IV Fiscal Policy Issues in Developing Countries
- International Monetary Fund. Research Dept.
- Published Date:
- May 1996
The strengthening in the growth performance of many developing countries in recent years has reflected fundamental changes in economic policies. A major role has been played by fiscal policies, which have been re-oriented toward the objectives of a more stable macroeconomic environment, stronger domestic saving and investment, and market-oriented structural reform. This re-orientation has required not only greater discipline in containing fiscal imbalances but also a scaling back of the role of government in the economy to ensure that state intervention does not impede private sector development. Addressing fiscal imbalances has also facilitated economic liberalization and structural reform by reducing governments’ budgetary dependence on distortionary taxes and controls.
Policy improvements, however, have been uneven across countries, and this is reflected in significant variations in economic performance. Some developing countries have made limited progress with fiscal consolidation, which has hindered their growth and increased their vulnerability to changes in external conditions. For these countries, policy measures on a number of fronts, including public sector reform, privatization of state-owned enterprises, trade and financial liberalization, and strengthening tax administration, would help raise living standards, increase the pace of development, and reduce the risk of falling further behind other developing countries that have made greater progress in adjustment and reform. All developing countries, however, face the challenge of sustaining improvements in policies to safeguard financial stability, to promote efficient resource allocation and growth, and thus to maximize the benefits and minimize the risks associated with increasingly globalized financial and product markets.
There has been a marked improvement in the fiscal positions of a large number of developing countries since the 1980s. For the developing countries as a group, the central government deficit-to-GDP ratio declined by 2½ percentage points between 1983–89 and 1990–95 (Table 14).42 For most countries, fiscal adjustment has been achieved primarily through the containment of expenditure, with a marginally greater reduction in capital expenditure relative to GDP than in current expenditure. With regard to revenue, for developing countries as a group the ratio of central government revenue to GDP has followed a downward trend since around 1980 (Chart 24). This mainly reflects the impact on oil producers of weakening oil prices and difficulties in revenue mobilization, although there has also been a trend in some countries toward lower taxation as a means of improving incentives for private sector activity and thereby promoting growth. Although saving and investment levels have been affected by a number of factors, including the availability of external finance, fiscal consolidation since the late 1980s and associated improvements in the economic environment are likely to have been a key factor contributing to the increase in private saving and investment in the developing countries as a group. Compared with the 1983–89 period, both the ratio of private saving to GDP and the ratio of private investment to GDP were about 3 percentage points higher in 1990–95; and between the same periods average annual GDP growth in the developing countries increased by almost 1 percentage point.
|Central government fiscal balance1||-3.4||-5.5||3.0|
|Central government revenue1||20.7||19.2||18.1|
|Central government expenditure||23.7||24.4||20.9|
|Current account balance||-0.8||-1.5||-1.5|
|Terms of trade2||1.7||-0.6||—|
|Consumer prices (median)2||11.8||8.7||10.1|Chart 24.Developing Countries: Fiscal Deficits
(In percent of GDP)
Deficits have fallen in most regions, although improvements in Africa, and the Middle East and Europe are more recent.
1 Including grants.
The decline in fiscal deficits in developing countries as a group conceals significant variations both among and within regions (Table 15). In Africa, despite substantial improvements in macroeconomic and structural policies in a number of countries, such as Kenya, Uganda, and more recently the CFA countries, the average fiscal deficit for the region widened in the early 1990s.43 Many African countries faced unfavorable external conditions for much of this period, with falling commodity prices and lower industrial country growth leading to substantial declines in the terms of trade, especially for sub-Saharan African countries. While some of these countries have managed to make modest progress in reducing fiscal deficits, in part because of external debt rescheduling agreements in the late 1980s and early 1990s, which altered the time profile of interest payments, most of the adjustment has been in the form of lower capital expenditure. In fact, for Africa as a whole, the ratio of current government expenditure to GDP increased by over 2 percentage points in the 1990–95 period, while the ratio of capital expenditure to GDP was marginally lower than in the mid-1980s. The average fiscal deficit declined in 1995 to just under 4 percent of GDP, but for most African countries the fiscal situation remains fragile, with many countries still heavily dependent on grants to finance large fiscal imbalances.
