Cheryl W. Gray and Johannes F. Linn
The public sector has grown rapidly in almost all developing countries during the past 40 years. Most governments—in Asia and Africa, particularly—responded to enormous social and infrastructural needs and weak private sectors after independence by assuming a major role in promoting investment and growth throughout the economy. Central government spending alone increased from less than 10 percent of GDP on average after World War II to almost 25 percent in 1985. State-owned enterprises have also expanded dramatically, and state and local governments have assumed major importance in some countries.
Many developing countries have faced fiscal crises during the past decade. Regardless of their economic characteristics—whether low-income, middle-income, oil exporter, oil importer, planned or market economy—many countries saw their public-sector deficits rise to unsustainable levels in the years before 1982. These deficits led them into heavy foreign borrowing and high inflation. As external financing dried up in the 1980s, fiscal retrenchment became unavoidable. The reduction in deficits since the early 1980s has been impressive (Chart 1), but many countries need to do more. The challenge is to contain deficits without falling further into recession or harming longer-term development prospects. The answer usually involves both raising revenue and cutting spending, while trying to minimize the economic costs of both.
While fiscal deficits have remained high on the agenda of policy reform, the urgency of this problem has distracted attention from the broader role of public finance in development. Public finance affects aggregate resource use and, together with monetary and exchange rate policies, influences the balance of payments, the accumulation of external debt, and the inflation, interest, and exchange rates. Public spending, taxes, user charges, and borrowing also affect the behavior of producers and consumers, and influence the distribution of wealth and income in any economy. The question is how to maintain the quality of government over the long term so that the methods of raising public revenues and the allocation of public spending promote—rather than hamper—growth and poverty alleviation.
This article, based on the World Development Report 1988 (see box on publication), examines the fiscal performance of developing countries in the past decade and suggests five broad ways in which developing countries could reform public finance:
- adopt prudent budget policies;
- reduce the costs of raising public revenues;
- increase the efficiency and effectiveness of public spending;
- strengthen the autonomy and accountability of decentralized public entities; and
- ensure consistency between public finance policies and poverty alleviation goals.
Although implementation of such fiscal reforms often faces daunting practical and political obstacles that should not be underestimated, recent reforms in a number of developing countries indicate that these obstacles can often be overcome.
Prudent budget policies
Moderate and sustainable fiscal deficits, with some allowance for cyclical variation, are far preferable to successive phases of rapid fiscal expansion followed by sharp fiscal contraction. Debt accumulation, capital flight, and loss of confidence during the expansion phase make adjustment during the ensuing contraction all the more difficult. The poorest members of society bear the greatest burden because they are unable to shield their incomes by moving assets abroad, and they are often the first to lose jobs and income opportunities during a recession.
A country can finance its public deficit in one of three major ways—by borrowing from abroad, borrowing domestically, or printing money. (In the short run, drawing down foreign exchange reserves or building up arrears to creditors are additional sources of finance.) Each of these approaches has its negative effects. Heavy foreign borrowing can lead to appreciating real exchange rates, widening current account deficits, capital flight, and growing foreign debt-service burdens. Heavy domestic borrowing by the public sector can put pressure on domestic interest rates and crowd out private investment. Money creation can exacerbate inflation if the money supply grows faster than the demand for monetary assets. “Moderate and sustainable” public deficits, therefore, are deficits at levels consistent with an acceptable external debt-service burden, reasonable domestic real interest rates, and low inflation.
