Policy, Research, and External Affairs t WORKING PAPERq, I Macroeconomic Adjustment and Growth Country Economics Department The World Bank October 1990 WPS 511 The Macroeconomic Underpinnings of Adjustment Lending Fred Jaspersen and Karim Shariff Macroeconomic policy and sequencing issues increasingly have been addressed explicitly in the design of recent adjustment loans, but there still is scope for: (I) strengthening the analytical framework and macroeconomic policy conditionality in adjust- ment loans, and (2) greater realism about the time and external resources needed to achieve adjustment and growth objectives. The Polcy, Research, and lxtemal Affars Complex distrbutes PRI. Aorking Papers todisseninate the findings orwork in progress and to encourage the exchange of ideas among Bank staff and all othcrs iniere-sted un development issues These papers carry the names of lhe aithors, rmflect onlh thcir sei.", and shou:d he uscd and cited accordingh! T'he finrings. inmerprciations. and corn lusions are thC authors own The. should not he autnhuted to the o'nr!d Banr. it,s Hoard of Dlircziors. iLs maragrment, or ans of ius mermber countnies Policy, Research, and External Affairs Macroeconomic Adjustment and Growth WPS 511 This paper is a product of the Macrocconomic Adjustment and Growth Division, Country Economics Department. It was prepared as part of a larger effort in PRE to assess the World Bank's experiencc with adjustment lend:ng and was a background paper for :he Bank's Report on Adjustment Lending 11: Policiesfor thze Recovery of Growth, submitted to the Bank's Board on Mlarch 6, 1990. The paper deals with the theoretical macroeconomic underpinning of adjustment programs and presents the results of empirical work on the design and implementation of Bank- supported adjustment loans. Copies of this paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact AludiaOropesa, room NII-019,extension 39075 (41 pages with tables). Drawing on conditionality and implementation policy, including public expenditure reform, has been information lor 184 World Bank adjustment loans to an important focus of the Bank's adjustment lending. 62 countries during the 1980s, Jaspersen and Shariff Rclatively Icss emphasis has been given to monctary examine the macrocconomic underpinnings of Bank- and exchange rate policies which have been a central supported adjustment programs. Thcy conclude that focus of INIF programs. Implemcntation of supply- macrocconomic policy reform and improved macro- side sectoral policies has been strongly affected by cconomic performance are critical to successful macrocconomic performance. W'herc macro balance implementation and sustainability of structural has not been reestablished, implementation rates have reform. been lower for all conditions. Where progress in climinating imbalance is taking place, rceform inertia Reducing macrocconomic imbalances is espe- has strengthened even aftcr loan disbursement has cially important for tradc rcform. If inflation is not been completed. reduced to a manageable level, therc is a danger that the cxchangc rate will be used as a nominal anchor Loan design has also becn an important determi- for domestic prices. If sustained over a long period, nant of implementation antd sustainability of reform. this may precipitate a balance of payments crisis and Where issues of sequencing have becn built into the make it impossiblc to liberalizc urade. design of adjustment programs, the implementation of sectoral reform has been stronger. Adjustment loan Reducing macrocconomic imbalances also has an conditions which are precisely defined and legally important bearing on implementation of sectoral binding for tranche releasc have had the highest rates reforms. To the cxteni that the govement cannot of implementation. For adjus:ment loans with a large reducc its fiscal deficit by increasing its own savings, numbcr of conditions, impilmentation has becn there is a greater likelihood that it will cut investment highest for the "corc" conditions that have been the in infrastructure to support sectoral rcstructuring. most critical for success of the program. Prolonged high rcal rates of intercst that result from unsuccessful stabilization efforts can hurt private After looking at recent experiencc with macro- investment and the restructuring of sectoral produc- economic con(litionality, the authors conclude that tion. If fiscal imbalances have not been eliminated macroeconomic policy and sequencing issues before liberalizing the financial systcm, it is likely increasingly have been addressed explicitly in the that the govemment will continue to rely on adminis- design of recent adjustment loans, but there still is trative controls to finance the public sector deficit, scope for: (I) strengthening the analytical framcwork undermining financial sector reform. and macroeconomic policy conditionality in adjust- ment loans, and (2) greater realism about the time and Analysis of thc Bank's adjustment loans indicates external resourLes needed to achieve adjustmcnt and that mac,occonomic conditionality has been relatively grox% di objectives. important and has increased ovcr timc. Fiscal Thc PRE Working Paper Scries disseminates thc findings of work under way in thc Bank's Policy, Rcscarch. and Extemal AffairsComplex. An objective ofthe scries ic to get these findings out quickly, even if pi escitations arc Icss than fully polished. Thc findings. interprctations, amd conclucions in thesc papers do not necessarily represent official Bank policy. Produced by the PRE Dissemination Ccnter THE MACROECONOMIC UNDERPINNINGS OF ADJUSTMENT LENDING Table of Contents Page I. INTRODUCTION: SUPPORT OF MACROECONOMIC OBJECTIVES UNDER ADJUSTMENT LENDING ........................., 1 II. THE CONCEPTUAL FRAMEWORK OF ADJUSTMENT: PROGRAM DESIGN - A TAXONOMY OF ADJUSTMENT POLICIES. 3 Stylized Facts. 3 Reduction of Aggregate Demand. 4 Change in Relative Prices. 6 Increase in Output............................................. 6 Debt Reduction........... : 6 Supply-Side Structural Reforms and Efficiency ............ 7 Adjustment Program Design ............ 9 III. IMPLEMENTATION OF MA :.OECONOMIC CONDITIONALITY IN BANK-SUPPORTED ADJUSTMENT PROGRAMS: 1980-89 ............ 9 Background ............ 10 Nature of Bank Conditionality and Implementation ............ 10 Macroeconomic Conditionality in Program Design ............ 11 Implementation of Macroeconomic Conditionality ............ 12 IMF Programs and Bank Macroeconomic Conditionality ............ 15 Inflation and Implementation of Macroeconomic Conditionality... 19 Economic Growth ............ 20 IV. FY89 ADJUSTMENT LENDING ............ 20 Background ............ 20 The Macroeconomic Underpinnings of FY89 Adjustment Lending ..... 26 Initial Conditions ............... ......................... 26 Preconditions for Adjustment Lending ..................... 27 Macroeconomic Policy in Program Design ................... 31 Conclusions and Issues Identified in the Case Studies .... 34 APPENDIX ................................................................ 40 Coding, Conditionality and Implementation .............. ............. 40 This report was prepared as a background paper for RAL2 by Fred Jaspersen and Karim Shariff. Jaume Ventura provided useful comments and Fernando Garcia provided valuable research assistance. Section III of this paper is based on the data base developed and maintained by the Industry Development Division in PRE. William Steel, Debbie Bateman, Diane Cougan and Mala Hettighe were especially helpful in providing this information. THE MACROECONOMIC UNDERPINNINGS OF ADJUSTMENT LENDING I. Introduction% Support of Macroeconomic Objectives Under Adjustment Lending (AL) One feature of quick-disbursing adjustment lending which distinguishes it from investment lending, which disburses more slowly, is its greater capacity to transfer hard currency resources. Adjustment loans maximize the present value of net transfets from the Bank to the recipient country for a given level of commitment. For this reason adjustment lending is the most appropriate instrument for assisting a country which is experiencing balance of payments problems. In the absence of external imbalanct or if such imbalance is the consequence of a transitory terms of trade shock which does not require a re-alignment of macroeconomic policies (i.e., the shock is short-lived and can be covered out of reserves or by short-term, compensatory borrowing), there is no strong rationale for the Bank to employ adjustment lending rather than investment lending since the latter is generally more effective where macroeconomic objectives are not of high priority and where shorter-term compensatory financing is available from other sources. If the external