|Central government fiscal balance1||-4.4||-4.8||-5.3|
|Central government revenue1||22.8||23.0||23.2|
|Central government expenditure||27.2||27.8||28.5|
|Current account balance||-4.7||-2.6||-3.8|
|Terms of trade2||-0.3||0.4||-0.8|
|Consumer prices (median)2||12.5||9.4||12.5|
|Central government fiscal balance1||-3.1||-3.4||-2.3|
|Central government revenue1||19.3||18.2||15.6|
|Central government expenditure||22.4||21.6||17.9|
|Current account balance||-1.4||-0.9||-0.5|
|Terms of trade2||0.8||—||0.3|
|Consumer prices (median)2||8.5||6.7||8.6|
|Middle East and Europe|
|Central government fiscal balance1||-3.2||-12.2||-7.2|
|Central government revenue1||40.3||28.9||27.3|
|Central government expenditure||43.5||41.1||34.5|
|Current account balance||7.7||-3.4||-2.7|
|Terms of trade2||4.7||-2.4||-0.8|
|Consumer prices (median)2||11.4||6.0||8.3|
|Central government fiscal balance1||-2.3||-4.4||-0.3|
|Central government revenue1||10.2||14.1||17.5|
|Central government expenditure||12.6||18.4||17.8|
|Current account balance||-3.8||-1.0||-2.1|
|Terms of trade2||1.3||-0.4||0.4|
|Consumer prices (median)2||13.2||12.5||14.0|
In the Middle East and Europe, although there has been a marked reduction in fiscal deficits since the mid-1980s—a decline of 5 percentage points in the deficit-to-GDP ratio—the average deficit, at over 7 percent of GDP, is still substantially larger than in other developing country regions. Many countries in the region were characterized by excessive growth of government outlays during the 1980s, following the jumps in oil prices and revenues of the 1970s and early 1980s. The subsequent fall in oil prices and the associated decline in government revenues left many countries with unsustainable expenditure programs and clearly unsustainable fiscal imbalances. The significant reduction in government expenditures that has since occurred is primarily due to lower capital spending: the ratio of current expenditure to GDP has been broadly stable at about 26 percent since the 1970s. There are, however, some notable differences from these average developments among the more diversified economies of the region. In Egypt, for example, a broadening of the tax base in recent years has led to increased tax revenues. Jordan, on the other hand, has been able to increase its capital expenditures as a percentage of GDP since 1992 by reducing current spending.44
The strongest improvement in fiscal balances has been among the developing countries in the Western Hemisphere, where several large countries, including Argentina, Brazil, and Mexico achieved fiscal surpluses in the early 1990s. For the region as a whole, fiscal deficits fell from almost 4½ percent of GDP in the mid-1980s to under ½ of 1 percent of GDP in 1990–95. In contrast to the period after the debt crisis of the early 1980s, when declining world interest rates and rising industrial country growth helped to lower the region’s fiscal deficits, progress with fiscal consolidation in the 1990s has been primarily due to stronger budgetary policies. Substantially lower inflation in the 1990s has also led to a marked reduction in central government interest payments in countries such as Argentina and Mexico. Reductions in central government expenditure and increases in public sector saving did not, however, lead to a corresponding increase in national saving, because of an offsetting decline in saving by the private sector. A continuing concern for a number of countries in the region is the extent to which capital inflows have substituted for shortfalls in the domestic saving-investment balance, which arguably may call for further improvements in fiscal positions to strengthen national saving.
The aggregate data for developing countries in Asia reflect the fact that many countries in the region have a long track record of prudent macroeconomic policies. For most of the period since the mid-1970s, fiscal deficits in many of the high-performing Asian economies have been maintained at levels well below those observed in other developing country regions. Most Asian countries have also managed to maintain relatively high private saving rates, which have helped them avoid resorting to monetary financing of fiscal deficits. Moreover, fiscal deficits declined further, on average, in the 1990–95 period, and in contrast with many other developing countries, this fiscal adjustment has been achieved largely by reducing current expenditure. This aggregate picture, however, again masks considerable variations within the region. In Pakistan, for example, persistent fiscal imbalances have contributed to low national saving and investment, thereby impeding growth performance, while in other South Asian countries, such as Bangladesh, although there has been some improvement in fiscal positions in recent years and domestic saving rates are higher than in many developing countries, they are substantially below saving rates in the faster growing economies of Asia.
Underlying Causes of Fiscal Deficits
For many developing countries, widening fiscal deficits in the 1970s and 1980s were largely a result of public expenditure expanding unsustainably fast—much faster than the tax revenue base of the economy. This unsustainable growth of public expenditure can be traced to a number of causes, including external factors such as increases in global interest rates, which raised the cost of servicing these countries’ growing external debt. But in many cases, upward trends in public spending were associated with government attempts to accelerate the process of development and industrialization through state involvement in eco- nomic activities that, as became increasingly apparent, could have been carried out more efficiently by the private sector. During the 1980s, governments in many African countries, in particular, provided extensive fiscal subsidies to loss-making enterprises, including those involved in activities such as the marketing and distribution of agricultural products—activities where there is little evidence of genuine market failure—as well as quasi-fiscal support to central banks involved in the allocation of subsidized foreign exchange and credit to domestic enterprises. State intervention in the allocation of credit was also prevalent among many of the successful Asian countries during the early 1980s; although in countries such as Korea subsidized credit was allocated largely to export industries that had to meet the test of international competition, while in Indonesia credit controls were significantly reduced.
For some developing countries, civil wars, political instability, and regional tensions limited the extent to which military expenditures could be reduced during the 1980s, a period when global military spending fell by over 20 percent in real terms. In comparison with the 1970s and early 1980s, the ratio of military spending to GDP in all developing countries—over 5½ percent in 1972–85—declined to under 4½ percent in the second half of the 1980s, although in sub-Saharan Africa it increased by ½ of 1 percentage point of GDP.45 More recently, military spending has declined further in most regions, including sub-Saharan Africa.
In many countries, poor governance and a lack of accountability of the public sector have contributed to inadequate control of government expenditure and failure to ensure that expenditures are allocated efficiently and equitably to serve society’s priorities. Expenditure on prestige projects, or spending that rewards politically powerful groups or benefits only a small minority—often to the detriment of expenditure on basic social services—stems largely from asymmetries in the political costs and benefits associated with taxation and spending. In countries that have democratic governments, expansionary fiscal policies have tended to be synchronized with the electoral cycle, especially on the expenditure side.46 Such politically motivated increases in spending may be difficult to reverse, especially when they result in higher employment in the public sector or in quasi-public institutions.
Among the countries with relatively successful growth performance, such as many Asian and Latin American countries, fiscal imbalances have often reflected increasing needs for investment in infrastructure, investment in human capital through health and education spending, and improvements in public services, including through public sector wage expenditures. Spending in these categories, when well-designed and efficiently allocated, can enhance the productivity of the private sector and promote growth. In some cases, private sector involvement in the provision of infrastructure and other public services may help moderate demands on public expenditure, but fiscal imbalances will usually call for expenditures elsewhere to be contained or reduced.
