Chapter

III Nature of Fiscal Adjustment

Author(s):
Sanjeev Gupta, Claire Liuksila, Henri Lorie, Walter Mahler, and Karim Nashashibi
Published Date:
June 1992
Share
  • ShareShare
Show Summary Details

While Section II focused on the aggregate measures used to assess the magnitude of fiscal adjustment during the program periods, the sources of fiscal adjustment are just as important as the size of the cut in the deficit, in terms both of reducing distortions (which may stem from the tax structure and expenditure policy), and of sustaining the measures utilized. A reduction in the fiscal deficit, achieved through cuts in investments for productive infrastructure or in expenditure on operations and maintenance, may not be sustainable over time, and indeed may be counterproductive, if such cuts lead, for example, to a deterioration of infrastructure and education and health facilities. On the other hand, expenditure savings can be effected by reducing “nonproductive” outlays such as prestige projects, subsidies that are not well targeted, or by trimming an excessively large civil service. Targeted cuts in both subsidies and the civil service can produce permanent savings without jeopardizing their objectives.

In the countries that reduced their deficits, the fiscal retrenchment was achieved through both revenue-enhancing and expenditure-cutting measures, although their relative contributions to the adjustment process have varied among countries. A country’s ability to undertake successfully a reduction in expenditure or an increase in tax revenue depends somewhat on its initial level of revenue or expenditure in relation to GDP. In particular, evidence suggests that it becomes difficult to raise additional revenue in developing countries when the revenue/GDP ratio is already around 20 percent, as in The Gambia, Kenya, Sri Lanka, and Togo. The tax base in most developing countries tends to be narrow, excluding relatively important sectors in the economy that are not fully monetized or incorporated in the formal economy, such as large segments of agriculture and services.15 Moreover, tax evasion by large segments of the formal economy is particularly widespread in the trade sector and among self-employed professionals and property owners. Thus, if the tax net captures only half of the economic activity, a 20 percent tax ratio translates into a 40 percent ratio on the average for those who pay taxes.

Revenue

More than half of the countries increased their tax revenue/GDP ratio vis-à-vis the base or pre-program year (Table 6, Chart 5). The tax ratio averaged 13.7 percent of GDP in the base year, declined to 13.4 percent in the pre-program year, and increased to 14.1 percent in the latest year of the program. This modest increase (0.7 percent of GDP over the program period) partly reflects the difficulty of raising the tax revenue ratio during a period of deteriorating terms of trade. It also reflects the widespread effort to reduce custom tariff rates and the proliferation of tax exemptions in response to recessionary conditions. The 12 countries that increased their tax ratio in relation to the base year did so by an average of 2.8 percent of GDP. The increase exceeded 5 percent of GDP in Bolivia, The Gambia, and Mozambique, and more than 3 percent of GDP in Ghana and Tanzania in comparison with the base year. Success in raising the tax revenue ratio was partly related to its initial level. Five countries that strengthened their tax effort had a tax revenue/GDP ratio of less than 10 percent in the base or pre-program years, while another 3 had ratios smaller than 15 percent. On the other hand, The Gambia, which already had a tax ratio of 20 percent in the base year (1986), succeeded in increasing it to almost 26 percent by 1990/91.

Chart 5.Indicators of Fiscal Adjustment: Tax Revenue1

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

1 For Guyana and Mozambique, in particular, large adjustments of the exchange rate have had an important effect on tax revenue.