shock is permanent, or the balance of payments problem is the consequence of internal imbalance stemming from mismanagement of domestic macroeconomic policies in the first place, and therL is a need to reorient such policies, adjustment lending may be more appropriate. It is for this reason that the Bank has emphasized that given the quick- disbursing nature of adjustment lending, it is important that there be "a clear and documented understanding with the government on its overall structural adjustment program, including short-term stabilization as wel'l as longer-term development objectives'.1 To ensure this, the Bank has recommended in the two recent reports on adjustment lending that a more explicit and monitorable approach be taken for Board presentation and subsequent tranche release. In particular, "both should be dependent on a documented management judgment that appropriate action was being taken to reduce external and internal financial instability.3 This judgment, it was noted, "should take into account evidence of actual performance as indicated by available short-term economic and financial indicators and also by the progress recently made in adopting relevant reforms indicated in the general understanding on the country's adjustment program. The judgment and supporting evidence would be included in the President's Report for proposed adjustment loans and in the notification to the Board of all subsequent tranche releases." The reports on adjustment lending have also noted the circumstances which would tend to increase the likelihood that a government's program be implemented over the medium-term. One such circumstance is that "the government publicly propose and justify the overall strategy of the adjustment program." Without this, it was agreed, the risk of weak implementation and subsequent reversal would be high. 1 This approach is to be monitored for 18 months starting in September 1988, after which management would reassess its adequacy with respect to improving performance. - 2 - Similarly, it was emphasized that adjustment programs need to be realistic. They hav to be "restrictive enough to be consistent with the financing available, but not so restrictive that it is likely to prove socially and politically unacceptable and therefore unsustainable." This is especially important for countries with a large external debt overhang. Greater realism was recommended in two additional areas: fuller recognition of (a) the effects of the global economic environment; and (b) the time that might be needed for supply responses. While improved economic performance over the medium-term is clearly a major objective of adjustment lending, unexpected external and internal shocks can adversely affect performance. There are also response lags following policy reform which are ex ante of uncertain duration that will determine the timing of the impact on actual economic performance. It is for these two reasons that policy (and not performance) conditionality is called for. Only over the medium-term would the Bank look for and require as a condition an improvement in performance. In a series of adjustment operations, the policy reform agenda would of course be adjusted in response to economic performance and progress in achieving program objectives. The most recent report on adjustment lending concluded that the relative macroeconomic performance of AL countries was better than that of those which had not received such loans despite the fact that the former suffered greater external shocks. The exception to this general finding was the Sub-Saharan Africa (SSA) and highly-indebted countries (HICs), where improvement in relative performance between adjustment lending recipients and non-adjustment lending countries is weak. It was concluded that while the SSAs and the HICs had been successful in reducing their external current account deficits after adjustment lending had been initiated, internal imbalances were greater and there was no significant progress toward resumption of adequate growth (in the absence of adjustment programs, of course, this situation is likely to have been much worse). Changes, not only in the design of adjustment programs and their implementation, but also in the existing debt strategy were recommended. Concerns have been expressed at Board discussions in recent years over Bank initiatives in the macroeconomic area which might duplicate Fund activities. Staff have clarified that because of the requirements of structural adjustment, the Bank has, of necessity, become increasingly concerned with a wide range of macroeconomic issues. Because the success of its adjustment lending will, to a considerable extent, be determined by the existence of a stable macroeconomic environment, the Bank must evalvate the macroeconomic policy framework within which such lending is taking place. Moreover, the Bank as a market-based institution must be concerned with the creditworthiness of IBRD borrowers and must assess their macroeconomic management. The recent agreement between the Bank and the Fund acknowledges explicitly the importance of these issues to the Bank (see paragraph 10, R89- 45 of March 31, 1989). -3- This paper assesses the Bank's experience supporting improved macroeconomic management through adjustment lending. It attempts to draw conclusions about the macroeconomic framework for such lending and about the design and implementation of macroeconomic conditionality in adjustment lending. Section II presents a conceptual framework for macroeconomic policy under adjustment lending which links stabilization, adjustment and, where relevant, debt reduction. It describes the linkages between macroeconomic policies, i.e., expenditure-reducing policies, expenditure-switching policies, and supply-side policies, and the sequencing of reforms at different stages of the adjustment process. It concludes by placing the macroeconomic obiectives of adjustment within the context of a taxonomy of program design for a typical adjustment operation. Section III focuses on the Bank's experience with the design aad implementation of macroeconomic conditionality in 184 adjustment loans to 62 AL countries during the past decade. It identifies some of the key operational issues concerning the macroeconoiic framework for these loans and draws conclusions about how the Bank can more effectively support improved design and implementation of this framework. Section IV examines for macroeconomic content a sample of 23 adjustment operations carried out by the Bank in FY89, the most recent year for which data is available. It identifies those FY89 operations which illustrate "best practice" Bank-wide in supporting improved macroeconomic policy management. It concludes with a summing up of the major issues faced by the Bank in such operations. II. The Conceptual Framework of Adjustment: Program Design - A Taxonomy of Adl"stment Policies Stylized Facts During the 1970s and early 1980s many of the countries which became candidates for Bank adjustment lending were pursuing expansionary demand management policies. Such countries typically had large fiscal deficits, rapid monetary expansion, accelerating inflation, unsustainable current account deficits and a large stock of external debt. In a number of these the cumulative effects of weak institutions and policy-induced distortions in factor and goods markets had created rigidities in these countries' economies reducing their capacity to adjust to external shocks. Distortion-inducing policies included price controls, over-valued exchange rates, administrative controls on interest rates which resulted in negative real rates of interest, and large fiscal deficits. Fiscal defic'ts were in many cases the consequence of external shocks, but also of unsustainable high levels of government spending, inelastic tax systems, direct subsidies to state enterprises, indirect subsidies to consumers through reduced prices of goods and services provided by the public sector, and of high interest payments on external and internal debt. -4- D'stortions created by these policies were frequently compounded by a trade regime which relied heavil- on quantitative restrictions to protect domestic industry and by adminis-rative controls at the sectoral level. Distortions in the financial sector, frequently the consequence of macroeconomic imbalances, were especially serious because of their far- reaching adverse impact on the allocation of resources throughout the economy. Figure 1 shows a typical country where adjustment is required. Policy-induced distortions result in the economy operating below the production possibility frontier (PPF). The level of expenditure (E) exceeds current production (Y). The result is an "ex ante" excess demand for both tradables (T) and nontradables (N). The former results in a current account deficit equal to T Tc. The latter gives vise to an increase in the price of nontradables (Pn) Pn absolute terms and relative to prices of tradables which are determined in world markets. The rise in the price of nontradables will lead to a real appreciation of the exchange rate. If the resulting current account deficit can be financed -- either by borrowing abroad or by reducing reserves, the real appreciat_on will stick, switching expenditure from nontradables to tradables and switching production from tradables to nontradables. The consequence is that the excess demand for nontradables will be corrected at the expense of a larger current account deficit. There will be no inflation, only a once-for-all rise in the price of nontradables. If external funds are not available, the nominal exchange rate will depreciate offsetting the increase in the price of nontradables and thereby preventing the real appreciation. In this second case, as long as the excess demand persists, a process of continuous devaluation and inflation will take place. This is the current situation in many of the countries that are now receiving ALs. For a country in this situation adjustment involves two actions. On the one hand, aligning aggregate demand and aggregate supply, that is, stabilizing the economy. On the other hand, bringing about structural changes that increase the level of efficiency and flexibility of the economy. Typically, a structural adjustment program has three objectives: (a) a reduction of aggregate demand; (b) a change in relative prices and (c) an increase in output. Reduction of Aggregate Demand Although in the medium-term the objective is to "grow" out of the crisis, it is not usually possible to raise production enough to correct the initial eixcess demand. The removal of distortions may even have a negative initial impact on income if rigidities have built-up in the economy. Therefore, the major emphasis in the short run will fall on reducing expenditure. Adjustment lending facilitates this in two ways. First, adjustment programs become more politically acceptable because expenditures will not have to be cut as much as would be the case in the absence of such lending. Second, acceptable investment levels can be achieved during the adjustment period. -5- E*cr' ut; I;t aC/ \ PPF \ p~ IC Tr~~r ~*IS Ptiwav f;cg -6- To succeed, expenditure reduction must be selective. Too large a reduction in investment jeopardizes the prospects for future growth. Too large a re!i^tion in private or public consumption imposes an unacceptable social cost anC may be politically difficult to implement. A careful balance among the different objectives has to be achieved. The conventional tools for managing aggregate demand are fiscal and monetary policy. The fiscal deficit found in most AL countries is the major element explaining the initial excess of demand which made the program necessary. A reduction in this deficit is usually central to an adjustment program. If this is achieved, it is possible to reduce the rate of growth of monetary aggregates and bring aggregate demand more in line with aggregate supply. Tight monetary policies must not be overdone, however, since excessively high real interest rates are incompatible with resumption of private investment and growth. Change in Relative Prices To complement aggregate demand reduction, policies aimed at bringing about a change in relative prices will help to restore internal balance and moderate the decline in expenditure. By increasing the relative prices of tradable gcLds to non-t.adable goods through a real depreciation of the currency, resources are reallocated or "switched' from the non-tradable goods sector, where there is excess supply, to the tradable goods sector, where they can produce for export or substitute imports. However, prior to the devaluation, it is important to look at the impact that other measures of the adjustment package will have on relative prices. Only after this is done can the amount of depreciation of the nominal exchange rate be determined. Changes in relative prices should be as smooth as possible. Overshooting of the nominal exchange rate (as well as undershooting) should be avoided. Increase in Output To the extent that output can be increased by exporting exports and increasing production of import substitutes, the decline in expenditure can be less, or perhaps avoided altogether. There are two ways of doing this: First, by using more efficiently the existing stock of factor inputs by removing distortions. Reforms directed at removing these distortions have been a main concern of the Bank's ALs. Second, by increasing the current stocks of factors of production, i.e., the attainment of a reasonable rate of capital accumulation. A crucial objective of the AL programs is to ensure that an investment level is attained which is compatible with &.b acceptable rate of growth over the medium-term. Debt Reduction Efforts to reduce macroeconomic imbalance have been complicated for countries with a large external debt. For such countries closing the gap between aggregate demand and supply has required a large fall in real expenditure despite efficiency improvements. Because external debt in most of the high debt countries is in large part public sector debt, servicing this debt has aggravated existing fiscal weakness adding to macroeconomic -7- imbalance. For those high debt countries which have lost access to foreign capital markets, there is a need to manage their budgets under tight external financing constraints. Excessive reliance on domestic borrowing haQ in many of these countries caused real interest rates to rise thereby .rowdxng out private investment. Where governments have relied more on money creation, this has resulted in increased inflatiorn and, in the absence of increased private financial savings, caused foreign exchange reserves to fall. High levels of external debt service hav': worsened existing rigidities on the expenditure side (frequently the consequence of a bloated public sector wage bV.l), thereby further reducing the government's room for maneuver to reduce the fiscal deficit. The consequence is that public sector investment has generally born the brunt of the adjustment. The deterioration of inrrastructure and public sector service which has resulted, has had an adverse impact on output further complicating the adjustment process. Viewed from another perspective, interest payments abroad on external debt have reduced national saving3, which in the absence of foreign savings have been a binding constraint on investment and growth. For countries in this situation it becomes difficult to restore macroeconomic balance and sustainable economic growth while servicing their external debt. Debt reduction complements the adjustment policies described in the previous section. Supply-Side Structural Reforms and Efficiency The typical adjustment lending country faces policy-induced distortions which limit factor productivity, reduce domestic and foreign -ompetition, and prevent goods prices from reflecting their opportunity cost. Such distortions are generally the result of weak institutions and inappropriate policies. Supply side structural reforms aimed at increasing efficiency and reestablishing adeqcate growth are focused on strengthening incentives and reducing constraints on factor mobility. Empirical work carried out for this paper suggests that such reforms are unlikely to be as successful in countries experiencing large macroeconomic imbalances and financial instability. Success in eliminating macroeconomic imbalances seems to be an especially important determinant of tb- 3urvivability of trade liberalization. If stabilization efforts are unsuccesErul in reducing inflation, the Government may, out of desperation, use the nominal exchange rate as an anchor for domestic prices2. In such cases tle real exchange rate will appreciate rapidly if domestic prices continue to rise. If this persists. it may eventually lead to a balance of payments crisis which ultimately undermines liberalization of the trade regime. Where governments are unsuccessful in stabilizing the domestic economy, the real exchange rate is allowed to appreciatf and inflation continues at relatively high rates, the real interest rate is likely to rise to excessively high levels. This may reduce investment 2 In 21 stabilization episodes of AL countries the real exchange rate appreciated in 10 of them (see Jim Nash, Trade Policy in Adjustment Lending, 1989, background paper for RAL2). -8- and prevent the restructuring of the economy in response to trade policy reform which is necessary if the liberalization is to be successful. Finally, if the government is not