In many developing countries, losses of public sector enterprises have frequently added to the government’s fiscal deficit: this is one reason why the central government’s accounts may flatter the public sector’s financial position, especially in the short run. In Kenya, for example, losses of the National Cereals and Produce Board during the 1980s—equivalent to about 5 percent of GDP a year—were eventually, though not contemporaneously, borne by the central government. In India, the cost to the central government of financing losses incurred by public enterprises amounted to almost 1½ percent of GDP a year in the early 1990s; subsequently, budgetary support to public enterprises was limited to just under 1 percent of GDP in 1995 and their access to subsidized bank loans was significantly curtailed. Public enterprises are often inefficient monopolies that are subsidized at the expense of the taxpayer, and whose activities do not need to be conducted in the public sector; their monopoly power frequently stems from entry restrictions on potential private sector competitors.
For a large number of public sector activities, such as central bank operations, and implicit subsidies in the form of government guarantees for borrowing by public enterprises, there are often no contemporaneous budgetary outlays; such quasi-fiscal operations frequently entail the creation of contingent or unfunded liabilities. This is another reason why the ratio of central government expenditure to GDP, which is typically about half that in industrial countries, may significantly understate the extent of public sector involvement in the economy. Difficulties in raising tax revenue through the budgetary process often help to explain why governments pursue fiscal objectives partly through quasi-fiscal means.47 Some activities of central banks and other public financial institutions, such as foreign exchange market operations and subsidized loans to public sector enterprises, although nontransparent and difficult to quantify fully, can entail considerable costs to the public sector as a whole. In Jamaica, for example, central bank losses from exchange rate guarantees exceeded 5 percent of GDP in the early 1990s. And in Mexico, the decline in the deficit of the nonfinancial public sector in the early 1990s overstated the improvement in the true fiscal position, partly because it did not reflect the quasi-fiscal costs of the large expansion in lending by the public sector development banks.
Apart from the losses of public enterprises and the costs of quasi-fiscal activities, expenditures may be devolved from central to local governments without corresponding financing. For all these reasons, improvements in the central government’s fiscal balance may misrepresent the true evolution of the public sector’s overall financial position. Reductions in central government expenditures may simply reflect a shift to quasi-fiscal activities and implicit government liabilities, rather than a genuine retrenchment of the public sector’s involvement in the economy. In Senegal, for example, while direct operational subsidies provided by the central government declined between 1985 and 1989, there was a large rise in central government overdrafts with quasi-public financial institutions. In such cases, the appearance of fiscal discipline in the central government’s accounts may be illusory, and an apparent improvement in the fiscal position may not promote macroeconomic stability simply because the public sector’s long-term financing requirement may fail to improve or may even deteriorate.
Difficulties in Revenue Mobilization
In a large number of developing countries, fiscal problems have been exacerbated by the presence of large sectors of informal (or “underground”) economic activity. Typically the result of the prevalence of price controls in many areas of the economy, high tax rates, and weak institutional and administrative structures, such informal activity often adds dynamism to an economy, but it also narrows the tax base. The lack of sufficiently developed tax administration and collection capabilities also limits the extent to which the formal as well as the informal sector can be taxed.
As a result of these difficulties, most developing countries have been heavily dependent on taxes on international trade and on domestic product markets (Chart 25). For developing countries as a group, trade taxes accounted for almost 30 percent of total tax revenue during the period 1975–90, compared with only 3 percent for industrial countries.48 In many of the more successful developing countries, especially in Asia and in Latin America, the relative importance of customs duties on imports and export taxes has declined in recent years, reflecting progress with trade liberalization and also domestic reforms. In many other developing countries, however, despite the negative economic effects of inward-oriented policies, trade barriers have been removed only gradually, in part because governments have remained dependent on trade taxes to finance spending programs.
Chart 25.Developing Countries: Sources of Tax Revenue
(In percent of total tax revenue)
Trade taxes still account for a large proportion of tax revenue in developing countries.
Source: IMF, Government Finance Statistics.
For countries that are heavily dependent on commodity exports, tax revenues are strongly affected by movements in commodity prices (Chart 26). When commodity prices rise—such as during the early and late 1980s—government revenues are boosted both directly in countries where commodity-producing sectors are state owned and indirectly through increased revenues from trade and income taxes. Governments in a number of commodity-exporting countries have tended to use these windfall gains to finance procyclical expenditures with the result that when commodity prices have declined, these countries have been left with unsustainable fiscal deficits. For some coffee-producing countries, such as Kenya and Tanzania, expansions induced by the coffee price booms of the late 1970s and early 1980s led to enduringly higher domestic and external debt burdens, largely because the windfall gains were used to fund public sector expenditures that yielded little or no return. Experience in many oil exporting countries has been similar. In fact, in Mexico, Nigeria, and Venezuela, following the oil price hike of the late 1970s, external borrowing increased long before oil prices began to fall.
Chart 26.Developing Countries: Commodity Prices and Tax Revenues of Commodity Exporters
1In SDRs: 1990 = 100.
2In percent of GDP.
Consequences for Economic Growth and Development
The relationship between fiscal policies and longerterm economic performance depends on a number of factors. Whether fiscal policies promote economic growth depends on the extent to which government expenditure is directed toward increasing the stock of productive physical and human capital; on the extent to which the provision of government services complements private sector activity; on the extent to which fiscal deficits crowd out private sector investment; on the effect of deficits on macroeconomic stability; and also on the parallel structural reforms needed to elicit the private sector’s supply response. In many countries, fiscal imbalances are the prime cause of low national saving and investment rates, while the development and efficiency of domestic financial systems frequently suffer from interest rate and credit controls, part of whose function is to facilitate the financing of fiscal deficits. In some instances, governments may be able to resort to foreign saving, often for relatively long periods, but the buildup of external imbalances with no counterpart in productive domestic investment increases a country’s vulnerability to sudden reversals of capital inflows with their disruptive consequences for domestic policies and economic activity.
