of a basket of five currencies—the U.S. dollar, the deutsche mark, the French franc, the Japanese yen, and the pound sterling.

To take account of the broader social impact of government spending, this paper also analyzes trends in SAF/ESAF countries’ expenditures on education and health and changes in social indicators.

The principal findings on the overall fiscal performance and revenue and expenditure policies and performance are summarized below.

Major Findings on Overall Fiscal Performance

Programs envisaged relatively modest adjustments in the overall fiscal balance—on average, the overall fiscal deficit was slated to decrease about 1 percentage point of GDP from the pre-program year. Programs sought greater change in the financing of deficits, aiming to shift away from domestic financing and the accumulation of payment arrears toward foreign financing (most of which was highly concessionary). On average, programs targeted an annual reduction in payment arrears of 0.8 percent of GDP.

During the program periods significant progress was made with fiscal consolidation, with fiscal deficits falling most dramatically in those countries where, on average, initial deficits were highest. For example, in countries classified as having high initial deficits (see Box 2 for definitions), the average overall fiscal deficit shrank from 13.8 percent of GDP in the pre-program year to 9.5 percent of GDP in the most recent year. Also, there was a broad movement away from domestic financing, by an average of 2.1 percent of GDP, but the pattern of financing did not shift as much as anticipated. The annual accumulation of arrears fell less than programmed.

While countries, on average, met the targets for the overall deficit, outcomes varied considerably by region. Transition countries outperformed expectations (albeit from excessively high initial deficits), economic performance in Asian and African countries fell slightly short of program objectives, and performance in Western Hemisphere countries fell significantly short of targets.

Box 2.SAF/ESAF Sample Countries and Definitions



Burkina Faso


Côte d’Ivoire

Equatorial Guinea

Gambia, The












Sierra Leone





Europe and Central Asia


Kyrgyz Republic


East and South Asia



Lao People’s Democratic Republic



Sri Lanka


Western Hemisphere





The pre-program year is defined as the year preceding the first SAF/ESAF arrangement for each country.

Revenue ratios refer to revenue-to-GDP ratios unless otherwise stated.

Initial revenue efforts (ratios) were based on total revenue as a percentage of GDP, classified as follows: low, between 5 percent and 9.9 percent; medium, between 10 percent and 19.9 percent; and high, more than 20 percent.

High-deficit countries are those with deficits greater than 10 percent of GDP; medium-deficit countries, between 5 percent and 9.9 percent of GDP; and low-deficit countries, less than 5 percent of GDP.

Major Findings on Revenue Mobilization

The analysis shows that improving the structure and administration of tax systems requires significant tax policy reforms, such as simplifying and rationalizing tariff and tax rates and introducing broad-based taxes on consumption, for example, value-added taxes (VATs). To improve tax and customs administration, these tax policy reforms need to be reinforced by appropriate institutional reforms.

SAF/ESAF-supported programs targeted, on average, a modest increase for both total revenue and tax revenue as a percentage of GDP. Programs in the CFA franc zone envisaged that revenue effort would remain unchanged. The revenue objectives indicated relatively large increases in the revenue-to-GDP ratio for countries with low initial revenue effort, smaller increases for medium initial effort countries, and virtually no increase for high initial effort countries (see Box 2 for definitions).

As regards program implementation, total revenue fell short of annual program targets by about 0.5 percent of GDP on average (somewhat greater when compared with the initial three-year targets), with adverse implications for expenditures. Underperformance was most notable in Africa. Latin American and transition countries, however, generally met revenue targets.

With respect to revenue components, the yields of consumption-based taxes and international trade taxes, although they rose during the program period, fell slightly short of targets, whereas those of income and profits taxes exceeded their targets by an annual average of 0.7 percent of GDP. Comparing the pre-program situation with the most recent data available shows that revenue components changed in the direction intended: tax revenue rose by 1.4 percent of GDP and nontax revenue declined by 0.4 percent. Half of the tax revenue gain was attributable to improved revenue mobilization from taxes on domestic goods and services.

A number of countries introduced or modified their VATs during program periods. Some of the VATs, however, contained undesirable features that led to mixed revenue results. In cases in which VATs conformed to widely accepted best practices, the results were generally positive.

Program countries that set specific benchmarks or prior actions for the reform of tariff policy and customs administration registered a stronger revenue performance than other program countries for international trade taxes. Insufficient data on revenue components precluded generalization of this result to other taxes.

Program shortfalls relative to targets were most pronounced for those countries with the lowest initial revenue effort. A number of these countries, however, did significantly improve their revenue performance, albeit less rapidly than programmed. Medium and high initial revenue effort countries performed better relative to targets.

Many SAF/ESAF participants that completed their programs on time also greatly improved their revenue yields. In general, however, the impact of interruptions in programs on revenue performance was ambiguous.

General Lessons for Tax Reform

Tax reform strategy should aim at simplifying and rationalizing tax and tariff structures and making greater use of broad-based consumption taxes, such as a modern VAT with, ideally, a single tax rate in the 15–20 percent range, a threshold to exclude small businesses, and the broadest possible coverage of goods and services.

Although some countries have resorted to short-term ad hoc revenue measures to meet immediate revenue needs, these measures should be viewed as temporary, and ESAF programs should remain centered on fundamental reforms in tax policy and administration.

