Chapter

III. Financial and Structural Policies: The Broad Front for Reform

Author(s):
F. Rozwadowski, Siddharth Tiwari, David Robinson, and Susan Schadler
Published Date:
June 1993
Share
  • ShareShare
Show Summary Details

The exceptionally weak economic performance and structural characteristics of the ESAF countries meant that a wide range of policy changes and reforms were needed. Generally, programs aimed at containing excess demand as well as addressing structural weakness, although in most countries the need for stabilization was not as dire as in many IMF-supported adjustment programs in Latin America and Eastern Europe. For most of the countries under review, the strategy underlying the structural reform program was to strengthen the financial position of the public sector and to reduce the government’s interference in the allocation of resources.

Typically, the reforms needed were closely interrelated. For example, in many countries reform of public enterprises and marketing boards was essential to reducing subsidies, strengthening the tax base, and addressing, in a lasting manner, weaknesses in the banking system. In turn, when currencies were overvalued, reforms in marketing boards and price controls required exchange rate adjustments if incentives to exporters were to be improved or maintained. Depreciations, however, could not be sustained in real terms without fiscal adjustment. Since there was no obvious place to break into this circle from initial positions that were typically highly distressed, the approach taken was generally as broad-based as possible.

In practice, the aggressiveness with which policies were implemented varied. In broad qualitative terms, there appears to have been more progress with reforms to exchange and trade systems, price controls, and marketing boards than in restructuring public enterprises or addressing weaknesses in the financial system. In most countries, central government deficits were reduced but often without the needed broad-based tax and expenditure reforms. Although the reasons for the differing success and pace of reforms were varied, a few general themes emerge from this review. First, the political commitment to reform was critical; in light of the history in most of these countries of centralized control and government ownership of productive activities, such commitment often required a break with deeply rooted views about the political economy. Second, vested interests often aligned themselves against reforms, and ways to soften the blow to these groups and persuade them of the longer term benefits of reform were important. Third, deteriorating terms of trade often impeded reform initiatives, both because they masked the benefits of reform, thereby eroding political support, and because they presented an urgent adjustment problem of their own. Fourth, administrative capacity was exceptionally weak, and many delays reflected inappropriate initial expectations of how quickly the technical expertise could be developed to design, implement, and sustain the reforms.

In this latter connection the IMF, the World Bank, and many other bilateral and multilateral agencies offered extensive technical assistance. The IMF’s particular contributions were concentrated in the areas of tax design and administration, central government budget formulation and expenditure control, financial sector reform (particularly the development of indirect means of monetary control, securities and money markets, and interest rate liberalization), and exchange system reform. The IMF also provided technical assistance to help in the development of statistics, especially in the areas of monetary accounts, balance of payments, and national accounts—areas where data deficiencies severely impeded the identification of policy problems and the monitoring of programs addressing them.

All SAF/ESAF arrangements were the outcome of strengthened collaboration among the country authorities, the World Bank, and the IMF under the umbrella of a policy framework paper (PFP) (see Box 1). Typically, different parts of the adjustment program were supported by regional development banks and bilateral donors, and it is rarely possible to segregate the effects of SAF/ESAF arrangements from the effects of International Development Association (IDA) and other credits. Nevertheless, it is clear that the PFP mechanism has enhanced the coordination of the IMF’s support for macroeconomic adjustment and structural reforms with the World Bank’s support for structural reforms in diverse areas such as trade, taxation, financial intermediation, public enterprises, public expenditure and investment, agricultural marketing, and the social dimensions of adjustment.

Public Sector Policies

Reforms of the public sector were typically geared toward strengthening its financial position and narrowing its focus to activities that the private sector could not perform efficiently. Frequently, however, the first of these objectives was hindered by the lack of consolidated public sector accounts that would identify the source of financial imbalance within the public sector. There were, however, indications in most countries that financial weaknesses permeated the various levels of the public sector. Consequently, reforms were targeted at all levels except, in most countries, local governments. Often, local government activity appeared minor relative to other segments of the public sector and. in any case, largely out of the control of the central government.

Central Government

Measuring the strength of adjustment of a central government is not as straightforward as it might seem. The improvement in a government’s financial position is, of course, one readily available indicator. But in SAF/ESAF programs, structural changes in the areas of taxation and the controlling and prioritizing of expenditures were often as important as improving the financial position itself; yet, these changes did not always produce immediate positive results. Even when they did, such results were sometimes masked by the effects of adverse weather or changes in the terms of trade or other exogenous influences on revenues and expenditures.