Significant increases in tax ratios can be found in countries that embarked on a comprehensive reform of the tax system (for example, Bolivia and The Gambia) or implemented policies that reduced domestic distortions and realigned relative prices (Bolivia, Ghana, Mozambique, and Tanzania) by adopting a more realistic exchange rate and freeing some administered prices. On the other hand, the tax base shrank in most of the CFA franc countries sampled, a reflection of the recessionary conditions induced by the combination of a sharp decline in terms of trade and an appreciation of the real effective exchange rate. In addition to lower export tax receipts,16 the decline in the terms of trade reduced the real purchasing power of incomes generated in the export sector, and hence the demand for other traded goods (including imports of consumer goods), which tend to be a significant part of the tax base in many developing countries. The appreciation of the exchange rate also reduced the import unit value in relation to the GDP deflator and hence the real value of customs duties and taxes.17 As a result, the tax ratio in the CFA franc countries (in relation to GDP) fell from an average of 14.3 percent in the base year to 11.7 percent in the latest year of the program, which largely explains the difficulties these countries faced in strengthening government savings. In some countries (for example, Niger and Senegal), the problem was exacerbated by import liberalization and tariff reform, where lower tariffs and barriers to trade were not accompanied by a much larger inflow of recorded imports owing to the prevailing recessionary conditions. If the five CFA franc countries are taken out of the sample, the increase in the tax ratio for the other countries becomes significantly higher, rising from 13.6 percent of GDP in the pre-program year to 14.8 percent of GDP in the latest year of the program.18

While there was some improvement in revenue performance compared with the base or pre-program year, the actual increase in tax revenue fell short of the program objectives in most countries, partly as a result of deficiencies in the design, implementation, and sequencing of tax structure reform. A widespread objective was to shift from reliance on taxation of imports and exports to domestic taxation by broadening the tax base. This objective was pursued in line with trade liberalization and the need to promote productive efficiency by reducing the level of effective protection. A number of countries have also used the customs tariff to raise budgetary revenue by imposing import surcharges and other customs levies. Since such measures clutter the customs tariff and increase distortions, indirect taxes on domestic transactions would be a better revenue-raising vehicle. However, such a switch assumes that the administration for collecting domestic taxes can bring into the tax net the multitude of transactions and taxpayers targeted by the reform, clearly a much more difficult task than collecting customs duties from a confined customs space. Hence, when tariff rates were reduced, without a concomitant removal of tax exemptions and a strengthening of tax administration, revenue was immediately reduced. A considerable period of time generally elapsed, however, before efforts to improve tax administration yielded significant increases in revenue. Thus, the experience of SAF/ESAF countries indicates that caution needs to be exercised in setting program tax revenue targets, and that substantial increases in tax revenue generally take time. For instance, in relation to the pre-program year, only 9 of 23 countries succeeded in raising their tax ratio by 2 percent of GDP or more.

Average revenue from nontax sources19 also showed some improvement, rising from 3.0 percent of GDP in the base year to 3.4 percent in the latest year of the program (Table 7, Chart 6). This revenue stems largely from public price stabilization funds that monopolize the exports of some cash crops (such as coffee, cacao, and phosphate) and imports of petroleum products and basic food staples. But it also stems from profit transfers from the central bank and public enterprises. Nontax revenue increased in about two thirds of the countries in relation to the base year. The increase in Bolivia, Burundi, and Tanzania was particularly noticeable. These results in some cases reflected windfall gains on domestic sales of imported commodities (for example, petroleum products) at fixed domestic prices while import costs declined. In other countries (for example, Pakistan), the increase in nontax revenue reflected increases in public utility rates. Yet, in a number of countries (for example, Benin, The Gambia, Ghana, and Togo) nontax revenue declined sharply with the decline in international commodity prices. Typically, producer prices for cash crops set by price stabilization funds did not decline as fast as their export prices, resulting in substantial losses for these funds, which have been financed partly by budgetary transfers and resort to bank credit and partly by the accumulation of domestic payment arrears.

Chart 6.Indicators of Fiscal Adjustment: Nontax Revenue

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

In general, revenue from nontax sources should be interpreted cautiously, as changes may not reflect underlying trends in the government’s ability to mobilize resources. Such revenue is sometimes artificially or temporarily inflated by various accounting procedures or one-time transactions. Among these are transfers to the treasury of public sector bank profits, which may be more fictitious than real,20 the collection of past arrears and dividends owed by public enterprises or publicly owned banks, and sales of government assets. Similarly, the sharp declines in nontax revenue in some countries were influenced by reclassification of revenue from one category to another and by the specific nature of fees and charges that were not adjusted to reflect domestic price movements.