Reform strategies should also seek to modernize tax and customs administrations, keeping in mind, however, that such institutional reforms take time to produce needed revenue gains.

The rise in revenue from international trade taxes in SAF/ESAF countries, contrary to long-term historical trends, is not necessarily inconsistent with structural reforms of the tax systems in SAF/ESAF countries over the average program period (about five years). While in some cases countries that collected more revenue from this source did so as a result of undertaking desirable macroeconomic or structural reforms (discussed below), others sought to safeguard revenue while domestic tax reforms were taking hold. This result has implications for the design and sequencing of tariff and tax reforms in the context of adjustment programs.

As noted above, some countries with low initial revenue effort failed to improve revenue performance as rapidly as envisaged. This result implies that programs designed to raise revenue over time in low-income countries should take account of administrative constraints. Generally, developing countries with low revenue-to-GDP ratios (around 10 percent) should emphasize efficient revenue mobilization as a key element in rectifying fiscal imbalances. On the other hand, at much higher revenue-to-GDP ratios (generally more than 20 percent), persistent macroeconomic imbalances would call for shifting fiscal consolidation efforts toward containing expenditures.

Major Findings on Expenditure Reform

Annual targets in SAF/ESAF programs have aimed, on average, to maintain total expenditure as a share of GDP while shifting the composition from current spending to capital spending. Relative to average spending in the three years preceding the program, these programs envisaged that capital expenditure would rise by an average 1.4 percentage points of GDP and that current spending would fall an average 2.2 percentage points of GDP. According to initial three-year-ahead targets, SAF/ESAF countries sought to reduce noninterest public spending by 1.9 percentage points of GDP.

Total expenditure averaged some 0.5 percent of GDP less than programmed, with lower-than-programmed spending traceable to shortfalls in revenue and in foreign financing. Revenue shortfalls led to lower-than-programmed capital expenditures in over three-fourths of the countries. Outcomes from the initial three-year-ahead targets deviated considerably more than those from annual targets.

Over the typical program period, SAF/ESAF countries succeeded in changing the composition of spending in the direction envisaged. The share of capital spending in the total increased and that of current spending declined, although changes often fell short of the extent envisaged.

Within current spending, the share of wages and salaries in GDP was, on average, slightly higher than targeted but substantially lower than in the pre-program period. The overall decline in wages and salaries as a share of GDP was due more to real wage shrinkage than to the employment cutbacks envisaged under civil service reforms.

The allocation of outlays by function changed significantly, with less spending on military, general public, and economic services (essentially subsidies and transfers) and more spending on education and health.

Expenditure management improved in a number of SAF/ESAF countries, aided by extensive technical assistance from the IMF. Shortcomings in expenditure management have persisted, however, and have impeded fiscal adjustment and structural reforms in many countries.

Temporary social safety nets were incorporated into many SAF/ESAF programs to mitigate the short-term adverse effects of price increases and reduced job opportunities. In some cases, implementation proved difficult because social policy instruments were lacking and targeting of benefits was imperfect. Evaluating the implementation and effectiveness of these safety nets suffers from insufficient data.

Countries with no program interruptions generally had lower expenditure-to-GDP ratios than programmed. Difficulties in timely completion of programs are linked to overruns in current outlays.

Education and health spending rose in real terms and in relation to GDP after the initiation of the first program in SAF/ESAF countries. Real annual spending per capita on education and health rose by 3.8 percent and 5.8 percent, respectively. In both cases, results varied considerably among countries. Key social indicators also improved in SAF/ESAF countries.

General Lessons for Expenditure Reform

Programs should incorporate explicit, monitorable quantitative targets for reductions in public employment. These targets should be based on actual numbers of workers rather than positions. Greater progress in this area will likely require larger input from the World Bank in support of civil service reform. In addition, programs should focus on a medium-term plan, rather than one-shot reductions, and on strengthening or creating institutions that ensure control over recruitment and the civil service payroll.

Considerable scope exists for increasing the level, efficiency, and benefit incidence of social spending. Progress in this area is currently constrained, however, by the limited data on functional categories of spending, particularly social spending. Thus, comprehensive and timely data on expenditures by function should be compiled and made available so that adjustment programs can incorporate targets that are realistic, easy to monitor, and supported by underlying analysis.

Social indicators in ESAF countries should be monitored on a continuous and systematic basis as they evolve.

The reform of budgeting and expenditure control systems should place greater emphasis on improving the quality of human capital, providing appropriate incentives for officials charged with carrying out the reforms and ensuring transparency and accountability. Programs should also consider making greater use of prior actions in this area.

Because revenue shortfalls often adversely affect expenditure composition and the accumulation of arrears, contingency measures on the expenditure side need to be considered systematically and a core budget of high-priority allocations that would be protected from ad hoc cuts specified.

Capital spending targets should be based on realistic expectations of the capacity for project implementation, and care should be taken to protect essential public investment from budget cuts.

Social safety net measures figure prominently in programs, but information concerning their impact on the targeted population groups is scarce. Greater effort should be made to follow up on these measures, particularly to ascertain whether programs are reaching their intended beneficiaries.

2Note that fiscal data for these countries were not always available in the needed detail for all periods considered; thus, the analysis was based, in some instances, on a smaller sample.

    Other Resources Citing This Publication