Impeding Growth Prospects
A comparison of developing countries ranked by the size of fiscal deficits during the mid-1980s shows that the group of countries characterized by small fiscal deficits (or surpluses) had markedly higher growth rates than countries that had moderate or large deficits (Table 16). Large fiscal deficits may of course be partly a result of slow economic growth, especially over short periods. Moreover, both large deficits and slow growth may in some cases be partly due to difficult external conditions, such as weak terms of trade or high global interest rates. Nevertheless, the experience of many developing countries suggests that weak fiscal discipline, indicated by persistent fiscal deficits, does reduce growth over the longer term. Countries that had large deficits in the 1980s experienced real GDP growth of only about 3 percent in the 1990–95 period compared with growth of almost 7 percent in small deficit countries. However, within the group of countries that had large deficits during the 1980s, a number of countries that undertook strong fiscal adjustment programs that were sustained, such as Chile and Uganda, shifted to markedly steeper growth paths. In many other countries where fiscal adjustment programs were not sustained, growth declined in the 1990s.
|Small Fiscal Deficit||Moderate Fiscal Deficit||Large Fiscal Deficit|
|Central government fiscal balance2||0.2||-1.4||-1.2||6.7||6.0||-3.5||5.5||-12.1||-6.0|
|Consumer prices (median)||11.0||7.5||9.2||12.6||8.2||8.4||12.0||10.4||13.0|
|Current account balance||-2.0||-0.3||-0.1||-2.1||-2.8||-3.3||1.6||-3.1||-3.6|
|Terms of trade3||1.1||-0.5||0.4||0.7||-1.0||-0.4||3.4||-0.4||-1.0|
|Real effective exchange rate3||0.7||-3.8||2.5||-0.8||-1.1||2.0||1.1||-2.5||-3.5|
An important channel through which fiscal deficits damage growth performance is by reducing national saving and crowding out domestic investment, through effects not only on the cost and availability of finance but also on the real exchange rate and international competitiveness. Tables 14 and 16 suggest that fiscal discipline may have helped to boost not just national saving, but private saving: for example, in small deficit countries, the ratio of private saving to GDP was almost 4½ percentage points higher than in the large deficit countries during the 1980s. Moreover, these differences in private saving rates have increased markedly during the 1990s. There may have been a number of forces at work here, including the longerterm effects of structural reforms on economic incentives and the confidence-building effects of greater macroeconomic stability. Fiscal consolidation has also helped to reduce governments’ dependence on financial repression, thereby stimulating private saving. The usual conclusion of econometric work in this area is rather different: it is that reductions in fiscal deficits increase national saving by less than the decline in public sector dissaving, because of a partly offsetting decline in private saving.49 This still carries the implication that a smaller fiscal deficit is likely to promote private investment. The extent to which this raises productivity and growth then depends on the productivity of private investment relative to the productivity of the public spending that is displaced. Providing that public spending is not reduced in key areas where there are significant market failures—such as those arising from externalities to the provision of basic health care, education, and essential public services—overall productivity and growth are likely to benefit.50
The contribution of public expenditure to growth and welfare naturally depends on its composition. Expenditure on improving the provision of primary education and basic health services, on productive investment in such areas as transport and communication, and on essential government services, can be very effective in enhancing growth in developing countries, whereas increasing civil service employment to expand governmental administrative functions, or expenditure on military equipment, may reduce the productive capacity of the economy by limiting resources available to the private sector. The broad classification of public expenditure into current and capital categories, although convenient, does not always convey the productive potential of different kinds of spending. The provision of health services, for example, through the wages of nurses and doctors, is part of current expenditure, while construction projects—including prestige projects with low productive potential—are classified as capital expenditures irrespective of their contribution to productive capacity, as are purchases of commercial property. These examples help to indicate why recent studies that have compared the effects on growth of current and capital public expenditures have arrived at ambiguous results, although it has been found that growth has tended to be higher in countries that have increased the share of certain kinds of capital expenditure—such as infrastructure investment—in total government expenditure. 51 Furthermore, if it is debt financed, public capital expenditure may crowd out private capital expenditures. In India and Senegal, for example, increases in public investment were associated with significant declines in private investment during the 1980s.
In many developing countries, fiscal imbalances have been the underlying cause of macroeconomic instability, which, in turn, has impeded growth prospects. In countries where fiscal imbalances are large and perceived to be unsustainable, the uncertainty about impending policy changes, including higher taxes, expenditure cuts, and changes in interest rates and exchange rates, is likely to weaken the private sector’s confidence and reduce private sector investment expenditures. In India, for example, during the foreign exchange crisis of the early 1990s that was caused largely by a rapid buildup of domestic and external imbalances, gross fixed investment declined by 4 percent in 1991–92 and recovered only gradually over the next two years.52
Large fiscal deficits in developing countries have also been associated with rapid inflation, essentially because they are typically financed to a large extent by monetary growth. During the 1980s, many Latin American countries, such as Argentina, Brazil, and Mexico, experienced very high rates of inflation largely as a result of substantial fiscal deficits that could not be financed by borrowing on either domestic or international financial markets. Although some countries are still characterized by relatively high inflation, average inflation in the developing countries has declined significantly in recent years, owing largely to strong fiscal consolidation programs. Impressive achievements in this regard are those of Argentina, whose 12-month inflation rate by late 1995 had been reduced to 1½ percent, from over 2000 percent early in the decade, and Brazil, which lowered average monthly inflation from over 40 percent in the first half of 1994 to about 1½ percent in 1995.