Expenditure

In comparison with the base year, the overall expenditure level (in relation to GDP) increased in more than half of the countries sampled (Table 8, Chart 7). However, expenditure was most significantly reduced in countries with high expenditure levels (around 30 percent of GDP) in relation to the base or pre-program years (for example, The Gambia, Malawi, Mali, Mauritania, Sri Lanka, and Togo). In the aggregate, total expenditure fell from 25.9 percent of GDP in the base year to 24.7 percent in the pre-program year, but remained virtually unchanged in the program period. However, both in comparison to the base year and during the program period, the CFA franc countries, with the exception of Niger, significantly reduced their expenditure levels.

Chart 7.Indicators of Fiscal Adjustment: Total Expenditure and Net Lending1

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

1 As noted in the text, the large increase in expenditure in Guyana and Mozambique is mainly linked to the effect of the depreciation of the exchange rate.

The average capital expenditure remained unchanged at 8.1 percent of GDP during the program period. When comparing the latest program year with the pre-program year, 11 of the 23 countries increased their capital expenditure/GDP ratio in line with the objective of enhancing fixed capital formation in productive areas (Table 9, Chart 8) (for example, Bolivia, Ghana, Lesotho, Nepal, Tanzania, and Uganda). Of these, 4 countries (Bolivia, Ghana, Tanzania, and Uganda) started off with very low public investment ratios (2–4 percent of GDP) while other increases were quite modest. In some countries, where capital spending was unusually high, part of the fiscal adjustment was secured by weeding out unproductive investments. But, in the majority of cases, despite the objective of promoting growth through higher productive investments, capital expenditure was lower than programmed, thereby facilitating the attainment of the overall deficit targets. There are four reasons for this result. First, a large part of the capital expenditure is foreign financed and very sensitive to supporting domestic policy in terms of the necessary technical and logistical work provided by the national authorities. Delays in such preparatory work or in disbursement by donors can lead to shortfalls in capital expenditure. Second, programs have at times set capital budget targets that in hindsight seem ambitious and unrealistic in relation to the implementation capacity of the countries. Third, bureaucratic and administrative delays in recipient countries, lack of government funds to finance the domestic portion of the costs, and delays in the release of counterpart funds of commodity aid by different donor agencies have often led to shortfalls in capital spending (for example, Bangladesh, Bolivia, Guyana, Niger, and Tanzania). Fourth, governments may find it easier to meet the overall targets by reducing capital spending than by cutting wages and salaries or by raising taxes. This may also create room for expansion in other expenditure categories if additional nonbank financing becomes available.21

Chart 8.Indicators of Fiscal Adjustment: Capital Expenditure

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

In some countries, raising capital expenditure was necessary to improve the core level of social and physical infrastructure, which would improve growth prospects. This was particularly true for Bolivia, Guyana, Madagascar, Mozambique, Tanzania, and Uganda. In other countries (such as The Gambia, Sri Lanka, and Togo) the reduction in capital expenditure reflected attempts to weed out nonproductive or low-priority investment. For instance, capital expenditure in Sri Lanka rose in the late 1970s, owing to increased foreign assistance and a sharp upswing in the prices of its primary commodity exports. But the higher capital expenditure may have exceeded the core expenditure necessary to support growth (as defined by the World Bank), suggesting that the cutback may not have hurt prospects for longer-term growth. However, in some countries such reductions may have reached the point where the maintenance and replacement of the existing capital stock was jeopardized. In Senegal, for example, public capital expenditure was reduced from 3.7 percent of GDP in the base year to 2.6 percent in 1989/90, despite higher program targets. This type of “adjustment” is clearly unsustainable and will have to be reversed in the coming years if an acceptable level of economic growth is to be assured.

In relation to the pre-program year, recurrent expenditure declined in 13 countries, with particularly large adjustments in Kenya, Lesotho, Malawi, and Mauritania. On average, current expenditure declined only marginally (from 16.4 percent of GDP in the pre-program year to 16.1 percent in the latest year of the program). Of the 10 countries that did increase their recurrent expenditure, Bangladesh, Nepal, and Uganda started off with very low levels (7.9 percent, 5.9 percent, and 6.2 percent of GDP, respectively) and may have moved toward a more sustainable level of recurrent expenditure (Table 10 and Chart 9). In the case of Mozambique and Sri Lanka, civil strife inflated military expenditure, while in other countries there was a marked slippage in expenditure control.