Nevertheless, the imperative of stabilizing the economy meant that each of the countries did target an improvement in the broadest measure of the central government’s net Financial influence on the economy—the deficit including official grants. In a few countries, however, where sizable reductions in the deficit had occurred in the three years prior to the SAF/ESAF period, the targeted additional improvements were very small. In fact, the government’s financial position including grants did improve in 13 of the 19 countries (Chart 5 and Appendix Table A9). In three of these countries-Guyana, Lesotho, and Mozambique—the cumulative improvement relative to GDP was more than 9 percentage points; although the deficits of these countries, except Lesotho, remained large, the reduction represented a distinct break from recent experience. For the other ten countries, the reduction in the deficit relative to GDP was more modest: on average the cumulative change over the average five years of their SAF/ESAF arrangements was about 2 percentage points of GDP compared with an average pre-SAF deficit of about 6 percent of GDP4 For the six countries that experienced a deterioration in their fiscal position, the decline amounted to about 1½ percentage points over a cumulative average five-year period, compared with an average pre-SAF deficit of about 4 percent of GDP.

Chart 5.Central Government Financial Profiles

(In percent of GDP)

Source: Appendix Table A9.

1 Data not available for Bangladesh.

While the central government deficit including grants is the best statement of a government’s financial impact on the economy, the adjustment effort—to the extent that it can be captured in a financial balance—is better reflected in the government’s primary current account excluding grants.5 The directions of the changes in primary current accounts largely matched those of the overall balances (Chart 5 and Appendix Table A9). Typically, however, the primary current account improved less or deteriorated more than the overall balance. This reflects the fact that, on average, scheduled interest payments fell during the SAF/ESAF period, while grants rose by slightly more than capital spending; these two factors together accounted for about 40 percent of the improvement in the overall balance of the central government including grants in the 13 countries where an improvement occurred. The drop in interest payments resulted from declines in domestic interest rates, as inflation in some countries fell, and from lower scheduled interest payments on foreign debt in countries where the exchange rate was stable.

Increases in revenues generally accounted for most of the reduction in primary current account deficits (Chart 6).6 These increases mostly stemmed from higher tax receipts and often reflected two main influences: first, during the early stages of the SAF/ESAF period, policy changes that eliminated the parallel market and thereby increased the share of economic activity in the tax net: and second, structural reforms to the tax system. Most countries undertook some tax policy measures to broaden the tax base and improve tax administration, but only in a few countries were these changes part of more comprehensive reforms. Mozambique stands out as an exception; sweeping changes in the tax system and a strengthening of tax administration contributed to a doubling of tax revenues relative to GDP Other significant measures were the introduction of value-added taxes in six countries—Bangladesh, Bolivia. Kenya, Malawi, Niger, and Senegal—mostly just prior to or early in the program period.

Chart 6.Central Goverment Revenue and Grants

(In percent of GDP)

Source: Appendix Table A9.

1 Data not available for Bangladesh.

Most of the countries that suffered sharp reductions in their terms of trade owing to falling export prices—Bolivia, Burundi, Ghana, Madagascar, Niger, and Uganda—were reasonably successful at insulating revenues. Although receipts from export-related taxes fell sharply in these countries, total revenues generally rose slightly relative to GDP, The two exceptions were Madagascar and Niger, where political disturbances contributed to a breakdown in tax administration, and revenues dropped to less than 10 percent of GDP In some countries, the effect of the drop in export prices was cushioned by exchange rate changes, which lessened the impact on the export tax receipts and raised receipts from ad valorem import taxes. Generally, however, these countries had to act aggressively through other tax measures to compensate for the loss of export-related tax receipts. Even though these efforts succeeded in preventing a fall in revenues relative to GDP. overall fiscal deficits usually exceeded targets.

Current, noninterest expenditure was restrained in most countries: spending relative to GDP declined in more than half the countries, and there was a large absolute reduction in Guyana (Chart 7). For most countries, efforts to restrain expenditure focused on subsidies and civil service wage bills. The scope for reducing subsidies was closely tied to reform of public enterprises, marketing boards, and price controls, which are discussed later in this section. The net effect of these reforms is not transparent, however, because subsidization often took such indirect forms as tax exemptions or preferential credit for public sector entities or portions of the private sector. Civil service reform of varying degrees look place in 13 countries. The reforms focused on creating a smaller but more productive civil service by eliminating ghost workers and redundant tasks. As a first step, many countries undertook a civil service census to assess the magnitude of the problem. A few countries then proceeded to alter salary structures to attract or retain experienced staff. Several countries froze recruitment or actually reduced staff. However, the scope for action on the civil service was constrained in some countries by large increases in capital expenditures, which generated additional recurrent staffing requirements for maintenance.

Chart 7.Central Government Expenditures

(In percent of GDP)

Source: Appendix Table A9.

In most ESAF countries, comprehensive public expenditure reviews were needed to guide fiscal consolidation and to increase the efficiency of public spending—both recurrent and capital. In almost all countries, efforts were made to improve the prioritization and monitoring of government outlays. For recurrent expenditures, systems for classifying and reconciling accounts between the Treasury, other government accounts, and the central bank were improved. For capital expenditures, rolling public investment programs or reviews of public investment were typically carried out by the World Bank. The implementation of these programs and recommendations was frequently part of the macro-economic conditionality of IDA credits. These efforts have met with varying degrees of effectiveness and need to remain an important focus for future programs.

