Peter S. Heller
Developing countries encounter balance of payments difficulties for many reasons. Sometimes the difficulties are external in origin—for example, a sudden increase in energy prices. But they are frequently caused by inadequacies in domestic economic policies, for example when rapid growth in public expenditure results in a fiscal deficit that is accommodated by expansionary monetary and credit policies. These policies in turn lead to domestic cost and price increases, making exports (and import substitutes) less competitive and weakening the balance of payments, especially if the exchange rate is inflexible. Measures to restore a better balance between the demand and supply of resources in the public sector thus form a crucial element in many policy packages adopted to resolve balance of payments difficulties.
In principle, a better balance between supply and demand in the public sector can be achieved by raising budgetary revenues, or by cutting budgetary expenditures, or by some combination of both. The Fund devotes considerable attention to assisting countries to make their tax systems more buoyant, to reduce the disincentive effects of taxation, and to administer their tax systems more effectively. In practice, however, it is sometimes very difficult or undesirable to raise revenues: if taxes are already fairly high, an increase in tax rates may seriously depress incentives, with undesirable effects on development; moreover, it is not always certain that revenues will increase in the short run. As a practical matter, therefore, policy measures must often be directed to checking or reducing government spending. For these reasons, government expenditure policies are often important elements in programs of external adjustment supported by the Fund.
Other aspects of the Fund’s work with member countries also call for discussion of government spending policies. Even where there is no Fund-supported adjustment program the Fund will, in the course of its regular surveillance activities, assess its member countries’ macroeconomic policies. Changes in expenditure policy may be seen as necessary when a deficit is perceived as too large and the country has difficulty mobilizing additional revenues. The attention of the authorities will be drawn to specific expenditure policies where these give rise to particular structural problems (such as unsustainable growth in a particular category of expenditure). In many instances, the Fund provides technical assistance to carry out detailed appraisals of countries’ expenditure policies or budgetary and expenditure control procedures.
Because government expenditure policies play an important role in the allocation of resources, the accumulation of capital, the distribution of income, the balance of payments, and the realization of other national objectives, changes in such policies have wide-ranging consequences and are generally controversial. To help clarify the issues that arise in deciding among alternative expenditure measures, this article examines the economic criteria that may be used for analyzing expenditure policy, and describes the types of expenditure measures commonly adopted in Fund-supported adjustment programs.
Framework for analysis
A comprehensive analysis of the internal and external imbalances of an economy must include an evaluation of the government’s overall expenditure policies. This evaluation requires answers to several key questions:
• Do the different elements of the government’s investment program realize an adequate social rate of return and do they efficiently further the government’s medium-term economic objectives?
• Which groups in society benefit from the different components of the government’s recurrent expenditure program and at what costs are these benefits provided?
• Is the balance between investment and recurrent expenditure appropriate, both in macroeconomic terms and at the sectoral level?
• Does the government, broadly defined, produce particular goods and services efficiently?
• What are the dynamics of growth of particular categories of expenditure?
• How sustainable are particular expenditure policies?
• What effect will a given expenditure (say a subsidy on a largely imported product) have on the balance of payments?
Ultimately, policy recommendations flow from an appraisal of the relative productivity of different expenditures, the effects of these expenditures on the allocation and distribution of resources, the relative costs of achieving different objectives, and the relative priorities of the authorities as between different objectives. Precise and comprehensive answers to these questions would require analytical and statistical resources beyond the means of policy analysts in most developing countries. An economist can nonetheless carry out an approximate or intuitive analysis of the implications of particular policies for resource allocation and distribution, ask whether they are consistent with the stated policy objectives, and point out the financial costs of those policies.
To analyze government expenditure, it is often helpful to concentrate on the main sources of growth in expenditure and on broad distortions in the expenditure structure. Here the World Bank and other aid agencies, with their sectoral knowledge and familiarity with country development programs, are valuable sources of information.
Judging among sectors. Available analyses of the productivity of programs in the public sector are likely to be of varying quality, particularly when the activities range from the production of marketable goods and services to the provision of social services and public goods (such as defense or public administration). Such analyses rarely provide an adequate empirical basis for judging the relative productivity of different sectors. It may be difficult to judge the appropriateness of the sectoral balance of expenditure, but abnormally large or growing shares of expenditure on, say, administration or defense may call for analysis and discussion.