Most of today’s heavily indebted countries increased their fiscal deficits in the late 1970s and early 1980s. For example, the deficits of both Argentina and Mexico grew from less than 7 percent of GDP in 1978 to over 15 percent in 1982. Countries in this group financed their deficits by foreign borrowing until 1982, and their current account imbalances widened in step with fiscal expansion. When the capital inflow abruptly halted in 1982, financing shifted to domestic sources. Inflation accelerated as money creation increased, and public domestic borrowing helped drive real interest rates to over 20 percent in some countries. Recession followed cuts in investment and non-interest current spending (e.g., wages, other goods and services, and subsidies); these cuts had to compensate for the rising interest burden on foreign and (often more expensive) domestic debt. Capital flight worsened the dilemma. People who had transferred savings abroad before 1982 in response to unsustainable deficits and overvalued exchange rates hesitated to repatriate them given the uncertain environment. The reduced pool of domestic savings meant that budget cuts had to be even deeper.
The World Development Report 1988 is a staff report of the World Bank. It was prepared by a team led by Johannes F. Linn and comprising William R. Easterly, Cheryl W. Gray, Emmanuel Y. Jimenez, Govindan Nair, Anthony J. Pellechio, Zmarak M. Shalizi, and Eugene L. Versluysen. The Bank’s International Economics Department prepared the main projections and statistical material.
The work was carried out under the general direction of W. David Hopper and Stanley Fischer.
Other countries (such as the Republic of Korea and Thailand) followed more prudent fiscal policies during the past decade. They avoided unmanageable debt accumulation or reliance on money creation by reducing deficits quickly in the early 1980s, generating good returns on public investment, and maintaining export-oriented trade and exchange-rate policies. As a result, these countries—which might easily have joined the problem debtors—steered clear of debt troubles even though they did borrow heavily.
Countries that depend on commodity exports faced cyclical swings in commodity prices and these had a large impact on their external and fiscal accounts. Some countries, such as Cote d’lvoire, Kenya, Mexico, and Nigeria, used their growing revenues from commodity booms in the late 1970s to finance increased expenditures, often of dubious economic merit. The boom also led to appreciating real exchange rates and growing indebtedness, as the countries borrowed against future earnings to finance their sharply increased spending. But they failed to reduce spending quickly when commodity prices fell. The result was a sharp increase in fiscal deficits and inflation. As other exports stagnated, the accumulated debt imposed higher debt-service burdens.
In contrast, some countries, such as Botswana, Cameroon, and Indonesia, followed a more cautious strategy. They saved part of their boom revenues, in some cases by paying off external debt, and allowed spending to increase only moderately. (In Botswana it actually declined as a share of GDP.) Inflation and exchange rate appreciation were modest, and performance of other exports remained satisfactory. Any fiscal adjustment needed during the subsequent commodity downturn was undertaken rapidly. Recession was avoided in all three cases. The lesson for commodity exporters is that they would do well to err on the side of caution, assuming a boom to be temporary rather than permanent (see “External Shocks and Fiscal Policy in LDCs” by Ke-Young Chu in the June 1988 issue).
Chart 1.Public sector deficits 1979-85, unweighted averages for 23 developing countries
Persent of GDP
Source: Wald Development Report. 1938. World Bank.
For countries that now face unsustainable public deficits, macroeconomic stabilization is a top priority. Yet stabilization efforts must be coordinated with structural measures to reduce distortions and promote growth or these efforts may prove unsustainable in the longer run. Stabilization programs that rely on higher tariffs, restricted imports, and reduced investment can stifle structural reform and growth. Conversely, adjustment programs that attempt to reduce these or other distortions can be destabilizing unless accompanied by complementary fiscal reform.
Raising public revenue
Raising public revenue is expensive. The costs include not just the direct costs of administration, but also the indirect costs (such as reduced savings or investment) that arise from distortions in economic activity. Some methods, however, are far costlier than others. Certain principles should guide governments in reforming their revenue systems.
To the extent possible, the costs and benefits of public services should be linked through user charges. These might be charges related to consumption (public pricing) when specific consumers can be identified, or charges related to benefits (such as “valorization” schemes to recoup the cost of public infrastructure investments from beneficiaries in Latin America) when the benefits of a public service are concentrated in a particular area. User charges can promote greater economic efficiency by making beneficiaries weigh the economic cost of a good or service against its benefit. Users who have to pay for services are also more likely to monitor carefully the performance of suppliers.