Rehabilitating and extending infrastructure was begun under the SAF/ESAF arrangements in several countries. In Guyana, Mozambique, and Uganda—where negligence, war, or forces of nature had destroyed much of the infrastructure—a vigorous effort to rebuild and rehabilitate was started. Capital expenditures (relative to GDP) increased in 10 of the 19 countries but, on average, remained broadly unchanged. Excluding Mozambique and Uganda, the average declined by nearly 1 percentage point of GDP. Declines often resulted from poor implementation capacity, delays in meeting conditions for donor financing, difficulties in mobilizing counterpart funds, and in a few countries, the cutting of capital expenditures to meet deficit reduction targets.

Nonfinancial Public Enterprises

Each of the SAF/ESAF-supported programs included a commitment to at least initiate public enterprise reform. Reform typically involved three stages: first, preparing studies to identify problems and propose solutions; second, implementing temporary measures to address the most egregious problems while legal changes for more far-reaching reforms were enacted; and third, carrying out fundamental reforms. A variety of such reforms were envisaged: enterprises considered strategically important were to be restructured or subjected to conditions approximating commercial standards: nonstrategic loss-making enterprises were to be liquidated; and nonstrategic enterprises with commercial value were to be privatized.

All these reforms encountered obstacles. Frequently, political commitment was lacking. In most of the countries under review, the public enterprise system embodied an economic ideology in which the state was to spearhead development in large part through its ownership of major enterprises. Recognition of the drawbacks of this strategy has only recently taken hold in some countries and is yet to come in others. Further important factors constraining progress were fears of output and employment losses; practical problems such as political unrest and the need to pace privatizations so as to avoid compromising the market value of enterprises to be sold and to develop some form of social safety net to protect redundant workers; and the lack of administrative capacity to implement reforms even when a plan for action did exist. In retrospect, the World Bank and IMF staff, as well as the authorities, have tended to underestimate the time required to design and implement reforms.

Only a few countries—Bolivia, The Gambia, Guyana. Senegal. Sri Lanka, and Togo—made significant inroads into the problem of public enterprises. While enterprise reform could not be considered complete in any of these countries, each did formulate a comprehensive reform program, enact needed legal changes, and implement parts of the program. With the exception of Guyana, these countries had seriously attempted—although usually in a halting fashion—to address problems in the public enterprise sector prior to the SAF/ESAF period. They therefore had formulated an agenda for reform even if some refinements for the SAF/ESAF period were needed. Also, except in Bolivia, legal changes required to proceed were not major. The substantive policies implemented in these countries took two main forms:

  • Restructuring (principally to improve management and reduce workforces), usually reinforced by “performance contracts” between enterprises and the government that established commercial standards for enterprises. Guyana took a different route in employing management contracts with foreign companies to strengthen the operations of several major enterprises.

  • Divestiture, although often of relatively small enterprises, or ones that did not constitute the heart of the problem in the enterprise sector. Exceptions were Guyana, where divestment was extensive, and Sri Lanka, where a large retrenchment of the labor force took place. Enterprise liquidation on a significant scale occurred only in Guyana, Senegal, and Togo.

Indicators of the financial performance of the enterprise sector are generally inadequate: in Guyana and Bolivia, where data do exist, performance appears to have improved markedly. Nevertheless, public enterprise reform in most of these countries is far from complete. In general, these countries need further divestiture and liquidation as well as a better definition of financial standards for enterprises that will remain under public ownership. Improving the data on the financial position of the enterprise sector will also be critical to monitoring the success of these efforts.

The other countries under review have not proceeded far in addressing the problems of public enterprises. A few—Burundi. Guinea, Malawi, Mauritania, Mozambique, and Niger—made progress in formulating plans to restructure or privatize some enterprises and in securing necessary legal changes. Others either did not have an articulated plan for reform or spent much of the SAF/ESAF period preparing such a plan. By and large, public enterprises remained a significant burden on the government budget for these countries, and any improvements in the financial position of enterprises tended to result from increases in their administered prices.

Marketing Boards and Trading Companies

For many countries, the most important reforms of the marketing system affected marketing boards for export commodities.7 These reforms, in which support from the World Bank played an important role, generally had three aims: (1) to eliminate the monopoly of publicly owned marketing boards in the purchase, transportation, processing, and marketing of export commodities; (2) to allow market forces to determine producer prices; and (3) when marketing boards were not abolished, to improve their financial profiles. Similar objectives for reform applied to trading companies for domestically consumed goods (imported and domestically produced). Reforms of marketing boards and trading companies were one leg of the general effort to place greater reliance on markets to guide production and consumption—the other two were decontrolling prices and liberalizing exchange and trade restrictions.