successful in reducing the fiscal deficit, sustainability and deepening of trade reform m&y be jeopardized since there will be pressure to overturn tariff liberalization which results in a loss of import tax revenue3. Sectoral reforms also depend heavily on reduction of macroeconomic imbalances. The more unsuccessful the government's fiscal deficit reduction effort, the greater the likelihood that public investment in infrastructure which complements sectoral restructuring will be cut. Macroeconomic instability is generally associated with an appreciating or volatile real exchange rate. When the real exchange rate discriminates against tradables or is unstable, sectoral output, profitability and reforms are likely to be adversely affected. Excessively high real rates of interest which are generally associated with unsuccessful stabilization efforts or programs which lack credibility, can have serious a;'verse consequences for restructuring of sectoral production and ultimately on the sustainability of sectoral reform. Financial sector reform has also been difficult to carry out in countries experiencing acute macroeconomic imbalance. While macroeconomic instability may have underscored the shortcomings of the financial sector, thereby helping to focus the government's attention on the need for financial sector reform, the high real interest rates which have generally followed liberalization, have generated financial distress among borrowers which in turn has eroded the financial viability of financial institutions leading to a roll-back of liberalization efforts. Protracted excessively high real rates of interest have alsc. complicated macroeconomic management in the short-term by increasing the interest burden of government's domestic borrowing and in the medium-term by requiring an infusion of government resources to shore up the financial system. These adverse effects of high real rates of interest on macroeconomic management and on the financial system itself has led the Bank to place more emphasis on rationalization rather than on decontrol of interest rates. Efforts to liberalize the financial system before the government succeeds in stabilizing the domestic economy are unlikely to be successful for other reasons. Where the government has failed to bring an underlying fiscal problem under control for a prolonged period, the authorities have generally channeled resources from the financial system to the Government through direct controls which have undermined the viability of the financial system. If the fiscal imbalance has not been eliminated priz to liberalization of the financial system, it is likely that the government will continue to rely on administrative controls to finance the public sector deficit thereby undermining financial sector reform. If the financial system is dominated by government-owned banks and the government controls interest rates directly through its shareholder power In the commercial banks, there will be a further undesirable reduction in the monetary policy role of the Central Bank in favor 3 Jim Nash, Trade Policy in Adjustment Lending, 1989. -9- of the Ministry of Finance. The clear lesson of experience is that far- reaching financial sector liberalization can survive only within the context of macroeconomic stability. Adiustment Program Design The following taxonomy classifies adjustment policies according to their principal objective. This classification is used in the next section to assess the relative importance and implementation performance of macroeconomi. adjustment policies. Adjustment Program Design I. Absorption Reduction Policies Fiscal Policy (deficit reduction targets) Monetary Policy (money supply targets) II. Expenditure-Switching Policies Exchange Rate Policies Wage Policy III. Supply-Side-Growth Oriented Policies Trade Policies Sectoral Policies Industry Energy Agriculture Financial Sector Reforms Rationalization of Government Finance & Administration Public Enterprise Reforms Social Policy Measures III. Implementation of Macroeconomic Conditionality In Bank-Supported Adiustment Programss 1980-89 This section analyzes the Bank's experience with implementation of macroeconomic conditionality in Bank-supported structural adjustment programs. It focuses on the design and implementation of such conditionality in 184 adjustment loans to 62 countries during 1980-894. Specifically, it considers the macroeconomic prerequisites for structural reforms, emphasizing the importance of an appropriate sequence and mix of macroeconomic and supply- side policies to achieve internal balance, external balance and growth- oriented adjustment. 4 For a discussion ct the methodology see the Appendix. - 10 - Background The Bank began -9s lending for structural adjustment in 1980. Although at first expected to be short-lived, adjustment lending increased to 312 5f the Bank's total lending commitments in FY89. The scope of adjustment lending aiso widened with the introduction of sectoral adjustment loans (SECALs) which account for almost all of the increase in adjustment lending since FY82. Increasingly, Structural Adjustment Loans (SALs), which contain comprehensive macroeconomic policy conditionality, gave way to more narrowly focused SECALs; increasingly SECALs themselves have included macroeconomic conditionality. In FY89, the Bank undertook 35 adjustment-lending operations, of which 27 were SECALs. Since 1980, 62 countries have received adjustment loans from the Bank. Because adjustm-nt is a process which in most countries takes a number of years, and because subsequent stages of the process cannot be initiated until the objectives of earlier stages have been substantially achieved, a cointry will generally need a series of operations, each supporting specific policy actions. Nature of Bank Conditionality and Implementation A typical adjustment lodn disburses in two tranches. Some conditions are required for effectiveness, some for second tranche release, and some are included as special covenants which are not binding for tranche release. Conditions of disbursement relate to policy instruments that target specific objectives. For example, expenditure and production switching policies encourage both exports and efficient import substitution by raising the domestic price of tradables relative to the price of non-tradables. These policies taken together are designed to reduce macroeconomic imbalance, to improve efficiency, to strengthen institutions and to mobilize additional resources for investment. While conditionality had initially been based on the premise that policies which reduce macroeconomic imbalances and price distortions will stimulate savings, contribute to increased investment, and accelerate growth, the weak growth performance of adjustment-lending countries has heightened awareness that other measures are also needed. These include institutional reforms, programs aimed at reducing the uncertainty associated with the adjustment which private investors face, specific investments aimed at alleviating real constraints in the economy, &nd where necessary, debt reduction. Each AL with its broad mix of policies is placed within the context of a country's own reform program, which itself is broader and generally includes some elements supported by the Fund. There is a high degree of variability among conditions with regard to their specificity. The Bank has moved to using more specific conditions, partly guided by experience. More specific conditions have higher implementation rates. The average number of actions has almost doubled from about 30 in FY79-85 to 56 in FY89. About one-third of all actions are legally required and are conditions which appear in loan agreements. The remainder - 11 - are listed only in the President's Reports. While more than a quarter of all policy actions are precisely specified -- only 102 of the actions are precise and included in the loan agreement, an average of about four conditions per loan. The data on implementation which is presented in this section indicates the extent to which a condition was fulfilled during the lifetime of a loan. Of the 184 loan sample, only 97 loans have been in place long enough to make it possible to assess the experience with implementation. The sources of information on implementation were mainly Supervision Reports, Tranche Release Documents, and where available, Project Completion Reports and Program Performance Audit Reports. These sources were supplemented by additional information obtained from Bank staff familiar with the country's adjustment program. While the process of coding conditionality is straightforward, the process of coding implementation is more complex. Progress in implementing each item of conditionality during the period of the loan was classified