There is significant scope for raising additional public revenues from higher user charges in developing countries. Many public services for which charges are both feasible and appropriate—including higher education, hospital care, electricity, water supply, and urban transport—are often provided free or at highly subsidized prices to all users. Uniformly low prices imply that high-cost services are much more subsidized than low-cost ones. Because the poor tend to have little access to high-cost services, the subsidies disproportionately benefit the relatively rich. Our research indicates that a disproportionately large share of higher education subsidies is received by upper income groups in selected developing countries of Asia and Latin America. A similar pattern holds for health care, in part because most health facilities are in urban areas although the poorest tend to live in rural areas. The average health sector subsidy for urban households in China, Colombia, Indonesia, and Malaysia, for example, is up to five times that for rural households. Charging more for these services—while maintaining subsidies only for the poorest groups—could increase efficiency while raising revenues to fund the expansion in basic services that is so urgently needed.
Even though user charges can often be increased, general taxes will continue to be necessary to finance public goods that benefit the citizenry at large. Tax revenue as a share of GNP has increased over the past decade in many developing countries in response to the need for fiscal adjustment. Taxes on international trade are still the major source of central government revenue in low-income countries, particularly in Sub-Saharan Africa. But the revenue share of trade taxes has been declining in favor of domestic taxes. Among the domestic taxes, commodity taxes such as sales, excise, and value-added taxes are more important than income taxes. This is in contrast to industrial countries, where income taxes are often the most important source of revenue.
Although some tradeoffs among objectives remain unavoidable, tax systems can be reformed to raise revenues while reducing economic distortions and the burden on the poor. Simplicity in tax design is essential, because most developing countries have severely limited administrative resources. Thus, administrative feasibility should always be a basic concern, and strengthening tax administration is an important aspect of any tax reform.
Tax systems satisfying these criteria will typically include more streamlined company and personal income taxes, a value-added tax (often at the manufacturer’s level only, for simplicity), and a few excise taxes on luxury items or socially undesirable goods such as liquor and cigarettes. Where import duties are judged to be necessary, they should be low and more uniform. Marginal tax rates should be low to minimize distortions and promote compliance. Taxes should be simply structured, with few rates and few exemptions. Extensive use of tax incentives to achieve particular social goals rarely works in practice, because they tend to create or exacerbate economic distortions, and severely complicates tax administration. High threshold exemptions in the individual income tax and exemption of unprocessed products from the value-added tax will go far toward keeping the tax burden on the poor low while concentrating administrative resources where they are most productive in raising revenue.
Recent tax reforms in developing countries such as Colombia, Indonesia, Jamaica, and Malawi have concentrated on expanding the tax base, thus avoiding higher tax rates and adverse effects on incentives. To make the tax structure more transparent and to ease administration and enforcement, they have favored fewer rates and fewer exemptions. In Jamaica, the reformed personal income tax has only one rate—35 percent—and a high personal exemption level that exempts most households from the tax. Indonesia adopted a VAT with only one rate (10 percent), and it abolished all special incentives in its company tax and instituted a simplified rate structure with a maximum rate of 35 percent. All of these reforms have tried to promote equity by improving the collection of taxes from the wealthy and by avoiding taxes on the poor rather than by fine-tuning the rate structure; taxes with steeply progressive rates are hard to collect in developing countries.
Chart 2.Central government spending as a •hare of GDP, by region 1975-85
Persent of GDP
Source War id Devotopmeni Report. 1988 World Bank.
Note: Figures rapntaanr GDP mightsd group awagat for lack of data, China, Japan, Ngaria. and a number of relativly small countries are excluded from the sample in this chart
Central government spending grew substantially until 1982 in most developing countries but then tended to decline as a share of GDP until it grew again in 1985 (see Chart 2). Interest payments have been the fastest growing component of government spending in recent years, while capital spending appears to have been the category most prone to cutbacks during recent austerity periods.