Often, reforms to marketing boards were contingent on large exchange rate adjustments and changes in the exchange system. This was true for two common situations. In the first, marketing boards’ losses were due largely to their attempts to counter the effects of an overvalued domestic currency by setting domestic producer prices above the official domestic currency value of world market prices. If export volumes were to be preserved while the financial position of marketing boards was strengthened, a depreciation, often with higher foreign exchange retention rates for producers, was essential. A second common situation was one in which producer prices had been held below the domestic currency value of world market prices, thus bolstering the financial positions of frequently inefficient marketing boards. Except when large savings could be generated by eliminating marketing boards or improving their efficiency, depreciations were also needed to raise returns to producers without sacrificing the financial position of the marketing boards.

Several countries—The Gambia, Guyana, Niger, and Uganda—came close to completing comprehensive reforms of their export marketing boards, although in Niger the problem was of a relatively small magnitude. The pattern of reform was similar in these countries: a step devaluation (except in Niger, where exchange rate action is precluded by its membership in the West African Monetary Union, or WAMU) that permitted an attractive level of producer prices to be established while the financial position of the marketing boards was improved; the elimination of guaranteed minimum producer or support prices; permission for private participation in activities previously monopolized by marketing boards; and the termination of subsidies on agricultural inputs. As a result of such reforms, most producer prices in these countries are now either freely determined in markets or set by marketing boards on the basis of world prices they receive. Generally, marketing boards have been stripped of activities that could be performed more efficiently by the private sector; in most of their remaining activities they do not hold monopoly positions.

In other countries where marketing boards were a problem, reforms were more piecemeal. In many of these countries, steps were taken to improve the financial positions of marketing boards (shedding labor and reducing input subsidies provided through the marketing boards) and to reduce the rigidity with which producer prices are set. The latter often involved moving producer prices closer to world prices or shifting to minimum indicative prices. In a few countries, private participation was introduced in some activities previously monopolized by marketing boards for both exports and domestically consumed goods; limited divestiture took place in two countries. Nevertheless, in all these countries, marketing boards retain a significant degree of monopoly power and producer prices are still administratively controlled or heavily influenced by the marketing board.

Price Controls

In some countries where controls on the availability of goods and foreign exchange were initially widespread, prices were de facto free of control owing to the prevalence of parallel markets. In these countries, removing official controls served mainly to draw parallel market transactions into the official net. Although this process usually enhanced fiscal revenues, its incentive effects on output were comparatively weak. However, when price controls had been less distortive or commodities subject to control had been monopolized by the state, price liberalization had significant real effects on production and consumption decisions.

Substantial progress was made in decontrolling prices, particularly in the agricultural sector, and this change appears to have had positive effects on output. In Bolivia (except for petroleum), Guinea, Madagascar, and Lesotho, prices were largely free prior to the SAF/ESAF programs and remained so throughout the arrangements. In The Gambia, all price controls were lifted during the arrangement. In four other countries (Guyana, Malawi, Niger, and Uganda), price controls on all except a handful of goods considered strategic—principally petroleum and/or utilities—were terminated. For petroleum products, changes in international prices and in the exchange rate were mostly passed on to domestic consumers in a short time. In the remainder of the countries, general controls on prices of goods and on profit margins were abolished and replaced by selective controls.

Exchange and Trade System

Exchange and trade reform—replacing government control over the allocation of foreign exchange with market mechanisms—was the area where most countries were able to make the most forceful changes. In general, the administrative capacity needed to make these changes was less than for many other types of reform. Also, as with price decontrol, reforms to the exchange and trade system tended to meet with relatively little resistance, particularly when a sizable share of the transactions had taken place in parallel markets. In these circumstances, the principal effect of liberalization was to attract foreign exchange flows from parallel markets to official channels, thereby raising government revenues and official foreign exchange receipts. When activity had been more constrained by official controls, the salutary effects of liberalization—particularly strengthening incentives for export production and rationalizing import consumption-were usually quickly evident.

Because exchange and trade restrictions had usually been aimed at officially controlling the allocation of foreign exchange, reforms to exchange and trade systems needed to be supported by changes in exchange rate policy. In the short run, exchange and trade liberalization usually reduced the equilibrium value of domestic currency; however, once the benefits of liberalization had had their full effect on the volume of exports, the equilibrium real exchange rate should have risen. In fact, the countries that undertook the most forceful reforms did devalue or allow the value of their currencies to float down. There has, however, been little sign of any rebound in the equilibrium real exchange rate—probably because of concurrent pressures on the external position from other sources.