as: (a) little or no progress; (b) partial progress; (c) substantial progress (substantial steps taken to move progress more than half-way); (d) implementation fulfilled; and (e) implementation more than fulfilled. Macroeconomic Conditionality in Program Design Because of the importance of complementary reforms in different policy areas, recent ALs address a wide variety of policy areas as they did when SALs were dominant. Macroeconomic conditionality defined as demand management or absorption reduction and switching policies has been relatively important in adjustment loans. Table 3.1 indicates that for all adjustment loans which the Bank has made since 1980, fiscal policy conditions were present in 51X of all loan agreements. This has increased over time for the early intensive adjustment lending countries (EIAL)5 from 47Z in FY79-85 to 532 in FY89. For the other adjustment lending countries, fiscal policy conditionality has risen from 332 to 582. Such conditionality has been especially important in the Sub-Saharan African Countries (692) where budgetary problems have been acute. Fiscal reforms included measures to reduce the deficit, to enhance the efficiency of raising revenue, and to increase the effectiveness of government expenditure. Reductions in deficit have been at the core of successful AL operations. Relatively less attention has been given to monetary, exchange rate and wage policies, although the importance of these too has been rising over time. Monetary and exchange rate policies have been a central focus of IMF programs. When the Bank and Fund have concurrent programs with a country, a frequent occurrence, these policies are usually left to the Fund. In those cases where there is a concurrent Fund program and exchange rate conditions are also included in the Bank loan, it may be because the Fund has different exchange rate objectives than the Bank. 5 Defined as countries that have received two SALs or three adjustment operations or more starting in 1985 or before. - 12 - Implementation of Macroeconomic Condit onality More than half (66?) of all conditions in adjustment programs are judged by Bank staff to be implemented fully or more than fully by the time of final tranche release. This rate increases to 84? if conditions on which progress has been substantial are also included. Loans since 1985 have higher average rates of implementation than those made earlier. Table 3.2 shows that performance in implementing conditionality varies widely across types of policies. For the five or six critical actions in the program (coded on the basis of the President's Report), the implementation rates are as good on average as for all conditions in the legal agreement. Implementation of macroeconomic policy conditionality is above the average for all conditions (e.g., 74? of fiscal policy is fully implemented; 82? substantially implemented; 752 of exchange rate policy fully implemented; 85Z substantially implemented). This may in part be explained by the additional leverage in support of macroeconomic policy reform from concurrent IMF programs. Implementation rates of supply-side policy categories have, as noted, been strongly affected by macroeconomic performance. In some instances supply-side policies have been incorporated into adjustment lending conditionality during early stages of the country's adjustment. In a number of cases this has been done before internal or external balance has been reestablished. In these circumstances implementation has been lower than the average for all conditions. Implementation rates for public enterprise reform have closely paralleled progress in fiscal policy management. Such reforms have been implemented more consistently where fiscal deficit reduction has been an important objective of macroeconomic policy. More than one-third of recent AL contain PE reform measures. Such reforms have made it possible for state enterprises to reduce their drain on the central government budget by cutting their losses through restructuring aimed at increasing efficiency and lowering costs as well ds bringing the price of goods and services which they produce more in line with long-term marginal costs. Some countries have made significant progress in divesting public enterprises. In others, it has been difficult to deal with financially-ailing entities: closing them down has meant increased unemployment. Without prior rehabilitation, they are not attractive to potential buyers. In addition, domestic capital markets are frequently unable to support the purchase of such enterprises. The temptation is therefore for governments to continue to run these entities. When the circumstances warrant continued public ownership, reform measures have aimed at improving efficiency and accordingly profitability. Reforming management and allowing competition from private and foreign firms helps to raise efficiency. This was the aim of the 1989 Public Enterprise Reform Loan (PERL) to Mexico, for example. While the incentive of tranche release motivated policy adjustment, the perceived benefits of adjustment and the desire to have ongoing AL also motivated reform. Indeed, after the final tranche release countries usually sustained or increased the rate of fulfillment. Of the conditions that were not at least substantially fulfilled, about one-third are substantially or completely fulfilled within a year. The tendency for reform inertia to Table 3.1: Importance of Macroeconmic Conditionellty in Adjusteat Loeding Operations Share of Losae with loan-agr.en_t coaditions or actions io various policy *reas (percent) Other Share of (2) All EIAL Countries AL Ceuntri-e loans W/ actions Contriee SSA HIs SAL SECAL Hybrid 79-5 S 1*-98 89 79-86 8-9 *cti.n to various (13U) (34) (64) (73) (110) (10) (5t) (49) (16) (9) (Cb) pelicg areas 113) Absorellto redction oulicdes Fiecal policy 51 69 41 78 84 30 47 51 63 83 so 67 Menstry policy (mosyI Sppy Targets" 1s 14 is 14 i6 10 7 16 13 11 24 42 II Slitching policies Ezcbanvg rtt 10 16 20 22 13 0 9 1t 20 11 22 46 sgs Pleicy 1i 2 S 26 0 20 4 6 7 11 29 22 III Su.nlw-aide erowth oriented selicies Trad pellet _6 5a 67 64 55 90 60 09 33 60 U5 79 Sctoral polici e lIod.try 22 s0 16 26 20 10 24 20 27 22 20 44 Eergy 1i 12 14 21 11 so 1s 14 7 22 16 27 Agrleiltaral 46 62 so 50 37 90 44 3U 27 U Ss 02 Fina_lcl sector U 20 31 40 26 20 16 5 27 44 40 51 3.tbimllgatleo of Govt fls _ a& *daloletratio 51 U7 60 71 U 10 U1 6U 40 44 66 72 Pubile EaterprIss reforms 44 5 34 49 40 40 21 49 U as b6 O Scial policy r0frmn 11 18 9 11 11 10 I 20 0 22 11 24 Other 23 42 17 n s2 10 7 27 20 a3 C1 49 Notes: 1. Nmbbre In brckekt ( ) are total number of loans. 2. All countrie. All conditioens celled for In *11 loan agreementa or other ctions called for In all President Rports. Source: eed on ranalysie ot 138 SAL and SECALa to 61 developing countries. - 14 - Table 3.2: Implementation of Adjustment Lending Conditionality In *11 loan Agresements 1/ Critical Actions 2/ Fully At least Fully At least Imple- Substantially Imple- Substantially mented Implemented mented Implemented I Absorption reduction policies Fiscal policy 74 82 72 89 Monetary policy [Money Supply Targets] 67 83 61 89 II Switching policies Exchange rate 75 8S 71 81 Wage policy 45 91 50 50 III Supply-side arowth oriented policies Trade policies 62 85 56 82 Sectoral policies Industry 72 92 53 65 Energy 69 80 72 80 Agricultural 62 81 49 74 Financial sector 73 89 79 92 Rationalization of Gov't finance & administration 63 81 54 68 Public Enterprise reforms 66 80 67 77 Social policy reforms 59 91 55 82 Total Loan agreement conditions 3/ 66 84 67 83 All conditions or actions 4/ 57 77 60 79 Notes: / The date on implementation Indicate the extent to whlch a condition or action was fulfilled at final trench, release. A total of 1,015 legal conditlons were graded on implementation. V Critical actiono are so Identflied because Bank staff deslinino the operation put particular emphasis on thm an becauso thoy were expected to ake a slgnificant contribution to adjustment In a short time. A total of 194 actions were coded as critical, of which 808 appeared as conditions in the loan agreements. / The implementation rate of conditions that appear In the loan agr_ment. / Averago Implementation of actions which apper in the President's Report or conditions in the loan agreemnt. A total of 2,281 actions were graded In Implementation. Source: osed on en analysis of 97 SALs end SECALs In 82 developing countries. The sourcoa of informtion on imploemntation were mainly supervision reports, tranche release docuewnts, and where available, Project Completion Reports and Program Portoremnce Audit Reports. - 15 _ continue after final tranche release is especially true for some key macroeconomic policies. Table 3.3 provides information