Chart 9.Indicators of Fiscal Adjustment: Current Expenditure

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

Reductions in recurrent expenditure have proved difficult to achieve. On the one hand, needs for health, education, the maintenance of infrastructure, and the delivery of other public services for a rapidly growing population are increasing. On the other, government employment often tends to be excessive, and manpower utilization inefficient, leaving substantial pockets of underemployed civil servants. Structural measures designed to curtail the growth of the civil service and to weed out areas of inefficiency are difficult to implement and take time to have effect. Moreover, government employment becomes a vested interest, and it is difficult for the government to scale down its size. This may explain why only four countries were able to reduce recurrent outlays by over 2 percent of GDP in three to four years of IMF programs, with most of the cuts affecting maintenance and purchases of goods and services rather than the size of the civil service.

The average total government wage bill in the SAF/ESAF countries increased marginally from 5.9 percent of GDP in the base year to 6.1 percent in the pre-program year, but was reduced slightly to 5.8 percent by the latest program year (Table 11). Wage drift, pressures to increase government employment in the face of rising population, and inadequate growth in employment opportunities in the private sector, which, in a number of African countries, have translated into automatic recruitment from civil service schools and guaranteed job security, have been systemic structural factors that have impeded the reduction and restructuring of the public sector wage bill. In a number of countries the government is still perceived as the employer of last resort. Nevertheless, in relation to the pre-program year, 13 of 21 countries reduced their outlays on wages and salaries as a percent of GDP, with particularly large reductions in Benin, Kenya, Mali, Malawi, and Mauritania (Chart 10). Notwithstanding the efforts at reducing the wage bill, the unsustainability of lax government employment and wage policies—often implemented in the context of years of high commodity prices—is particularly evident in Benin, Kenya, Senegal, and Togo, where the wage bills in the latest program year still averaged more than 8 percent of GDP. In Bolivia, the public sector wage bill rose over the program period; however, this increase partly reflected efforts to improve social services and the pay scales for health workers and teachers.

Chart 10.Indicators of Fiscal Adjustment: Wages and Salaries

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

Government interest payments, mostly to external creditors, ranged around 3.4 percent of GDP in most SAF/ESAF countries before programs were started (Table 12). In more than half of the countries for which data are available interest payments in the latest year of the program exceeded the base year or the pre-program year. On average, however, they were marginally lower than in the base or the pre-program year. This outcome was heavily influenced by sharp declines in cash interest outlays in Zaïre owing to the buildup of interest payment arrears. In comparison with the pre-program year, cuts in recurrent expenditure in 13 countries took place despite an often rising share of interest payments on both domestic and foreign debt. Banking sector reform and the associated liberalization of domestic interest rates have been an integral part of SAF/ESAF programs. Together with the high level of external debt inherited from the pre-program years, this has led to a substantial increase in interest payments. In the CFA franc countries, resort to the accumulation of domestic arrears to finance budget deficits understates the underlying domestic interest burden in recurrent expenditure. Of course, a part of the unpaid interest cost reappears in other expenditure categories, or in lower tax revenue, as suppliers make up for their own losses through higher prices or reduced tax compliance.

Disaggregated data on subsidies and/or transfers are only available for 15 countries (Table 13). Caution needs to be exercised in interpreting these data as some subsidies and/or transfers may have been classified in other budgetary expenditure items and are therefore not fully captured in the data. In general, the data show an increase in outlays on subsidies and/or transfers from an average of 2.1 percent of GDP in the base and pre-program years, to 2.5 percent in the latest year of the program. All of this average increase occurred in the first year of the program, and the average subsidy or transfer level has not changed from the first year to the latest year of the program. The increase in these expenditures in relation to the base year or pre-program years was particularly noticeable in Kenya and Tanzania, in part reflecting the poor performance of public enterprises and cooperatives. In Sri Lanka, where the increase was around 1 percent of GDP, additional expenditure arose from the Government’s decision to launch extensive poverty alleviation programs in 1989. These programs, by providing benefits to large segments of the population that could not be defined as poor, have been costly, and are being restructured by targeting transfers to the poorest segments of the population. Notwithstanding the aggregated results, the subsidy and/or transfer levels declined in about one third of SAF/ESAF countries. The substantial reduction in Mozambique was attributable to the freeing of agricultural prices, which reduced subsidies to the agricultural sector. Subsidies and transfers were also reduced in Niger, Senegal, and Togo in an attempt to lower the overall level of public expenditure, which was out of line with domestic revenue sources and the availability of foreign budgetary support.