Except in the three countries where exchange and trade regimes were broadly free to start with (Bolivia, The Gambia, and Lesotho).8 exchange and trade reform figured prominently. Bangladesh, Ghana. Guinea, Guyana, and Malawi moved to or close to complete freedom of controls over current international transactions. In Ghana and Guyana, this involved the effective dismantling of the import licensing system while allocating an increasing share of official foreign exchange receipts through market mechanisms—in Ghana, a Dutch auction later supplemented by foreign exchange bureaus, and in Guyana, a legalized parallel market. In these countries, complete unification of exchange markets with full retention of export proceeds was attained during the ESAF period. In Bangladesh, the official and secondary markets were unified through the gradual shifting of transactions from the former to the latter. Malawi and Guinea required less sweeping changes. In Malawi, control over imports through the exchange system was abandoned except for a short negative list of restricted imports, and tariff reform was implemented. And in Guinea, where the trade restrictions had been liberalized during a previous IMF stand-by arrangement, participation in foreign exchange auctions was broadened.

In Mozambique, Tanzania, and Uganda, there had been several earlier but unsuccessful attempts to unify official and parallel market exchange rates through devaluations of the official rate. The SAF and ESAF arrangements aimed at unification through more fundamental systemic changes—principally, introducing or increasing the scope of partial retention of export proceeds and establishing secondary markets for foreign exchange, financed by receipts from nontraditional exports and private transfers. Exchange system reform went the furthest in Uganda, where the parallel market was absorbed into a bureau system, and the few remaining restrictions were generous. However, the exchange rate established in auctions of receipts from donors remained below the bureau exchange rates throughout the arrangement. In Tanzania and Mozambique, there remained considerable, though much reduced, parallel market premia as well as surrender requirements and other controls. In each of these countries, reform of the exchange system was more forceful than that of the trade system. Each introduced or extended the coverage of an open general license system, but this measure generally followed exchange reform and left considerable restrictions and high tariffs in place.

Several countries—Niger. Senegal. Sri Lanka, and Togo—that entered the SAF/ESAF period relatively free of exchange restrictions needed to address still-extensive trade restrictions and to reduce high tariffs. Some progress was made in these areas through the elimination of several quantitative restrictions and reduction and rationalization of tariffs.

Even those countries whose exchange and trade systems were broadly free introduced measures to improve their efficiency. In The Gambia, foreign exchange bureaus were established, and in Bolivia, tariffs were reduced and rationalized. Reform in Lesotho was constrained by its membership in the South African Customs Union.

In several of the ESAF countries, progress made earlier in the reform of the exchange and/or trade system was later set back as delays in receipts from donors tightened the foreign exchange constraint. This back-tracking occurred in Kenya, Madagascar, and Malawi, where previous exchange liberalization was partially reversed or made subject to administrative interference. In two other countries—Burundi and Mauritania—reforms were more piecemeal from the beginning and focused rather narrowly on removing quantitative restrictions on imports or tariff reform while moving to more flexibility in the exchange rate.

In many countries, SAF/ESAF programs also laid the groundwork for promoting foreign direct investment through two principal types of change: first, the propagation of a more stable and liberal macro-economic environment with avenues for the legal and timely repatriation of profits, and second, changes in the regulatory framework to reduce discrimination against foreign direct investment. There is evidence of some increase in foreign investment in several countries. In many this was directed, in the first instance, to the services sector, especially tourism. Limited investment also took place in agriculture, especially horticulture. Except in Guyana and Bolivia, there is little evidence of increased investment in other tradables. However, real wages in many ESAF countries are low by international standards, and sustained and credible reforms, particularly of the exchange and trade system, should have the potential to attract greater foreign investment.

Exchange Rate Policy

Many of the countries under review had significantly depreciated the nominal value of their currencies during the two years before the SAF/ESAF period, and they had succeeded in depreciating in real terms also but by less than in nominal terms. Nevertheless, there was a perception that in most countries further depreciation was needed, for several reasons: to support ongoing trade and exchange reform; to correct residual overvaluation when previous efforts had been thwarted by inadequately supportive fiscal policies; and to respond to a deterioration in the terms of trade.

Currencies of all countries except those using the CFA franc (Niger, Senegal, and Togo) depreciated during their SAF/ESAF arrangements in nominal terms against the major currencies (Appendix Table A10). However, because several of these currencies—those of Bolivia, Malawi, and Mauritania—depreciated less than other (mainly local) trading partners, they appreciated in nominal effective terms (Table 1 and Appendix Table A11). Movements in the exchange rate of the French franc vis-à-vis currencies of other trading partners caused Togo and Senegal to appreciate in nominal terms and Niger to remain stable. The nominal effective rate for Lesotho remained stable because it is pegged to the South African rand, which dominates in the effective exchange rate weights.