Fiscal policy and inflation are also related by the fact that rapid inflation can provide an important, though unsustainable, source of budgetary revenue, through seigniorage, which consists primarily of the “inflation tax.”53 In the 1980s, a number of developing countries, including Ghana and Zambia, relied on the inflation tax as a significant source of revenue to finance their budgets. Over the longer term, however, higher inflation reduces the attractiveness of domestic monetary assets and this eventually limits the amount of revenue that can be generated. The decline of inflation has reduced the amount of revenue derived from inflation since the 1970s (Table 17).
|Small Fiscal Deficit||Moderate Fiscal Deficit||Large Fiscal Deficit|
|Real interest rate2||2.1||2.9||1.6||-5.6||0.8||1.3||-4.8||2.2||-2.5|
|Government borrowing from central bank||1.9||1.7||1.5||2.3||1.9||2.1||3.3||4.9||2.1|
|Private sector credit||29.7||41.7||53.7||20.8||29.7||32.5||15.4||21.8||20.7|
Governments in many developing countries have intervened in financial markets in attempts to finance deficits at below-market interest rates; this has been a major factor impeding the development of financial sectors and limiting intermediation in financial markets and the availability of credit to the private sector. Although some regulatory controls on financial markets are motivated by prudential concerns, many controls in developing countries, such as interest rate ceilings and statutory underrenumerated reserve and liquidity requirements on financial institutions, are primarily intended to ensure the provision of funding to governments at below-market rates (Box 7). It is difficult to estimate the revenue effectively generated by these policies since the extent of the subsidy depends on the differential between controlled interest rates and the interest rates that would prevail in the absence of controls, but in many countries, revenue from financial repression can be substantial.54
Many developing countries have managed to sustain relatively large fiscal deficits—in some cases averaging almost 10 percent of GDP over a number of years. Such deficits are unlikely to be sustainable, however, because government debt cannot grow faster than the economy in the long term.55 Moreover, delays in implementing adjustment measures can lead to substantial output losses when fiscal adjustment is forced upon governments by financial markets. During the 1980s, the fiscal deficit in Pakistan averaged over 7 percent of GDP and was financed largely through extensive controls on financial markets, relatively strong monetary growth, and external borrowing; growth also averaged about 6 percent a year in the 1980s. But in the early 1990s, adverse supply conditions increased the fiscal deficit to over 9 percent of GDP, and the growing external debt burden eventually led to a financial and exchange market crisis in 1993, followed by a sharp decline in growth to about 2½ percent. Large fiscal imbalances also contributed to the financial crisis in Turkey in 1994, which brought about a sharp decline in the availability of external finance and a deep recession; output declined by over 4½ percent in 1994 compared with over 8 percent growth in the previous year.
Benefits and Challenges of Fiscal Adjustment
In contrast with the 1970s and most of the 1980s, there is now widespread consensus on the need for fiscal discipline, on the associated benefits in terms of monetary and financial stability, and economic growth and development, and on the need for constraints to be placed on the size and role of the public sector. As the experience of many successful performers among developing countries illustrates, a key requirement for the promotion of growth is often to scale down the role of the state. For many developing countries, this requires reducing overemployment and increasing efficiency in the public sector, and implementing extensive privatization programs. But in many developing countries there is also scope—much more so than in industrial countries—to enhance revenues without worsening distortions and reducing efficiency, especially by broadening the revenue base and improving tax collection and administration. Some important reforms, however, including the liberalization of trade and the financial system, may have short-term negative implications for revenue that may need to be offset. For developing countries that have benefited from substantial foreign investment, both safeguarding financial stability and maintaining the momentum of reforms are essential for growth to be sustained.
Economic growth has been considerably higher in developing countries that have addressed fiscal imbalances decisively and implemented strong fiscal consolidation measures. A classification of developing countries based on the magnitude of fiscal adjustment—the change in the fiscal balance as a percent of GDP—between 1980–85 and 1990–95 shows that in the group of countries where fiscal balances improved the most, output was on average over 40 percent higher in 1990–95 than in 1980–85 (Chart 27). Although the strengthening of fiscal policies in these countries was in part attributable to favorable growth performances, prudent policies played a key role in sustaining relatively high growth. By contrast, in countries where fiscal balances worsened, growth has been significantly lower, and in some cases output has barely grown over the last fifteen years.
Chart 27.Developing Countries: Fiscal Adjustment and Growth
(Changes between 1980–85 and 1990–95)
Developing countries that have addressed fiscal imbalances have experienced large output gains.
1The magnitude of fiscal adjustment is evaluated as the change in the average fiscal balance between the period 1980–85 and 1990–95. Countries are ranked on the basis of the size of the fiscal adjustment.
Box 7.Quasi-Fiscal Activities in Developing Countries
In a large number of developing countries, economic policy objectives are frequently pursued through operations of central banks and other public sector financial institutions that may have important fiscal implications.1 Central bank lending to the public sector, including foreign exchange lending at below-market interest rates, and subsidized loans and loan guarantees extended by public sector financial institutions to specific sectors and groups of borrowers affect the public sector’s net financial position even though they may not show up in the central government budget as specific expenditure or revenue items.