Other recurrent outlays, including expenditure on social sectors and operations and maintenance, were above the base year levels, in about two thirds of the countries, in the latest program year (Table 14). The increase was substantial in Burundi, The Gambia, Mozambique, Senegal, and Tanzania. In a number of countries (for example, Bangladesh, Bolivia, and Mozambique), a deliberate effort was made to raise the main poverty alleviation components (health and education) and expenditure for operations and maintenance, which tend to fall dramatically under conditions of financial stress. Within health and education, efforts have focused on rehabilitating rural health centers (such as in Ghana), improving staffing of the primary health care facilities of the community-based health system (Malawi), and laying special emphasis on primary or basic education (Bangladesh, Bolivia, Malawi, and Niger). In some countries, attempts were made to reduce such unproductive expenditure as military expenditure (Pakistan and Zaïre), but on the whole, very little progress was achieved.

The relative contributions of expenditure-reducing and revenue-raising measures in lowering the average overall fiscal deficit and in strengthening the average current balance have depended on the reference year chosen. The average overall fiscal deficit improved by 1.7 percent of GDP in relation to the base year, while declining by only half that amount (0.9 percent) vis-à-vis the pre-program year. In the former case, nearly two thirds of the contribution to the deficit reduction (1.1 percent of GDP) came from expenditure-related measures, of which one half was from the recurrent component, whereas higher revenues provided the remaining adjustment of one third (0.6 percent of GDP, composed of tax measures of 0.4 percent and nontax measures of 0.2 percent). However, the composition of adjustment is radically different in comparison with the pre-program year; in this case, the bulk of the fiscal adjustment stemmed from the revenue side (around 0.8 percent of GDP), of which over four fifths was traceable to additional tax revenue (0.7 percent of GDP) and less than one fifth to nontax revenue. One plausible explanation for these differences in the role of revenue and expenditure measures in the adjustment process could be that the base year covered a period of relatively favorable terms of trade that permitted higher expenditure. With the subsequent deterioration in the terms of trade, expenditure was cut—a process that continued up to the year preceding the IMF program—which, however, proved inadequate for restoring internal balance. Thus, in relation to the pre-program year, the improvement in the fiscal position relied more on revenue measures.

Financing

In the pre-program period, a number of countries facing severe budget difficulties delayed payments to their suppliers and accumulated unpaid bills to cover expenditure. The increase in expenditure arrears temporarily reduces the reliance on domestic bank financing or on explicit borrowing from the public. However, it has a number of adverse consequences: (a) it complicates and/or thwarts the expenditure control processes; (b) it increases costs as suppliers of goods and services raise prices to cover the uncertainty of receiving payment; (c) it is not reflected in the interest rate structure, thereby resulting in a misallocation of capital; and (d) its adverse repercussions on the banking system cause in some cases liquidity crises and fragment the capital market. As indicated earlier, on average, the overall deficit was reduced by the latest program year by 1.7 percent of GDP compared with the base year and by 0.9 percent compared with the pre-program year. The financing data suggest that a part of this fiscal adjustment helped to accelerate a reduction in arrears by about 0.5 percent of GDP compared with the base year and by about 1 percent with respect to the pre-program year. Hence, the cash deficit and its financing declined by over 1 percent of GDP compared with the base year and remained virtually the same in relation to the pre-program year.22