regarding the sustainability of reforms once loan disbursement has been completed. It indicates that reforms initiated in conjunction with Bank-supported adjustment prQgrams generally continue after loan disbursement. Sustained improvement in implemeatation has also been the case for fiscal policy. As borrowers have taken steps to carry out administrative reforms aimed at strengthening public sector management generally and have gained greater control over public enterprises, expenditure policy planning, and the budgetary process, fiscal policy objectives have been easier to achieve. It may be unrealistic, however, to view institutional change as monotonic or linear since the likelihood of sustained reform is linked to government commitment which itself varies with the general state of the economy, with political constraints, and with the degree to which leaders agree with the substance of reform measures. Table 3.4 indicates that loan design is an important determinant of implementation. Actions which are only listed in President's Reports have relatively low rates of implementation (only 49Z are fully implemented). Actions which are precisely defined have higher rates of implementation even though they may not be legally binding. Conditions which are both precisely defined and legally binding have the highest rates of implementation (71% are fully implemented). This suggests that the greater the specificity of conditionality and the more legally binding the more likely it is that it will be successfully implemented. This is particularly true of SALs, where the average rate of implementation for actions listed only in the President's Reports is 472 while the rate on precise and legally binding conditions is 73Z. Greater detail may require a larger number of conditions. General conditions that summarize the spirit of the adjustment effort are also important. IMF Programs and Bank Macroeconomic Conditionality The Bank consults regularly with the International Monetary Fund. The Bank has generally refrained from adjustment lending if a cou.atry with macroeconomic instability does not have a Fund Program -- a Standby Agreement, an Extended Fund Facility, a SAF or an ESAF. But the Bank has made its own macroeconomic assessments. Over four-fifths of SALs and over three-fifths of SECALs have Fund programs by the time of effectiveness. In about 152 of these cases the Fund programs are subsequently cancelled6. The macroeconomic content of the Bank's adjustment lending is heavily influenced by the commitments which the borrowing country has entered into with the International Monetary Fund. Table 3.5 indicates the conditionality content of the Bank's adjustment loans where there is or is not a concurrent IMF program and the rates of implementation for adjustment programs with and without Fund programs. In those cases where the IMF does not have a program, Bank conditionality emphasizes areas of macroeconomic policy (exchange rate and monetary supply targets) which are areas of primary interest to the Fund. 6 Half of these are in turn revised probably with less strict requirements. - 16 - Table 3.3: Implementation of Conditions prior to and after Tranche Release Percentage of Conditions During the Loan Period 1/ Current Situation 2/ Fully At least Fully At least Imple- Substantially Imple- Substantially mented Implemented mented Implemented I Absorption reduction policies Fiscal policy 73 78 68 82 II Switching policies Exchange rate 78 78 60 90 III Supply-side growth oriented policies Trade policies 60 85 67 88 Sectoral policies Industry 57 86 73 88 Energy 71 79 92 92 Agricultural 65 85 69 86 Financial sector 78 88 72 95 Rationalization of Gov't finance & administration 56 77 67 90 Public Enterprise reforms 69 75 76 85 Total Loan agreement conditions 3/ 66 83 70 88 All conditions or actions 56 76 62 81 Notes: n / At final trencho releoa. 3V At least 12 months after final trench, release. !/ Based on conditons which apper only In the loan agreements. A total of e9s conditions were monitored. Source: Based on a subset of 97 SAL. and SECALS to 82 countr10s. - 17 - tabl .41 Spl-mentatien of condItIne (Average percent et final tronch. release) All conditions or actlonal Lean Agree_ents2 Precise' Fully At least Fully At least Fully At least implemented Substantially S plemnted Substantially Implemented Substantially Imple mnted pl lssted mpl mnted All 57 77 66 64 62 73 HICe 01 *2 72 s9 71 6t SSA 55 75 60 s0 56 71 SAL. 5s 73 645 ea s 74 SECAL gO sO so 84 07 J2 EIAL countries fY79-J5 S3 71 6O sO 60 75 FYU -9S gO 02 Go SO 62 79 Other AL countries fY79-S6 as O7 as 74 43 Is fYP6-Sn 70 94 '6 93 63 93 Actions only In Proc e and L og14 Proold nt @ Roports' Fully At loast Fully At loest implemented Substantially Implemented Substantially Iple*ented m pl mented All 71 so 49 72 mCie 7" 9'3 s 74 SSA 04 62 sO 70 SALs 73 67 47 69 SECALs 70 6O Sp 70 EIAL countries FY79-O6 71 e6 47 66 FYJ4-JJ 69 es 64 76 Other AL countries PY79-65 40 92 3O 59 fYJO-68 93 97 31 9S Source; Based on an analysis of 97 loans to 32 developing countries. All conditions called for In the loan agrement or other actions called for In the President's Report, except prior actions, arm not lneluded because thoy aro said to be fulfilled Opriors to negotiation. V, All conditions In th, loan agreement. 3/ Preelso conditions or actions are specific and usually quantitiable. O Precie and legal conditions are those which appear In the Loan Agreement and are specific In nature. S/ Actions which appear only In tho President's Report (excluding prior actions). 4w3n pa owl 1 * ..... - ua .-P 41s.m-- -GeftI,2se.i A - *.ie PA" aqi is *1d .1 00S Paf ee s i, S#[."ee z & *if, 01M) pa eW *I ** 1le1l : fgeAjrna - pft *..i AseWp- .'-$ a 9.4 a mg -e low n~~~~~~9 Is u- s~ am on a I Is nU 5111*8 131 u es 4u n 161 $Msa nmum , Us at It LIt i na8assu us n . MOa_ U U9 9 U ft 00 29~~~~~~m to ii, SI it iA.g 1 u is A.sup m t ,9 s ~ I n, maam& SlUlusi mami* al8013esu en ma ma u G b t'4 flmNM U2IMi mau mm .~ 3* II IA A'f AMI nn du 09 is "P P s s elINt 61i5 mmasin u n ,6j5uaS p s At' "w^' "a '" *fe so A,V ,,1 Al pyF e iwAjj Po ".PS3 i,n0.4 paSj t.w.a C yw.b-3~-" IC gw~ _* 1 -u u _ - - - ,,, - ~ ~~~~~~~~ ......... .......... . ~14.4~jd.s a-$W1d ONfli IN3N N03 HIIA SNOIIV83dO SNION31 JIM 311 _tjpa t*lAo *SC A.1 1.d W*s5 Us .a,am so -14e._l,to ,. "S., , 6,A.4 -A C %ao. m t - 19 - Interestingly, the data indicates that the share of loans with fiscal policyconditionality is greater where there is a concurrent Fund program. This is in part a reflection of the nature of the lending instrument. Most SALs have Fund programs in place. The Fund has traditionally relied on the Bank to carry out a review of public sector investment programs where it has aprogram. It is likely that this complementarity between the Fund and the Bank results in greater involvement of the Bank in issues of public finance where there is a Fund program. Supply side policy conditionality is also significantly affected by the presence of a Fund program. Rationalization of Governmert Finance and Administration and Public Enterprise Reforms receive more emphasis in Bank adjustment loans which are concurrent with Fund programs than those which are not. These areas receive greater emphasis because of the greater attention to fiscal policy in such loans. Financial sector polcies are given less emphasis where there are Fund programs primarily because banking system policies and broader financial sector policies are a natural extension of the Fund's primary area of focus. On average, though, there are insignificant differences in rates of implementation of Bank program conditionality with and without Fund programs. Inflation and Implementation of Macroeconomic Conditionality Sometimes the macroeconomic environment deteriorates during the course of the program. For example, the Argentine Export Development AL of FY87 was undertaken although inflation exceeded 1002 annually during the first four quarters following effectiveness. In total, seven SECALs and one SAL were undertaken despite an inflation rate exceeding 1002 annually during the first four quarters after effectiveness. Implementation of adjustment policy is influenced to a great extent by the macroeconomic environment. The results of the empirical work prese :ed in this section confirm that a supportive macroeconomic environment i one of the most important determinants of implementation in adjustment programs. Based on the average share of conditions implemented per loan, Table 3.6 indicates that implementation was significantly lower the more severe the internal macroeconomic imbalance (i.e., the higher the rate of inflation). While the presence of a concurrent Fund program is on average corre,lated with lower rates of inflation, the result is