Central banks lend to governments by providing overdraft facilities or directly purchasing government securities. When interest rates on loans to central governments and other public sector bodies are set at below-market rates, public expenditures will not be evaluated at their true opportunity costs, and this can lead both to excessive expenditure and to the selection of projects for which the true economic costs exceed their benefits. Apart from direct lending to governments, central banks frequently compel commercial banks to hold short-term government securities at below-market rates, by imposing statutory liquidity requirements. Central banks may also pay below-market rates on commercial banks’ reserve assets. These devices essentially represent taxes on the banking system, which tend to restrict the development of financial intermediation, increase the spread between borrowing and lending rates, and reduce saving and investment in the economy. Such forms of financial repression can also yield substantial revenue to the government. Between 1984 and 1987, for example, the Mexican government extracted close to 6 percent of GDP, or about 40 percent of total tax revenue, from controls on financial markets.2
In many developing countries, subsidized lending by development banks and other public sector financial institutions to specific groups, often without adequate collateral, is the most common form of quasi-fiscal operations. In India, for example, the central bank requires that government-owned commercial banks lend about 40 percent of their assets to small-scale farmers and other small businesses. To finance such lending, nonfinancial public sector undertakings are required to maintain deposits at below-market interest rates with the commercial banks and other public sector financial institutions. Extensive problems of loan recovery have plagued public sector banks that engage in such implicit subsidization since it promotes adverse selection. In some cases, the central bank has to step in by either providing funds to recapitalize bankrupt banks or assuming nonperforming assets. When faced with serious financial difficulties, state banks in Brazil have frequently raised overdrafts on the collateral of their legal reserves and taken large rediscount loans from the central bank. In early 1991, to improve the central bank’s portfolio, the government swapped such loans with its own securities to the extent of about 2 percent of GDP.
Despite a strong trend toward exchange market unification in developing countries, central banks in a number of countries still engage in multiple exchange rate practices, with different exchange rates applying to various categories of activities. A frequently used variation of this practice is to set the official exchange rate (cost of foreign exchange in terms of domestic currency) at a lower level than the market rate. All official transactions, such as debt-service payments on external debt owed by the government and public enterprises, along with payments for imports of state-owned enterprises, such as government- owned oil and electricity companies, are conducted at the official rate. The government may also require the surrender or repatriation of export proceeds of some or all commodities—in many countries, in fact, all exports are subject to surrender and repatriation requirements—and allow private sector imports of necessities (e.g., medicine) and capital goods at the official rate. Apart from distorting the allocation of resources and understating expenditure on public sector imports and debt-service payments in domestic currency terms, this also obscures the effective levels of taxation on exports and imports. During 1979–82, the Costa Rican central bank provided foreign exchange for certain imports at a lower rate than the rate it had paid for the foreign currency. In 1981 alone, these subsidies amounted to about 4½ percent of GDP. In 1987–88, the parallel market exchange rate of the Ugandan shilling ranged between U Sh 200 and 400 per U.S. dollar, although the Coffee Marketing Board was surrendering its export receipts to the central bank at the official exchange rate of U Sh 60 per U.S. dollar.
Governments also, through development banks and other public sector financial institutions, often provide exchange rate guarantees or subsidize exchange risk insurance. As a result of these subsidies, a strong incentive is created for the beneficiaries to increase their foreign currency liabilities, especially in countries where inflation is higher than in the main trading partner countries. While the impact of these subsidies on the fiscal deficit may not be immediate, the increase in the contingent liabilities of public sector financial institutions and ultimately those of the government may result in substantial fiscal expenditures during times of macroeconomic instability. In Chile, following the devaluation of 1982 in the aftermath of the debt crisis, the central bank provided subsidized foreign exchange to the private sector for external debt-service obligations, equivalent to almost 2 percent of GDP a year in 1983–85. At the same time, exchange rate guarantees amounted to about ½ of 1 percent of GDP a year.
In recent years, central banks in many countries that have experienced large capital inflows have opted to increase reserves and sterilize them rather than let the domestic currency appreciate freely. As a consequence, the domestic interest in such cases has often risen to levels higher than the return on foreign reserves, giving rise to losses for the central bank. In the early 1990s, quasi-fiscal costs of such sterilization operations were as high as ½ of 1 percent of GDP a year for some Latin American countries.3
Quasi-fiscal operations are typically undertaken to circumvent legislative and political constraints on fiscal policy, but they frequently have important fiscal consequences and, in general, reduce efficiency by distorting relative prices in the economy, thus impeding development. While it may be difficult to quickly eliminate many of the subsidies implicitly provided through quasi-fiscal operations, acknowledging such activities in central government budget statements would enhance the transparency of fiscal policy. Over the longer term, however, reliance on such subsidies, whether provided implicitly or explicitly, should be phased out in combination with the implementation of other structural reforms that enhance resource allocation and long-term growth.1For a more detailed discussion on the use of different quasifiscal instruments, see George A. Mackenzie and Peter Stella, Quasi-Fiscal Operations of Public Financial Institutions, IMF Occasional Paper (forthcoming).2For further details see Alberto Giovannini and Martha De Melo, “Government Revenue from Financial Repression,” American Economic Review, Vol. 83 (September 1993), pp. 953–63.3See Miguel A. Kiguel and Leonardo Leiderman, “On the Consequences of Sterilized Intervention in Latin America: The Case of Colombia and Chile” (World Bank: Washington 1993).
When faced with the need to reduce government expenditures, policymakers in many developing countries have frequently chosen the politically easier paths of cutting capital spending or allowing real wages in the public sector to be eroded by keeping nominal wage levels unchanged while prices are rising. Rarely have governments reduced public sector employment. 56 In a number of African countries, the growth of value added by government, measured primarily by increases in public sector employment, has exceeded the growth of private sector activity for long periods and sometimes by significant margins.57 In Kenya and Zimbabwe, growth in the private sector during most of the 1980s and early 1990s was almost 2 percentage points lower than the growth of the government sector.