Investment. Few countries systematically collect data for evaluating expenditure on public investment. Estimated rates of return are typically available for only a few projects, if at all. Those estimates that are available tend to be based on a limited analysis of a project’s profitability; they rarely consider the project’s place within the country’s macroeconomic development strategy or the limitations likely to be imposed by, say, national shortages of technical manpower, foreign exchange, or recurrent budgetary resources. It may thus be necessary to use other criteria for appraisal, such as the project’s likely contribution to export revenues or its potential to produce substitutes for imports (bearing in mind its domestic resource costs), the performance of completed projects in the same sector, or its potential for removing bottlenecks in infrastructure. The World Bank has developed the concept of the “core” investment program, consisting of the limited set of projects which can be completed quickly and with high return, with the available resources.
Recurrent expenditure. To evaluate recurrent expenditure for the provision of goods and services, the cost-effectiveness of different programs and the costs of achieving given objectives must be assessed. The beneficiaries of given programs must be identified, relying on whatever sectoral appraisal studies exist and on whatever proxy indicators can be derived from data on the output of budgets and programs. Recurrent expenditure is potentially as productive as investment expenditure; in some sectors (for example, highways), expenditures on maintenance have had much higher rates of return than expenditures on new projects. One approach to evaluating the productivity of recurrent expenditure is to examine how it is divided among items, both on an aggregate basis and within sectors. Such a classification can often be made, at least to a limited extent, from budget documents. Countries commonly seek to reduce recurrent expenditure unevenly, making severe cuts in purchases of goods and services and negligible cuts in the government workforce. Government staff are then unable to provide services in the field for lack of gasoline or spare parts for vehicles, while plant and equipment deteriorate for lack of maintenance. The consequences can be devastating for the productivity of government workers and for the utilization of physical infrastructure.
By examining the levels of inputs to a sector and the relationships between these inputs, over time, both in real and nominal terms, one can gauge the adequacy of these inputs. Comparisons may be made both in financial terms and in terms of the relationship between major physical inputs (such as the number of employees or vehicles) and output. Cross-country comparisons of the relative size of different categories of expenditure and of key input-output relationships (for example, expenditure per student in primary school) are also useful.
Certain categories of recurrent expenditure lend themselves to possible abuse. Often, an examination of the budget suggests large and questionable expenditure on items such as foreign embassies, overseas travel, and allowances. Travel by senior civil servants as a “backdoor” benefit proves costly to the government, both financially and in terms of lost productivity.
Wages, employment. In evaluating the size and structure of public sector employment, cross-country comparisons can be a useful supplement to the analysis of specific programs. Often there is a built-in impetus for growth in public employment (for example, where high school or university graduates are guaranteed public sector employment), raising obvious questions as to the productivity and affordability of public employees. Government wage levels are also a matter of concern. In many countries, government wages are significantly higher than is warranted by wages elsewhere in the economy. Conversely, where real wages are much lower in government than outside, the government’s productivity may be drastically impaired by difficulties in retaining staff or by pressures for corruption and for taking illicit outside jobs. Crosscountry indicators and other market data can be used to evaluate the level and structure of government salaries and wages.
Subsidies, transfers. Subsidies and transfers often constitute a significant share of expenditure. Governments may subsidize sales to consumers of basic commodities, such as staple foods, fertilizers, and petroleum products, holding their prices below market costs. Other budgetary subsidies finance the operating deficits of nonfinancial public enterprises, particularly those supplying essential services (for example, electricity, water, and urban transport). Some subsidies to public enterprises are the result of deliberate decisions—to hold the prices of their goods and services below market costs, or to allow high-cost production because this serves other policy objectives such as employment creation.
When subsidies are intimately related to pricing policy, several issues arise:
• How do the prices set by government affect consumption and production?
• Where low producer prices are the vehicle for maintaining low consumer prices for basic foodstuffs, is agriculture stagnant? Is less food being marketed officially; has smuggling increased or a “parallel” domestic market grown up?
• Is the budgetary cost of subsidized prices open-ended or capped; is it a growing portion of the budget?
• How do the subsidies on particular commodities affect income distribution; how efficiently do they reach their target groups?
• How do subsidies affect the balance of payments if the subsidized commodities are largely imported?
Subsidies and transfers not linked to pricing policy may be direct attempts to alter income distribution or to channel funds to particular groups within the society. In evaluating their distributional implications and costs, some subsidies and transfers are of particular interest, notably those with questionable results (such as university scholarships largely received by children of upper-income families, or subsidies on gasoline that mainly benefit middle- and upper-income groups); and those that seriously distort labor market decisions (for example, by reducing the perceived private cost of higher education).