The proportion of the deficit financed through foreign sources—measured in relation to GDP—increased rather sharply during the period under consideration. On its own, this increase is not an indication of rising foreign dependence because of significant exchange rate devaluations in many of the countries in the sample. However, estimates of foreign financing expressed in U.S. dollars comprising grants and loans (the latter adjusted for actual repayments) show that the flow of foreign resources to these countries increased from an average of $165 million in the base year to $311 million in the latest program year (Table 15). These resources helped finance an increasing proportion of total expenditures, from 23 percent in the pre-program year to 31 percent in the latest program year.23 However, this increase in external financing was not matched by an increase in public capital formation, which, as noted earlier, had been maintained at about 8 percent of GDP. Rather, it helped in undertaking structural reforms and reducing resort to inflationary borrowing from the banking system. Nevertheless, with a rising proportion of foreign resources and a relatively stable level of total expenditure, the dependence of the lowest-income countries on foreign funds has increased (Table 16).

In the aggregate, dependence on domestic sources for financing budgetary deficits in SAF/ESAF countries declined from 2.7 percent of GDP in the base year to 2.1 percent in the pre-program year and to minus 0.1 percent in the latest year of the program (Tables 17 and 18, Chart 11). The average financing from both bank and nonbank sources fell by 2.8 percent of GDP in relation to the base year, and by 2.2 percent vis-à-vis the pre-program year, with most of the reduction taking place in borrowing from the banking system. The decreased reliance on the domestic banking system for financing fiscal deficits is mostly due to two factors. First, the SAF/ESAF programs aimed at creating a stable environment for investment and growth by reducing inflation levels, through a reduction in overall credit expansion. They also aimed at freeing resources for the growth of the private sector. In a number of programs these two objectives have meant a reduction or elimination of bank financing of the fiscal deficit. Second, as noted earlier, the international community has supported the reform efforts of the developing countries by providing increased amounts of resources. Indeed, much of the commercial borrowing that had largely dried up since the debt crisis, has been replaced, and in some cases has been exceeded by official concessional assistance.

Chart 11.Trends in Average Financing

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

About three fourths of the countries in the sample reduced their dependence on domestic bank borrowing (Chart 12). Substantial reductions (in excess of 3 percent of GDP) were achieved in Bolivia, Burundi, The Gambia, Guyana, Lesotho, Madagascar, Mozambique, Senegal, Tanzania, Uganda, and Zaïre. In some instances (Madagascar, Mozambique, Uganda, and Zaïre), the reduction in domestic bank borrowing was facilitated by substantial increases in the net inflow of foreign resources. Attempts to reform the financial system and liberalize interest rates in SAF/ESAF countries apparently did not succeed in promoting the sale of government instruments (such as treasury bills and bonds) to the public at large. The second half of the 1980s was associated with widespread banking failures in developing countries, the accumulation of domestic arrears, and loss of confidence in the banking system and in public sector operations. Thus, the use of nonbanking sources to finance the total budgetary deficit actually fell by 0.6 percent of GDP. Nevertheless, about one third of the countries increased their use of nonbank private sector sources for financing the deficit, with significant increases in Guyana, Mauritania, Pakistan, and Sri Lanka. While a shift from bank to nonbank financing tends to reduce the rate of monetary expansion and inflation, excessive reliance on nonbank sources could be counterproductive, as it may crowd out the private sector from access to savings. It may also impose an additional interest burden on the budget so that the government must bid up interest rates to attract a greater amount of private savings.

Chart 12.Indicators of Fiscal Adjustment: Domestic Bank Financing

(In percent of GDP)

Sources: Data provided by the authorities; and IMF staff estimates.