Real wage reductions in the public sector, however, like capital spending cuts, can be counterproductive. They may reduce public sector efficiency, especially in key areas such as tax administration, the enforcement of tax regulations, law and order more generally, and the operation of state-owned utilities. To the extent that declining real wages of civil servants lead to a widening of differentials with respect to private sector pay, this expedient may also foster corruption and rent-seeking activities.58 In Tanzania, for example, real wages of civil servants had declined by 1980 to one fifth of their 1970 level, and this was widely regarded as a major factor contributing to absenteeism and corruption in the public sector. In contrast, among many of the relatively successful developing countries, such as Chile, Korea, and Singapore, public sector wages have been maintained at levels that are broadly comparable with the private sector.
Although reductions in public sector employment may be politically difficult, especially in countries where alternative employment opportunities appear limited in the short term, a number of countries have managed to implement extensive civil service retrenchment programs by providing severance payments and job-search assistance to departing workers. Between 1987 and 1991, Ghana reduced its civil service workforce—including through the elimination of “ghost workers”—by about 10 percent, offering severance payments of two months’ salary for every year of service and additional assistance, including employment counseling and food-for-work programs, for those who failed to secure alternative employment. In some cases, in the short term, net budgetary savings from such civil service retrenchment programs may be small but over the medium term the efficiency gains and savings can be substantial.
In most developing countries, reducing employment in state-owned enterprises is a prerequisite for their restructuring and privatization. A large number of such enterprises engage in activities that the private sector could undertake more efficiently. In many African and Middle Eastern countries, such as Senegal and Egypt, state-owned enterprises are engaged in food production, while in India, state-owned enterprises are extensively involved not only in heavy industry and mining but also in the production of consumer goods. For the developing countries as a group, nonfinancial state-owned enterprises account for over 10 percent of GDP.59
There has been relatively little progress in most developing countries in introducing more competition and easing entry restrictions on potential private sector competitors to state-owned enterprises. In spite of some large privatizations, the number of enterprises run by the state has not declined markedly. In some countries, policymakers have sought to improve the efficiency of state-owned enterprises by instituting performance contracts for managers. In most cases, however, such measures have led to little or no improvement in efficiency and profitability, in part because the enterprises, being state owned, continue to face conflicting objectives. In Senegal, for example, performance contracts for managers of the stateowned electricity company detailed more than 20 criteria on which performance would be judged, but the government lacked the power to enforce penalties if the criteria were not met. In fact, profitability declined because the government did not force other stateowned enterprises to pay for their electricity consumption on a timely basis. By contrast, in countries where state-owned utility companies have been sold to the private sector, as with the privatization of telecommunications in Argentina, Chile, Malaysia, and Mexico, productivity has increased, in part because of higher investment.
Among many commodity-exporting countries, the motivation for state intervention in commodity-producing sectors has frequently been macroeconomic stabilization. In the face of volatile export revenues, governments are concerned about the impact of foreign exchange flows on the real exchange rate and the international competitiveness of the traded goods sector. Intervention has taken a variety of forms, including export taxes imposed when commodity prices are high, requirements on exporters to hold foreign exchange earnings at the central bank, and investing proceeds from commodity price booms to promote diversification of exports. Experience suggests, however, that such intervention has tended to introduce further distortions and instability into the economy. Moreover, public sector management of windfall gains from commodity price booms frequently entails dealing with competing claims, such as from government ministries, special interest groups, and state-owned enterprises, which are not always resolved in such a way that the increase in revenues is invested in programs that enhance longer-term growth performance.
Since the 1980s, fiscal consolidation programs in many developing countries have also entailed improvements in the structure of taxation and in tax administration systems. A number of countries have rationalized and consolidated tax structures to reduce distortionary taxes, especially taxes on international trade, and have moved toward broad-based consumption taxes. In India, for example, the maximum tariff on imports was reduced from 400 percent in 1990 to 50 percent in 1995, and average tariff rates fell from well over 80 percent to under 30 percent over the same period. In Indonesia, prior to the introduction of the value-added tax (VAT) in 1985, revenue from indirect taxes amounted to about 1 percent of GDP, whereas by the late 1980s such revenue was equivalent to over 3 percent of GDP. Reforms of tax structures have also included marked reductions in personal and corporate income tax rates. In Thailand, the maximum personal income tax rate was reduced from 65 percent in 1980 to under 40 percent in 1993. While there is widespread acceptance that such reforms help improve allocative efficiency, and can therefore lead to higher revenue over the longer term, their success depends on limiting ad hoc exemptions that tend to narrow the tax base and reduce efficiency. Lower personal income tax rates can also help to promote a more equitable tax system in countries where, because of poor tax administration, high marginal tax rates are applied in an ad hoc and discriminatory fashion. Improving tax enforcement and compliance can allow countries to lower tax rates and at the same time increase tax revenue. In Chile, the difference between estimated potential revenue from indirect taxes and actual revenue declined from almost 25 percent in 1981 to 17 percent in 1993 as a result of more efficient tax administration.60
Concerns about inadequate tax revenues have also slowed the pace of trade and financial liberalization in some developing countries. Countries that have traditionally traditionally been dependent on trade taxes for a significant portion of revenues have dismantled trade barriers only gradually, despite the potential long-term benefits of liberal trade regimes. Although trade liberalization may lead to shortfalls in revenue in the short term—and in these cases expenditure reductions may be difficult to avoid—some trade liberalization measures can be implemented without significant declines in revenue. In fact, lifting quantitative restrictions may well increase revenue as the imports liberalized would most likely be subject to tariffs. Moreover, even in the case of tariff reductions, the fall in revenue may be small. In countries that have liberalized exchange markets, especially where official and unofficial exchange rates have been unified, the higher domestic currency price of imports at the liberalized market exchange rate may well allow tariff rates to be reduced with no loss of revenue.61