Table 1.Developments in Effective Exchange Rates(Annual average percentage change)
CountrySAF

Period
ESAF

Period
Combined

SAF/ESAF

Period
CountrySAF

Period
ESAF

Period
Combined

SAF/ESAF

Period
I. Countries where nominal effective exchange rates depreciated during their SAF/ESAF arrangements
BangladeshKenya
Nominal–5.40.4–3.5Nominal–4.4–5.2–5.0
Real–3.4–3.8–3.5Real–5.0–4.0–4.2
BurundiMadagascar
Nominal–12.1–2.2–10.5Nominal–29.11.2–12.2
Real–10.60.4–8.9Real–23.0–4.2–12.2
Gambia, TheMozambique
Nominal–32.83.9–10.1Nominal–62.1–28.1–51.0
Real–12.9–2.0–5.8Real–36.3–8.9–26.5
GhanaSri Lanka
Nominal–40.4–10.2–17.2Nominal–10.2–1.1–8.0
Real–22.9–1.9–6.5Real–1.95.3–0.2
GuineaTanzania
Nominal–41.7–28.8–58.5Nominal–28.76.9–22.7
Real12.2–19.9–9.3Real–35.6–1.1–29.9
Guyana1Uganda
Nominal–50.6–50.6Nominal–56.6–36.8–45.6
Real–21.9–21.9Real7.6–26.6–14.5
II. Countries where nominal effective exchange rates appreciated during their SAF/ESAF arrangements
BoliviaSenegal
Nominal13.473.159.0Nominal5.34.44.7
Real–3.6–5.6–5.2Real1.5–4.6–2.6
Malawi1Togo
Nominal3.43.4Nominal3.08.77.2
Real4.54.5Real–5.8–2.9–3.6
Mauritania
Nominal1.112.98.0
Real–6.5–1.2–3.4
III. Countries where nominal effective exchange rates were unchanged during their SAF/ESAF arrangements
LesothoNiger
Nominal–0.4–0.1–0.3Nominal–0.90.4–0.1
Real–1.12.0–0.4Real–7.5–7.1–7.3
Source: IMF staff estimates.

No SAF arrangement.

Source: IMF staff estimates.

No SAF arrangement.

Developments in nominal effective exchange rates were only partially reflected in real effective exchange rates. Although all of the 12 countries that depreciated in nominal effective terms experienced a real effective depreciation during the SAF/ESAF period, only 7 saw real depreciations of more than half the nominal depreciation. In these countries-Bangladesh. Burundi, The Gambia, Kenya, Madagascar, Mozambique, and Tanzania—domestic demand was significantly restrained by relatively large fiscal adjustments and/or by declining export prices. Other influences that are likely to have contributed to substantial real depreciations in some countries were a lack of extensive wage indexing and a relatively large amount of unused labor. These circumstances probably helped contain pressures on wages and consumer prices following the nominal depreciations. Among the seven countries whose currencies appreciated or remained stable in nominal effective terms during their SAF/ESAF arrangements, all except Malawi experienced real effective depreciations. This situation reflected the lower level of inflation in these countries compared with that in trading partners.

For most of the 19 ESAF countries, relative consumer price indices are not always the best indicator of the price incentive for exporters. More accurate indicators are hard to construct, but the ratio of export unit values to the consumer price index (a proxy for domestic costs) can be informative. This measure suggests that in all countries except Lesotho, Malawi, Niger, and Togo, the price incentive to exporters improved, notwithstanding often large deteriorations in the terms of trade (Appendix Table A12). For most countries—even several that experienced the largest terms of trade losses—gains in this more specific indicator of export competitiveness were greater than those in relative consumer price indices. Most of the improvement in the price incentive for exporters occurred in the early years of SAF/ESAF arrangements; since 1989 almost every country has seen a leveling off or slight deterioration in this measure. This trend reflects both some diminution in the depreciation of nominal exchange rates and the more pronounced weakening of export prices for some countries since 1989.

Financial Sector Policies

Under SAF and ESAF arrangements, financial sector policies were aimed at constraining credit growth, especially to the public sector, while enhancing the role of market mechanisms in allocating credit, shifting to indirect means of monetary control, eliminating quasi-fiscal activities of the banking system, and strengthening financial institutions. Virtually all SAF- and ESAF-supported programs had an agenda in these areas, but progress in implementing them generally proved slower than expected. In general, delays arose because of weak technical and administrative capacity; the long time required to develop competitive banking structures; and impediments posed by inadequate reforms in other areas, particularly public enterprises, which banks were often forced to support.

Putting aside Bolivia and Uganda, where extremely rapid money growth was reduced sharply, the growth of broad money in the countries under review remained broadly unchanged from the period prior to SAF/ESAF arrangements; in eight countries, growth diminished (relative to the prior rate) and in another eight, growth increased (Table 2). Developments in domestic credit were heavily affected by progress in reducing the financing requirements of the public sector. The growth of domestic credit dropped or was unchanged in all countries, and the accumulation of foreign assets rose. Conclusions on the composition of credit are not possible, owing to the lack of comparability in the classification of private and public sector credit.