Tight budget constraints and the high cost of raising revenue means that it is vital to set priorities and achieve quality in public spending. Priorities can be set by considering what governments do best and what markets do best. Governments must provide “public goods” that benefit all citizens, such as law and order and national defense. They should be involved in the provision of goods and services with large benefits to society, such as primary education and basic health care. Direct investment or regulation is needed to control monopolies that result from either a single source of supply or large returns to scale relative to the size of the market—water supply, sanitation, and power, for instance. Finally, government subsidies on goods and services consumed by the poor are sometimes justified, but to contain the cost they should be accurately targeted to those most in need.
Public spending in the developing world should be more sharply attuned to these priorities. In education, countries (particularly the poorest) need to expand and improve the primary sector. In health, more public resources should be allocated to basic health measures such as immunization and prenatal care. Public spending on these basic services is not only socially more profitable than spending on higher education, non-essential drugs, and expensive curative hospital care, but it is also more equitable, because the relatively wealthy are the primary users of the more expensive services. Although there is still a need for public sector involvement in the provision of higher education and hospital care, more of the costs of these services should be recouped through user charges. In urban services, public provision of roads, water, electricity, and sanitation is critical, whereas bus services and housing infrastructure can often be more efficiently provided by the private sector. In rural infrastructure, roads, potable water, irrigation, and electricity are areas in which the public sector has been and should continue to be involved; but in each case spending can often be shifted toward more cost-effective techniques. Such reforms can expand the access of the poor to basic services while increasing the contribution of the public sector to economic growth and development.
Setting priorities is only the first step. All dimensions of investment projects—economic, technical, administrative, financial—must be appropriately designed, and implemented in a policy environment that provides incentives conducive to good performance. Priorities and quality must also be considered in allocating recurrent public spending. Adequate spending on operation and maintenance will often be more important than new investments. Hiring fewer civil servants and paying them competitive wages will generally be preferable to using government as the employer of last resort. As mentioned earlier, subsidies are more efficient when targeted to the poor rather than dispersed across the entire populace.
Improving the efficiency and effectiveness of public spending requires reform of fiscal planning, budgeting, implementation, and monitoring. Fiscal planning ideally involves the formulation of a phased investment program, projections of current spending needs, and an assessment of revenue availability and borrowing requirements over three to five years, all set in the context of a consistent macroeconomic framework. The annual budget would then be a one-year slice of the medium-term plan. For plans and budgets to promote effective decision making by individual public agents, the trade-offs among agencies, programs, and projects must be explicit, and the budget constraint for each agency, once set, must be “hard”—that is, an agency must not be permitted to exceed a budget on its own initiative.
Although the capacity to carry out medium-term fiscal planning and comprehensive annual budgeting is limited in most developing countries, some have coped well. Botswana, for example, has developed procedures to ensure that careful attention is paid to the recurrent cost implications of its investment spending. Chile and Thailand have been thorough in using economic analysis (primarily cost-benefit analysis) to screen potential investments. The Gambia, Central African Republic, Costa Rica, and a number of other countries are controlling and trimming government payrolls through hiring freezes, civil service censuses, and voluntary retirement schemes; Ghana and Sri Lanka are attempting to rationalize the civil service wage structure. Mexico is moving toward targeted food subsidies. These and other examples show that progress in improving the efficiency and effectiveness of public spending is possible.
Decentralized public entities
Decentralizing both decisionmaking and accountability can help to link costs and benefits, and thus improve efficiency. Local decisionmakers can be more flexible and responsive to the needs and preferences of their constituents; equally, citizens can better watch over local entities than over central ones. The extent of feasible decentralization varies from sector to sector. Social service providers, such as schools and clinics, can usually be given greater responsibility for delivery and for cost recovery, although some degree of central control is needed to maintain standards and to provide targeted subsidies where necessary. State-owned enterprises (SOEs) should be granted the autonomy to cover their costs through rational pricing, with managers held accountable for the quality of services and for the financial viability of their enterprises. Means of decentralization in urban and rural infrastructure are discussed in the accompanying article by Emmanuel Jimenez.