The average external financing from both grants and loans fell from 6.5 percent of GDP in the base year to 6.1 percent in the pre-program year, but rose subsequently to 8.1 percent of GDP in the last year of the program (Tables 19 and 20). The average increase of over $100 million from the pre-program year, however, conceals the marked variation across countries. For instance, Bangladesh, Kenya, and Pakistan received over $300 million more in the latest program year than in the pre-program year, while Ghana, Mozambique, Uganda, and Zaïre each obtained an increase of about $200 million during the same period (see Table 15). While this trend illustrates, on the one hand, the support that the rest of the world has provided to the developing countries implementing structural reforms, on the other it reflects, as indicated earlier, a trend of rising dependence on foreign resources. On average, foreigners financed over 31 percent of total expenditure in the latest program year—up by about 8 percent from the pre-program year (see Table 16). This proportion stayed at a constant level between the first year of the program and the latest available year, suggesting that donors maintained their support to these countries throughout the program period. Of concern perhaps are the large increases (between 20 and 50 percent) in the proportion of spending covered by foreign sources in some of the SAF/ESAF countries (Ghana, Madagascar, Mozambique, and Uganda)—a level of dependence that may be difficult to sustain in the years to come (Chart 13).

Chart 13.Indicators of Foreign Dependence

(Base year = 100)

Sources: Data provided by the authorities; and IMF staff estimates.

Foreign grants (expressed in domestic currency) increased in over two thirds of the countries, rising by an average of 0.6 percent with respect to the base year and 0.7 percent with respect to the pre-program year. The increase was over 3 percent of GDP in Burundi, Guyana, Lesotho, Madagascar, Mozambique, and Uganda. A similar picture emerges with regard to foreign loans; the increase was 0.9 percent of GDP with respect to the base year and 1.3 percent in comparison with the pre-program year. Increases over 3 percent of GDP can be found in The Gambia, Guyana, Mozambique, Tanzania, Uganda, and Zaïre. However, as with grants, part of the increase in these countries is of an accounting nature to the extent that foreign financing was undervalued in the past owing to an overvalued exchange rate.

The fiscal programs in SAF/ESAF countries have attempted to reduce the incidence of payment arrears and the carrying forward of unpresented checks by incorporating measures that strengthened expenditure monitoring and control procedures. About half of the countries (10) in the sample experienced episodes of domestic arrears accumulation, particularly during the second half of the 1980s. Consequently, fiscal programs in these countries have attempted their phased liquidation in the calculation of the cash deficits. On average, arrears ranging between 1 and 2 percent of GDP were cleared by the latest program year in these ten countries. In Kenya, repeated attempts at eliminating government domestic arrears (checks issued but not presented for collection) were unsuccessful, and the budgetary deficits in some years were financed by increasing arrears.

15

For a description of the characteristics of the tax system in a developing country, see Vito Tanzi, “Quantitative Characteristics of the Tax Systems of Developing Countries,” in The Theory of Taxation for Developing Countries, ed. by David Newbery and Nicholas Stern (New York: Oxford University Press, 1987).

16

These taxes have been mostly implicit, as they frequently took the form of transfers to the government from commodity stabilization funds.

17

Thus, even if the elasticity of the tax system with respect to real domestic income had been unity, the decline in the terms of trade and the appreciation of the exchange rate would have led to a decline in the tax revenue/GDP ratio. For a detailed discussion, see Vito Tanzi, “The Impact of Macroeconomic Policies on the Level of Taxation and the Fiscal Balance in Developing Countries,” Staff Papers, International Monetary Fund (Washington), Vol. 36 (September 1989), pp. 633–56.

18

Excluding Guyana and Mozambique.

19

Nontax revenue includes interest, rent, profit, and dividend receipts as well as administrative fees and fines.

20

If public sector banks extend credit to nonperforming public enterprises their “profits” should normally be written off against losses on their nonperforming portfolio. Instead, they are transferred to the government, which will eventually have to bail out these banks.

21

Reducing capital spending may also reduce foreign financing. Hence, expanding other expenditure categories, while meeting overall ceilings on bank credit expansion targets, would typically be done by resorting to nonbank financing.

22

This can be clarified with the help of the figures below. The level of financingin the latest program year was higher than the overall deficit, which facilitated a reduction in payments arrears.

Base YearPre-Program YearLatest Program Year
(Averages, in percent of GDP)
Overall deficit9.18.37.4
Total financing9.28.28.0

23

Part of this increase may be due to past undervaluation of foreign assistance in some countries owing to the maintenance of an over valued exchange rate.

    Other Resources Citing This Publication