not statistically significant at tne 5% level. Table 3.7 indicates the implementation rates for different types of adjustment policies according to whether the adjusting country has low, moderate or high rates of inflation. It takes a longer view of the inflationary environment and the uncertainty associated with greater price instability. It indicates that macroeconomic policies are more apt to be implemented where there is low or moderate inflation. Policies aimed at changing real relative prices (switching policies) tend to be more difficult to implement, the higher the rate of inflation. Exchange rate policies (real exchange rate targets) were the most adversely affected by high rates of inflation. For this type of conditionality implementation rates fell from 80% implemented with low inflation to 36Z with high inflation. It is likely that this reflects the growing concern of policy makers with the inflationary effects of a nominal devaluationa and increasing reluctance to use the nominal exchange rate as an instrument of economic policy as inflation rises. Overvalued or unpredictable - 20 - real exchange rates are common in countries with high inflation, such as Brazil and Yugoslavia. With extremely high rates of inflation governments are more apt to use the exchange rate as a nominal anchor thereby sacrificing adjustment to stabilization objectives. Most supply side policies are also adversely affected by inflation. Trade, agricultural pricing and social policy reforms appear to be the most difficult to implement in a high inflation environment. Economic Growth Table 3.8 relates implementation of conditionality to economic growth. It indicates that where countries have been successful in reestablishing adequate, sustainable rates of economic growth, implementation rates are substantially improved across all policy categories. This suggests a causal relationship between growth and the degree of success a government is likely to have in implementing adjustment policies. An equally plausible case can be made that the more successful a government is in implementing adjustment policies, the better its growth performance will be other things being equal. Causality probably runs in both directions. IV. FY89 Adjustment Lending Background During FY89 adjustment lending continued to grow in importance in the Bank's total lending. The Board approved 34 adjustment operations for a total of $6.4 million, up from 29 for $5.0 million in FY88 (Table 4.1). Adjustment lending as a share of total commitments rose from 26Z in FY88 to 30? in FY89. Within total adjustment lending, SECALs continued to grow in importance (20 SECALs were approved in FY89, up from 18 in FY88). The relative importance of adjustment lending was greatest in Latin America (482), followed by Africa (24Z), Asia (14Z) and EMENA (14Z) (Table 4.2). The main foci of FY89 SECALs were industry and trade (nine), financial sector (seven), ai.ni agriculture (three). SALs continued to focus on broader macroeconomic adjustment policies. In Africa six SALs were committed in FY89; in LAC, four. No slower disbursing, sector loans which finance a time slice of a country's sectoral investment program (which RAL I had noted were a form of lending particularly well suited for dealing with medium-term institutional reforms) were approved by the Board in FY89. It was initially envisaged that the Bank's adjustment lending to a country would continue for several years through a series of operations (up to five). The number of countries with multiple operations increased further in FY89 (from 40 in FY88 to 47). Turkey had the largest number of multiple adjustment loans (10). The Africa Region had the largest concentration of multiple operations, followed by LAC and EMENA (Table 4.3). _rna aWI ttt: mter with *w eoi AdiWutmwl toWx t7tomMtietn ot1 -satitlen itr U2 cew rr concurrot for Ibch the 1w progrinnt ii Fra0r- li P rogsr_ _a tulip wd at tffact leWI at teet a of Comeitled U4 Average rate of to s ipag7esitd *OSDJIUtIIIIYFirst 12 at,. *we@@@@oNN@ %Iw^e00i f*w "0none .. e ee.. . .. ee *..feaeim..m........... .... le to *o sut n 222 22 3 -10 to10 S 2? Fs 92 21 Is 4 SAL 22 12 n1 "IeAi, to to 0 St CI 12 9 Il 12 SAL is Sr 73 261to50 UCAL 26 0 141 to 2 sm to as It 12 9 3 - So teo 550 Sam 53 73 SAL 7 52 o notes: (1) Ith rate of Inflation is taead an tma avrag tml pawcantag chabga of the CP Index for the qartrawe Weriod bagiming the rtr of Lown effectivenes. (21 th, rate of iqptinwrtatlen is boad an t, aerae shwre of condit -a_ ttpt _ntad per town. In this reard only cridltlaw tiated In the toen Wgreant wre creldared. A totet of 53 YUtL ad 35 SxLs were grrdd on Iaspitntation. t!) Includx Stwid-by arrwmgta* Extetndd furid facititiee, Strututtral tAdtment faecUlties. and trismed StructuraI Adjetmnt facilltlex. (41 About half of the camreltd funda Progrew wor. amheqauntty renewed, within three Onthb. preou_t7y with revisin. Faur of five cxiceited progres ti high inflatlon situmtlon were renewed. So-co Used - an wtralye by of IS5 SALes ard SCAls to 61 de.eloplg7 cor,ntrles. intatimon dat* bao on e q rtrly dae draan iro IFS. Table 3.7 1lamm1tl od Zflstion Average Ross of sf lull. Oveer Loen Period (1) Low tli... than 1061 l oderste (bet..n 1011 and 306) Hish (greeter tm M0I Fully end Fully and Full1y *n loplemen"ie Subetotielly l_lI_nd Substanieally I _plemented Substantially Implmuented 1_un 1_n I. Absorption Reduction Policies (2) Fiscal Policy 61.0 95.2 79.1 00.4 o5.0 60.0 II. Sietching Polici se Eachonge Ret. 100.0 100.0 64.6 92.3 40.0 00.0 Xll. Supplp-Sid., Orowth-Orie.t*e Policies Trdeu Policieo 84.7 6U.2 62.3 ".9 as. a .9 Sector-l Politi.. Industry 60.0 72.4 76.S 100.0 100.0 100.0 EnrOV 70.0 90.0 71.4 71.4 100.0 100.0 Aericultural 76.9 92.3 4.9 66.1 62.9 15.7. Financial Sector 7J.3 91.3 75.0 65.0 64.7 100.0 Retionelieation of Govenment financ* end Administration 83.3 93.3 56.7 73.8 40.0 68.0 Public Enterprise Retore1 65.7 94.1 69.8 90.6 6. 60.8 TOTAL LegCl Conditions 71 8 91 3 60.3 84.8 508 . 7.5 All Conditiona of Action. 59.4 aV.9 57.4 76.6 49.2 71.6 tk&2: (1) The people of adjustment *rogram is di.ided into tiero. groups -- tloo eaperiencing lowI moderate, or high roaes of inflsaion during th- loan period. Here the interest is in the nu mbe of conditions implemented as a sherp of all conditions over th, entire asape. The rate of inflation during the loan period a calculetod we tho *.arage rote of inflation dur;ng the three year .or*od beginning the ;*ws prior to ffectivenove Infl.t-on data is based odn th. percntage change of tho CPI ourjc Onsed on a panpe of 57 adiustment loan. to 26 doeeloping countries. Table a I*S auatat.os snd * Ad,Dt4k (49 to 6X1 LOW to 301 pb ci, t.Lk too to -5 Fs IV At l...t at Least a of At l..t I o 1ae,tod Sbowntiolly il_aant S.bmtantiIly trted Sbeastnin a*l Ip lJlaeatod zmpUaaeete Ruplesantd . oaratios raductio. Fiscal policy so n2 *01 7a fUudgat Doticit Rsd.ction) ftmotop polIcy so 100 7o 7IO ICO 100 1h,,q Seply TiaGota) 11. SeWitchi l ici Eachrne rate 76 loo 8 100 3 50 I wepolicy so 83 47 67 0 0 - 111. hmh-- ".t'h ori_nted soliciaa Tr." policda U at 0o n 47 of Sectorsl Policiea I etry 50 67 I00 100 0 1CO 6e argl so0 as 6o o0 7? 77 *.i;rlte,al es so 44 64 29 46 Fi_esc;-l a-ct 82 el 70 es 48 66 Rationalial;ian of goeeaat finance A adiniatr.tia. 0 74 62 71 S T Public sntepri. refores so 78a 5B a 70 76 Social polioy eforms 70 OD 4 63 0 0 I/ Gr-thU .t. d.f,nd* e. Lho thnn.l p.rcent*qg chan in QW denotd n 190 conet local c-.,,ner Som tea- -- .a "bset of 57 *eJdotee"& ane Table 4.1s: Fn9 ADJUSThENT OPERATIONS STATUS REPORT/ (as of June 30, 1989) IBRD IDA Total Board Status Country Name Type Board Date Ntiber (USSn) (US51) (US$m) Approved Argentina Trade Policy 11 Loan SAD 10127188 1 300.0 .0 300.0 8angladesh Energy Sector SAD 0411/I89 1 .0 175.0 175.0 Industrial Sector Cr. Suppl. SAD 03113189 0 .0 2.5 2.5 Benin, People's Rep. SAL I SAL 05123189 1 .0 45.0 45.0 Bolivia Finan. Sector SAD 03113109 0 .0 11.3 11.3 Burundi Agric. Sys. SAD 05125189 1 .0 33.1 33.1 Caseroon SAL I SAL 06114189 1 150.0 .0 150.0 Chad Financlel Rehab. Cr SAD 07127188 1 .0 16.2 16.2 Transport Sector AdJ SAD 04125189 1 .0 60.0 60.0 Costa Rica SAL 11 SAL 12113/88 1 100.0 .0 100.0 Gambia SAL 11 SAL 06108189 1 .0 23.0 23.0 Ghana FtH. Sector Adj SAD 03113189 0 .0 6.6 6.6 SAC 11 SAL 04116189 1 .0 120.0 120.0 Cuinea Bleaau SAC 11 SAL 051i8I89 1 .0 23.4 23.4 Hondurae SAL I SAL 09/l5S/8 1 50.0 .0 50.0 Indoneale Private Sector Devt. Loan SAL 06106189 1 350.0 .0 350.0 Kenya Finamcial Sec. Operations SAD 06127189 1 .0 120.0 120.0 Ind. Sec. Ad). SAD 03113/89 0 .0 53.7 53.t l-o P. 0. R. Struct. AdJ. Credit SAL 06114189 1 .0 40.0 40.0 Hada6ascar Public. Sec. AdJ. SAD 03113189 0 .0 1.4 2.4 Halawi Ind. Trade AdJ. SAD 03113189 0 .0 5.2 5.2 1 mali Human Resource