Local entities and SOEs will require financing other than user charges, however, if they supply public goods or subsidies. Local governments can streamline their tax systems and—through more accurate property valuation and better administration—broaden the base and improve the collection of property taxes. Borrowing by local governments or SOEs may be warranted to fund investment whose benefits emerge in later years, but major borrowing throughout the public sector should be subject to central approval, especially where government guarantees are given. In a sample of 99 developing countries, direct foreign borrowing by SOEs between 1974 and 1981 accounted for more than one fifth of the total foreign debt of those countries as a group; in fact, SOEs have accounted for an even larger share of total debt (over one half in Brazil, the Philippines, Portugal, and Zambia), because governments have passed on much of their own borrowing to them. Potentially costly government guarantees, whether implicit or explicit, have generally stood behind all SOE borrowing. When the state agricultural marketing board in Senegal was liquidated in 1980, for example, the government assumed bank debt equivalent to 15 percent of GDP.
Central government grants and subsidies are often justified, usually on equity grounds, but they should be designed to provide incentives for efficient cost recovery or local revenue collection. Unconditional grants can discourage local revenue initiatives, as has occurred in Nigeria and Brazil. Regular and reliable auditing by central authorities can increase local accountability for the use of borrowed funds, grants, or subsidies.
Financial flows within the public sector are usually complex and often confusing. Transparency in these transactions can be increased by accounting explicitly for all subsidies, equity investments, dividends, taxes, payments for goods and services, borrowing, or debt guarantees. Financial obligations between agencies should be enforced. Arrears in one account jeopardize the financial stability of the creditor and often lead to compensating arrears in other accounts. This causes confusion and lack of control.
Strengthening local government—like most of the other reforms suggested above—calls for improved administration and management. Efforts to improve the quality of public sector employees through recruitment, training, and performance incentives will always be of central importance in fiscal reform. For many commercial SOEs, greater competition from private providers or increased private sector involvement can also help to reduce inefficiency as well as the budgetary burden. Because the barriers to full and rapid privatization are often daunting, intermediate solutions—such as subcontracting, leasing, or permitting private competitors to enter the market—are often more feasible.
Public finance and poverty goals
Reducing poverty remains a critical challenge of development policy. Public finance can be a powerful tool for this purpose if policies are designed in the manner described above. Fiscal prudence sets the groundwork for growth—the precondition for defeating poverty in the long run. Moreover, the poor often bear the biggest direct burden of imprudent policies.
On the revenue side, the poor can be exempted from income taxes by setting a high personal exemption level, as has been the case in Jamaica and Indonesia. Exemption of unprocessed products (particularly food) from sales taxes softens the effect of such taxes on the poor. Targeted subsidies can be used to exempt the poor from user charges. In some cases modest user charges on higher cost services may actually help the poor by permitting increased investment in, and thus expanded access to, publicly provided essential services, such as potable water, primary education, and basic health care. The primary education budget in some African countries—for example, Cote d’lvoire, Mali, Senegal, Tanzania, and Togo—could be increased by more than 20 percent if stipends for living expenses paid to higher education students were terminated.
The most important fiscal instrument for directly relieving poverty is public spending. Expanding the resources devoted to low-cost urban and rural infrastructure can lead to immediately improved living standards, even if they are financed through user charges. Strengthening local governments is crucial because they are usually the main providers of urban services. Subsidies targeted for the poor can be maintained even in times of fiscal austerity. In the long run, investment in human resources—including basic health care and nutrition, primary education, and family planning—can significantly improve the lot of the poor and support growth in developing countries.