Table 2.Money, Interest, and Credit
Increase in Broad Money (Annual averages, in percent of beginning of period broad money)Increase in Domestic Credit (Annual averages, in percent of beginning of period broad money)Interest Rates and Inflation Pre-SAF or Pre-ESAF4 (In percent)Interest Rates and Inflation at end-1991 (In percent)
Nominal5Real6Nominal5Real6
Pre-SAF or pre-ESAF1SAF2ESAF3Pre-SAF or pre-ESAF1SAF2ESAF3Loan rateDeposit rateLoan rateDeposit rateInflation (CPI)Loan rateDeposit rateLoan rateDeposit rateInflation (CPI)
Bangladesh2816133013816125111211547
Bolivia48456863883913–1666381921415
Burundi167315631076341510619
Gambia, The20321232–49–32019–22–2252261413212
Ghana625326636–22221–9–1033272015910
Guinea73030443915121315–31–306523209712
Guyana793563111513–44–451053129–28–2982
Kenya17819239211410410291517410
Lesotho17201208–51481–41325139—114
Madagascar1819198–3–21315–21151211329
Malawi2014192016411518145112
Mauritania141012111142129418127834
Mozambique1551342643262016–54–561633434–1–135
Niger5962–8–301061511–41671561
Senegal393910–9116615167167
Sri Lanka111622162113181071020141059
Tanzania7203641354421914–11–153326234121
Togo8–847–211451451671561
Uganda17716556164116494030–58–6123739359628
Mean (excl. Bolivia and Uganda)23211825851412–6–93221157215
Median18191922103151341523149410
Source: IMF staff estimates.

Average during three years preceding first SAF- or ESAF-supported program.

Average during period from first SAF-supported program to year preceding first ESAF-supported program.

Average during period commencing with first ESAF-supported program.

Interest rates at end of calendar or fiscal year preceding the first SAF- or ESAF-supported program.

Commercial bank loan and time deposit rates. The rate used is a representative rate, the midpoint of published rates, or where available, the weighted average of interest rates.

Real interest rates, given by rr = 100((1 + rn/100)/(1 + cpi/100) – 1) where rr is the real interest rate, rn is the nominal interest rate, and cpi is the CPI inflation rate.

Pre-SAF or pre-ESAF averages are for one year only.

Source: IMF staff estimates.

Average during three years preceding first SAF- or ESAF-supported program.

Average during period from first SAF-supported program to year preceding first ESAF-supported program.

Average during period commencing with first ESAF-supported program.

Interest rates at end of calendar or fiscal year preceding the first SAF- or ESAF-supported program.

Commercial bank loan and time deposit rates. The rate used is a representative rate, the midpoint of published rates, or where available, the weighted average of interest rates.

Real interest rates, given by rr = 100((1 + rn/100)/(1 + cpi/100) – 1) where rr is the real interest rate, rn is the nominal interest rate, and cpi is the CPI inflation rate.

Pre-SAF or pre-ESAF averages are for one year only.

Much progress was made in establishing positive real interest rates. Increases in real lending rates were slightly greater than those on real deposit rates in line with the objective in many SAF/ESAF programs of improving the profitability of banks. Real interest rates on time deposits rose in 16 of the 19 countries during the SAF/ESAF period, but were still negative or barely positive in 6 countries at the end of the period. Most of the latter were high-inflation countries, a fact that underscores the difficulty of establishing positive real interest rates in an inflationary environment. Real loan rates rose or were unchanged in all countries and were positive at the end of the period in all but two. In general, except in Bolivia, The Gambia, Ghana, and Sri Lanka, where interest rates were largely free of control, interest rates were, at least initially, raised administratively. Subsequently, most countries rationalized the structure of interest rates substantially, and three countries—Burundi, Kenya, Malawi-progressed to full liberalization. Even in countries that fully decontrolled interest rates, however, un-competitive banking structures often prevented proper responses to market forces.

The shift to indirect means of monetary control has progressed more slowly than interest rate liberalization. About half the 19 ESAF countries took significant steps, including changes in reserve requirements, measures to make central bank refinancing more market oriented, the introduction of central bank securities, sometimes with regular auctions, and the development of active secondary markets, interbank markets, and money markets. However, because countries have been reluctant to abandon direct monetary controls until these new systems are tested and trusted, these systems have tended to supplement rather than replace traditional means of control. The IMF has provided extensive technical assistance in this area, but further development of local expertise is needed.

The elimination of quasi-fiscal activities and improvements in banking practices was slow. Over half the countries reviewed removed quasi-fiscal portfolio restrictions on commercial banks. In some of these, however, selective credit controls and moral suasion over banks’ lending practices persist. Officials in many countries were concerned about the vulnerability of certain sectors—particularly agriculture—if preferential credit practices were eliminated.

The restructuring and strengthening of financial institutions was another area where progress was slow. In general, countries were relatively quick to identify institutions requiring restructuring or liquidation and to pass new prudential regulations. but implementation of restructuring or rescue operations was often delayed and enforcement of new regulations proved difficult. One of the important sources of delay in both connections was the continuing demand for credit on the part of weak public enterprises.