Expenditure linkages over time. Particularly in countries with high rates of inflation, it is necessary to consider what the government’s expenditure structure implies for growth in spending over time. Governments may link certain categories of expenditure (for example, wage rates or the interest rate on outstanding government securities) directly to the rate of inflation. By maintaining such linkages, a government runs the risk of losing control over a sizable share of its budget, shifting the burden of restraint on to other elements of its expenditure. Since this may have disastrous consequences for the productivity of government programs, such linkage policies may call for a fundamental reappraisal.
In evaluating investment programs, too, it is critical to ensure that they are not sowing the seeds of future budgetary disorder. The investment costs of a project in the current budgetary period may be very small compared to those in subsequent periods. The relationship between investment and recurrent costs must be closely examined. Some investments, perhaps financed by external donors, may not be very costly to start with but may require significant recurrent expenditures from the domestic budget in the future.
Execution. In some countries, budgetary imbalances may be caused as much by problems in budgetary execution as by policy. Problems of execution may include expenditure for purposes not sanctioned in the budget, improper use of payment vouchers, high-cost purchasing and procurement, and payments to so-called “ghost workers.” Procurement policies may not be designed to ensure that purchases are from the least-cost sources of supply. If the budget is poorly monitored, excessive commitments may build up without being noticed, causing pressures for supplemental appropriations. Budgetary arrears are frequently used to allow higher expenditures than were programmed, at the expense of forced “credit” from the private sector.
For a number of reasons, Fund staff missions do not—and are not able to—undertake as comprehensive and detailed an analysis as that suggested by the framework discussed above. Many areas of expenditure (such as the government’s investment program or the policies within a particular sector) are beyond the Fund’s competence and expertise and are often the focus of considerable attention by the World Bank. In other areas, lack of time, staff resources, or reliable data often precludes a systematic and detailed evaluation. But within these limitations, the Fund staff attempt to assess government expenditure policies in the broader context of what is necessary to achieve a sustainable balance of payments adjustment in the short to medium term.
The relative importance of the different aspects of expenditure discussed above obviously varies among countries and over time. There is no single, orthodox approach to appraising the structure or level of public expenditure. Assessments of government expenditure may differ in the weight they attach to particular policy issues or to the criteria outlined above, in part because of differences in the relative weights assigned by the countries themselves. Fund staff missions routinely seek the views of the World Bank on the quality of investments and their relative priority. In preparing for a Structural Adjustment Loan, the Bank may provide a detailed analysis of government expenditure policies.
In their appraisal and discussion of expenditure policies with a country’s authorities, Fund staff are keenly aware of the importance of promoting economic growth. Governments generally seek to foster growth by channeling government expenditure to the development of infrastructure and human resources and to direct investment in particular activities or industries. However, sustainable government expenditure policies to achieve these goals can only be pursued within the limits of available resources—that is, domestic savings, external capital inflows, and a viable balance of payments position.
What of the distribution of income? It is often difficult to evaluate or measure the precise distributional consequences of particular policies. An adjustment program, designed to deal with a fundamental imbalance in the economy, may entail some losses in real income for certain sectors and groups in the short term. It needs to be kept in mind, however, that the situation that engendered the need for an adjustment program is itself likely to have depressed incomes, possibly with distributional consequences worse than those of the adjustment program. (See “The effects of adjustment,” by Wanda Tseng, Finance & Development, December 1984.) Experience has shown that, in the medium term, a rational effort to adjust and a movement toward relative prices that reflect scarcity values (as signaled by the market, in market economies) have better consequences for real incomes, and in particular the real incomes of the poor, than a failure to undertake such policies.
In general, the Fund staff, in collaboration with the authorities, seek to determine the overall magnitude of deficit reduction that would be necessary and consistent with the adjustment program as a whole. It is for the authorities of the country to determine which taxes are to be raised or which expenditures are to be reduced, reflecting their own socioeconomic and political perceptions regarding alternative measures to rectify a fiscal imbalance, and their own implicit evaluation of the costs of alternative choices. The formulation of an adjustment program may sharpen their recognition that some of their priorities are not tenable, in the face of overall macro-economic and fiscal constraints. In particular cases, however, especially when specific items of expenditure have been shown to be a major cause of the fiscal deficit or of a balance of payments disequilibrium, the staff may more actively recommend specific measures for reducing expenditure.
Once an adjustment program has been formulated, some of the measures, including expenditure measures, may be implemented before the program formally enters into force, in order to make the prospective macroeconomic adjustment more credible. More frequently, however, specific expenditure policies are implemented during the program period; the Fund reviews the extent of their implementation and may focus on specific problems that may arise. Difficulties in implementing particular measures are reviewed in the context of the overall policies pursued and the progress achieved in restoring external balance.