Developing countries that are at a relatively advanced stage in liberalizing trade regimes and implementing market-oriented structural reforms have attracted large capital flows in recent years. In some cases, however, the capital inflows may be attracted primarily by high domestic interest rates stemming from an unbalanced mix of monetary and fiscal policy. For these countries, fiscal tightening will help to safeguard financial stability by improving the domestic saving-investment balance. For all such recipient countries, however, large-scale capital inflows have been accompanied by higher domestic spending and higher imports, which have increased fiscal revenues. Such increases in revenue may well be unsustainable, and as suggested by Mexico’s experience, policymakers may need to aim for structural fiscal surpluses to provide a buffer that can be used in the event of large outflows. The experience of Mexico also underscores the role of fiscal policies in determining the credibility, and ultimately the sustainability, of pegged exchange rates. Devaluations that are not accompanied by strong fiscal measures are unlikely to gain credibility or deliver effective adjustment.62
While fiscal policies can bring substantial longerterm growth benefits, fiscal consolidation measures such as reductions in public sector employment, tax increases, and the privatization of state-owned enterprises can have adverse effects on the economically weaker social groups in the short term. The removal of generalized subsidies on basic necessities, the relaxation of price controls, and currency devaluations can cause real incomes of domestic consumers to decline in the short term. The adverse effects of adjustment and reform on the poor and vulnerable need to be addressed through well-targeted and cost-effective social safety nets. In a number of countries, such as Mozambique and Zambia, cash compensation schemes have shielded vulnerable groups during periods of rising prices and at the same time permitted a strengthening of budgetary positions. In Jordan, a generalized subsidy for selected food items was replaced in 1990 with a food coupon scheme that allowed coupon recipients to purchase fixed quantities of basic necessities at below market prices; the budgetary costs of food subsidies declined from over 3 percent of GDP in 1990 to about 1 percent in 1994. These examples illustrate the fact that, despite practical difficulties, especially in poorer countries, of targeting the most vulnerable groups—difficulties that often arise from weak administrative capacity and in some cases from inadequate political support—social safety nets can help alleviate many of the short-term adverse consequences of fiscal consolidation.
In comparison with many industrial countries, where the unfunded liabilities of publicly provided pension schemes currently imply considerable increases in future tax burdens, developing countries are at a relatively early stage of establishing extensive pension programs for the elderly. In some rapidly growing developing countries, however—especially many Asian countries with aging populations—the need to offer sufficient security to retired persons is likely to place growing demands on public sector expenditures. The experiences of public sector provision of pensions in a number of countries suggest that developing countries in the process of implementing pension programs should encourage greater private sector participation to avoid excessive fiscal burdens, which frequently arise from poor public sector administrative capabilities. During the 1980s, the rate of return on publicly managed funds in Egypt, Peru, Venezuela, and Zambia ranged between -12 percent and -37 percent a year; in Zambia, more than half the contributions were used for administrative expenses. The establishment of sound privately managed systems, however, will often require a strengthening of the financial system and of the official regulatory and supervisory capability. Pension reforms in many Latin American countries, such as Chile in the early 1980s, and more recently Argentina, have been designed to overcome the problem of rising contributions required to finance overextended benefit commitments. In addition to the fiscal burden, excessive contributions—higher than could possibly be realized in future benefits—will tend to discourage formal sector employment and to reduce saving and capital accumulation.
* * *
Many developing countries have made substantial progress with economic policies since the 1980s and are now much better placed to reap the benefits of increased integration of goods and financial markets in the global economy. The resilient growth performance of developing countries in the 1990s is indicative of the progress that has been made. Although domestic policies have improved in a number of areas, improved fiscal policies have been a prime source of greater macroeconomic stability and increased confidence for investment. Compared with the 1980s, fiscal imbalances are markedly smaller in a large number of countries, despite significant differences in the underlying causes of fiscal deficits. For some countries, deficits have stemmed largely from inappropriate policy responses to terms of trade shocks, but in most cases, fiscal deficits reflect a legacy of excessive and unnecessary state involvement in economic activities, which in turn has impeded growth of both the economy and tax revenue bases. Among the more successful performers, rapid and sustained economic growth has increased the need for public sector provision of infrastructure to keep pace with private sector development. Despite differences in the underlying causes of deficits, fiscal consolidation measures have consisted primarily of expenditure cuts, reflecting a widespread consensus among policymakers on the adverse effects on growth of higher taxation. In some developing countries, rapid growth has also enabled governments to reduce taxation without reducing expenditures on essential public services.
While sound fiscal policies have been established in a number of countries, in many others it is necessary to speed up the process of fiscal consolidation and ensure that programs already initiated stay on track. And in parallel with disciplined fiscal policies, radical structural reforms are still needed in many countries to elicit the private sector supply response and raise growth performance. With more open trade regimes and increased integration of capital markets, the costs of macroeconomic imbalances are likely to be even greater than in the past. Indeed, as some countries have already seen, adjustment forced by financial markets can require sharp changes in expenditure policies and lead to large output losses. All countries need constantly to evaluate expenditure priorities and to limit the extent of public sector intervention and the associated distortions that may hinder private sector activity. For many of the strong performers, it will be imperative to ensure that public sector commitments to extend social security programs, including pension schemes, do not give rise to excessive fiscal burdens. These countries have the opportunity to avoid the costly mistakes of many industrial countries where pension entitlements have contributed to increasingly unsustainable fiscal burdens.