Social Dimensions of Adjustment

In view of the low per capita incomes and the inadequate availability of social services in the 19 program countries (Table 3), reform initiatives have paid increasing attention to alleviating poverty and mitigating the social costs of adjustment. The World Bank and the IMF have collaborated in this area; the Fund staff has focused on budgetary and macroeconomic implications and the Bank staff on longer term poverty reduction. The Bank has greatly increased emphasis on mitigating the social costs of adjustment and promoting sustainable poverty reduction strategies. IDA adjustment credits have increasingly specified measures to reduce social costs of adjustment and improve longer term social policies. The IMF has also made efforts to improve policy design and integrate social safety nets in Fund-supported programs.9

Table 3.Social Indicators
PopulationHealth and NutritionUrbanizationEducation
GNP per Capita 1990 (In US$)Average annual growth 1980–90 (In percent)Population aged 0–14 yrs 1990 (Percent of total)Population per physician 1984Daily calorie supply 1989Population in urban areas 1990 (Percent of total)Primary net enrollment 19891Female primary education 19892
Bangladesh2102.342.96,3902,021166378
Bolivia6302.542.51,5301,916518389
Burundi2102.845.621.0201,93265180
Gambia, The260
Ghana3903.446.820,3902,2483381
Guinea4402.546.12,132262645
Guyana330
Kenya3703.849.910,0502,1632494
Lesotho5302.743.418,6102,2992072122
Madagascar2303.045.59,7802,158256495
Malawi2003.446.711,3402,139125081
Mauritania5002.444.611,9002,6854769
Mozambique802.644.11,680274578
Niger3103.347.239.6702,308201757
Senegal7102.946.72,369384872
Sri Lanka4701.432.35,5202,2772110093
Tanzania1103.146.724.9702,206334898
Togo4103.548.18,7002,214267263
Uganda2202.548.72,15310
Average3482.845.214,6052,171265781
Memorandum items:
Low-income economies3502.035.25,8002,4063878
Middle-income economies2,2202.035.82,2502,860608990
High-income economies19,5900.619.94703,409779795
Source: World Bank, World Development Report 1992 (New York: Oxford University Press for the World Bank, 1992).

Percent of children of primary school age actually enrolled.

Females per 100 males enrolled in primary school.

Source: World Bank, World Development Report 1992 (New York: Oxford University Press for the World Bank, 1992).

Percent of children of primary school age actually enrolled.

Females per 100 males enrolled in primary school.

Increasing production and employment is the only-way to reduce poverty in low-income countries for the long term. For most of the ESAF countries—particularly those in Africa, where agriculture is the mainstay of the economy—this often means improving conditions and incentives for agricultural production. Gains in this area for many of the countries may well have been the most significant influence of ESAF programs on poverty. These gains frequently owed much to such structural reforms as increases in agricultural producer prices and reforms affecting marketing boards, which contributed to both higher production and the redistribution of income from rentiers to the poorer rural sector: such benefits were important in Bangladesh, Ghana, Guinea, Niger, Tanzania, Togo, and Uganda. Improvements in infrastructure, such as agricultural extension services, marketing, credit extension, and transportation, also supported the supply response. The lowering of inflation is also likely to have benefited the poor, who typically pay a disproportionate share of the inflation tax.

Notwithstanding the generally favorable effects on the poor of improved macroeconomic performance, adjustment often involved short-term social costs to both the poor and other segments of the population. Measures were built into programs to compensate for these. Where adjustment involved retrenchment of employment in the civil service or public enterprises, severance pay and/or retraining programs were provided in Bolivia, The Gambia, Guinea, Mauritania, Senegal, Sri Lanka, and Uganda. Where subsidies on consumer items were lifted or reduced, these were often replaced by better-targeted assistance in the form of food distribution (Bangladesh. Mozambique, Malawi), food stamps (Sri Lanka), or income supplements for the urban poor (Mozambique). In Lesotho and Kenya, the poor were exempted from increased user fees for health and education. Because weak information bases made it difficult to evaluate the social impact of adjustment, several countries began to improve the data base or analytical framework for such measures.10

Many countries also introduced measures specifically directed toward alleviating poverty. The ambitiousness of the measures varied among countries and was typically constrained by limited administrative and financial resources. Several approaches were tried: employment programs, often in labor-intensive public works, to provide work and income to the poor: increased spending on social programs, mainly in health and education with a bias toward expenditures that benefited the poor and rural areas: improvements in the economic and social infrastructure; and in a few countries, transfers to vulnerable groups, usually in the form of food or cash. The most ambitious scheme of this last sort was the Jana Saviya program in Sri Lanka: others were food grain distribution in Bangladesh, basic health and primary education programs in Bolivia, targeted assistance for the poor in Guyana, agricultural and nutritional programs in Malawi, an outreach program in Lesotho, and income supplements for the urban poor in Mozambique.

    Other Resources Citing This Publication