What policy actions have countries actually taken to rationalize government expenditure in the context of Fund-supported adjustment programs? Almost 90 percent of 78 programs surveyed (a sample of programs agreed upon between 1980 and 1983) contained specific or general recommendations for cutting recurrent expenditures. Approximately two thirds called for cuts in capital expenditures and net government lending. In one fifth of the programs surveyed, policies to improve or augment investments were introduced. Almost two thirds of the programs called for improvements in expenditure administration and policies to limit the growth of subsidies and wages.
Program commitments on expenditure vary in specificity. Often, a country will simply commit itself to limiting the growth of an aggregate category of expenditure, without detailed specification of the policies to achieve such reductions. Other programs focus on narrowly defined categories of expenditure.
Investment. Fund-supported programs have occasionally involved increased capital expenditure, with emphasis on removing key bottlenecks to economic growth. There has been a tendency to support investment programs funded by grants and loans on favorable terms from bilateral and multilateral donors. Some programs have also required reconsideration of the structure of the investment program, particularly when a cut in investment outlays is warranted. This arises when projects are perceived to be of questionable productivity, imply exorbitant recurrent costs, or are of low priority relative to other claims on the budget. Some programs have emphasized the rehabilitation and maintenance of existing public infrastructure at the expense of new investment, even if this means giving up some external aid tied to new investment.
Wages, employment. To limit increases in government wages, authorities have frozen or postponed adjustments to the wage scale, held nominal increases in wages below the rate of inflation, reduced fringe benefits (for example, housing entitlements), and, in rare cases, reduced nominal wages. By tapering wage increases, permitting low wages to rise more than high ones, governments have often attempted to protect their lower-wage employees somewhat from the consequences of necessary wage restraint; this has, however, often led to difficulties in retaining senior personnel whose wages were held down. Occasionally, to help restore government productivity, where wages have been seriously eroded, Fund-supported programs have provided for increases in the real wages of the civil service.
Governments have frozen and occasionally reduced public sector employment in the hope of eliminating redundant or underemployed staff. Efforts have also been made to limit net recruitment into the civil service and to remove “ghost workers” from the payroll. Countries with systems of guaranteed employment for secondary school and university graduates have occasionally withheld this guarantee.
Subsidies, transfers. Subsidies may significantly contribute to a fiscal imbalance, and the resulting price distortions may result in a significant misallocation of resources. Policies to reduce budgetary subsidies have thus been a common component of Fund-supported adjustment programs. Often, countries have committed themselves to increasing the prices of specific commodities or basic services within a given period, establishing mechanisms to ensure that changes in world market prices or in exchange rates are passed through promptly to producers or consumers. In some cases, countries have attempted to soften the effects on the incomes of those losing subsidies, by gradually phasing in price increases for particularly sensitive commodities or services. Occasionally, programs have included policies designed specifically to safeguard the poorest groups. Where operating subsidies have sustained particular inefficient public enterprises, governments have sought to improve the financial return on operations through cutting employment, modernizing their procedures for management, and improving those for financial control—often with World Bank technical assistance. At times, the investment policy of these enterprises has come under close scrutiny, particularly when their investment programs are large and of questionable productivity. Finally, in some Fund-supported programs governments have closed unprofitable public enterprises or sought private buyers for them.
Other government costs. Governments frequently indicate their intention to limit expenditure on the broad category of “other purchases” of goods and services. Less typically, authorities may specify the particular categories of expenditure they intend to curb. Cuts in expenditure on overseas travel and allowances as well as in the number of overseas embassies and foreign missions have featured in several adjustment programs. In many countries, non-wage outlays have been reduced below the minimum consistent with operational efficiency. In such cases, the Fund has often sought to increase expenditure on materials and supplies in the face of governments’ reluctance to reduce employment and wages.
Budget execution. In connection with Fund-supported programs, governments have introduced a wide range of procedures to control expenditures. They have tightened cash disbursement procedures, implemented systems to monitor monthly or quarterly expenditures, and required operating ministries to keep to quarterly targets.
In conclusion …
Unsustainable government expenditures are often a cause of balance of payments difficulties and, therefore, controlling such expenditures is frequently an important element in policy packages adopted to bring about adjustment. In Fund-supported programs the Fund staff, in cooperation with officials of the member country, and often drawing on work done by the World Bank, analyze the country’s fiscal policy (including government expenditure) and may make certain recommendations. The ultimate choice, however, will be governed by the size and structure of the budget deficit, the severity of the overall adjustment that is needed, and, most important, the government’s priorities concerning the various elements of its expenditure policy.