SECAL SAD 06127189 1 .0 26.0 26.0 Mexico Fin. Sec. Ad). SAD 06113/89 1 500.0 .0 500.0 Ind. Restructuring SAD 04/27189 1 250.0 .0 250.0 Ind. Sector Policy SAD 06113189 I 500.0 .0 500.0 I Pub. Enterp. Ref. I SAD 06113189 I 500.0 .0 500.0 Morocco SAL I SAL 12101188 1 200.0 .0 200.0 Mozambique Rehab. III SAL 05118/89 1 .0 90.0 90.0 Nepal Structural Adj. 11 SAL 06127189 1 0.0 60.0 60.0 Nigeria Trade & Inveat. Policy Loan SAD 12122/88 I SOO.0 .0 500.0 Pakistan Agric. Sector Loan SAD 00102188 1 200.0 .0 200.0 Energy Sector Loan It SAD 0629189 1 250.0 .0 250.0 Financial Sector Adj. SAD 03128189 1 250.0 .0 150.0 Philippines Financial Sector Loan SAD 05104189 1 300.0 .0 300.0 Seneg a SAL IlI SAL 03113189 0 .0 5.5 5.5 Somalia SALIASAP 11 SAD 06101189 1 .0 10.0 70.0 Tantanit Ind. Trade AdJ. SAD 03)13189 0 .0 12.5 12.5 Ind. & Trade Adjust. Cr. SAD 12/31188 1 .0 IS.0 135.0 Togo SAL III SAL 03113189 0 .0 .1 .1 Tunisia ASAL 11 SAD 06101189 1 84.0 .0 84.0 Uganda IDA Reflow Supplemental SAL 03113189 0 .0 1.7 1.7 ERCI Supplemental SAL 04120189 0 .0 25.0 25.0 Uruguay SAL 11 SAL 06108189 1 140.0 .0 140.0 Venezuela SAL SAL 06115189 1 402.0 .0 402.0 Trade Policy Loan SAD 06/15189 1 353.0 .0 353.0 Total Approved 34 5279.0 1162.2 6441.2 Total 34 5279.0 1162.2 6441.2 Type: SAI. - Structural Adjustment or Program Loan Revised 612190 SAD - Sector Adjustment Loan Supplemental and 8-Lans are shown as 0 In the number column. - 25 - Takla 4.2: PY89 ADJUSTYINT LENDING No. of No. of S Total SALs Secal Secal. Total Lendin8 LAC 4 692.0 6 2414.3 3106.3 48.2 AFRICA 6 483.7 8 1039.7 1523.4 23.8 ASIA 3 450.0 2 477.5 927.5 14.3 EMENA 1 200.0 4 684.0 884.0 13.7 TOTAL 14 1825.7 20 4615.5 6441.2 100.0 Table 4.3: NUMBER OF COUNTRIES WITH MULTIPLE ADJUSTMENT OPERATIONS Number of Countries Number of Countries with two or more vith three or more Adjustment Loans Loans AFRICA 22 14 ASTA 6 4 LAC 13 10 EEMNA 6 4 REVISED 8/2/90 - 26 - The Macroeconomic Underpinnings of FY89 Adjustment Lending To assess the macroeconomic underpinnings of FY89 adjustment lending, economists in the Macroeconomic Adjustment and Growth Division reviewed 23 adjustment loans to 18 countries7. The sample includes six SALs and 17 SECALs which are included in the list of countries which the Division covers as part of its on-going review work. Background case studies were prepared for each of these loans and are available on request. They are based on reviews of Initiating Memoranda, President's Reports and Loan Agreements, the agendas and minutes of Operations Committee meetings, and discussions with regional staff. The summary of findings which follows focuses on three areas: (a) the initial macroeconomic conditions at the time the loan was committed; (b) fulfillment of preconditions of adjustment lending; and (c) the design of the Bank's macroeconomic conditionaiity. Initial Conditions To what extens were the countries in the sample experiencing significant macroeconomic difficulties at the time the loan was committed? Reviewers found that almost all of the countries were experiencing some form of macroeconomic imbalance 'n FY89. These difficulties ranged from hyperinflation, acute external imbalance and accelerated decline of output in Argentina to mild stagflation in Uruguay. In some countries the economy had been stabilized, but thus far it had not been possible to reestablish growth (Bolivia). In some of the sample countries, macroeconomic imbalances were becoming more severe (Brazil, Pakistan and Cameroon). In others, the situation was improving (Ghana, Mexico, Uruguay and Philippines). In all of these countries, with the possible exception of Indonesia, a plausible case could be made that there is a need for Bank support in the form of adjustment lending to either reverse deterioration of the macroeconomic environment or to sustain the adjustment effort and help ensure against policy backsliding. Of the 18 countries in the sample, only Indonesia appears to have overcome its macroeconomic imbalances. The country had already undertaken a major adjustment effort and now appears to have recovered from the oil price shock of 1986. Growth of non-oil output has been restored to over 52 annually, inflation is under 102, and the deficit in the current account of the balance of payments has been reduced to under 32 of GDP. The Region's adjustment lending strategy for Indonesia is to use adjustment lending as an ex post reward for actions already taken, rather than as an inducement for undertaking new stabilization or other macroeconomic policy reforms. While there may be special circumstances which justify continuation of adjustment lending for the next several years as envisaged in the Region's 7 Even countries with reasonably consistent programs are not without problems. Some high debt countries such as Bolivia are unable to reestablish adequate rates of economic growth despite the existence of a relatively stable macroeconomic environment; in this regard, even for these countries, adjustment programs were not fully consistent. - 27 - Five-Year Lending Program, the Indonesia case does raise an important question: At what point following restoration of macroeconomic balance is it appropriate to discontinue adjustment lending? Preconditions for Adjustment Lending In analyzing the macroeconomic underpinnings of each adjustment loan in the sample, the assessments focused first on whether the preconditions for adjustment lending had been met. There are three such preconditions: (a) that the government "own' the program; (b) that a consistent macroeconomic adjustment program be in place at the time of Board presentation and that there be a recorded agreement between the Bank and the borrower on the key elements of a medium-term adjustment program, including short-term stabilization measures; and (c) that the program be realistic and adequately financed. Most countries were found to have met the first precondition of adjustment lending. This was the case for five of the six SALs reviewed and for 13 of the 17 SECALs. However, reviewers encountered practical difficulties in making this determination. First it is difficult in most cases to find direct evidence which provides an unambiguous indication that the government is on the record publicly as "owning" the program. Second, as a consequence of this, reviewers were forced to rely heavily on what they could glean from documents prepared to justify the loan. These documents are part analytical and part advocacy in nature; the presentation is such that it is frequently difficult to distinguish the two. Judgments made by reviewers may reflect the positive interpretation of events which is generally contained in these documents. The results shown above may, therefore, overstate the degree of Government ownership of the adjustment program. The second precondition has two parts: (a) that a consistent adjustment program exists at the time of Board presentation; and (b) that there be a recorded agreement between the Bank and the Borrower on key elements of the short-term and medium-term program. The reviews indicate that about half of these countries met the first part of this precondition. In a few cases it was reasonably clear that a country had a consistent adjustment program in place at the time of Board presentation (Bolivia, Uruguay, Philippines)8. It was equally clear that such a program did not exist in others (Argentina, Brazil, Cameroon, Pakistan). For most countries in the sample, however, the judgment was more difficult. For some countries, the existence of a Fund program provided some indication that a consistent stabilization program exists. In Venezuela, for example, the government has a strong commitment to adjustment and has in place a reasonably consistent short-term stabilization and medium-term 8 Even countries with reasonably consistent programs are not without problems. Some high debt countries such as Bolivia are unable to reestablish adequate rates of economic growth despite the existence of a relatively stable macroeconomic environment; in this regard, even for these countries, adjustment programs were not fully consistent. - 28 - adjustment program. The authorities have worked closely with the Fund and the Bank; an IMF Standby Agreement is currently being replaced by an EFF. Negotiations with the commercial banks have not yet been completed, however, and other elements of the external financing program are in doubt. Because of apparent or real political constraints, the government has had to act less forcefully in some areas of reform (liberalization of interest rates) than it had hoped initially. Even where an IMF program did exist in FY89, this did not ensure that a consistent macroeconomic program was in place (Brazil). To judge whether a consistent macroeconomic program existed, reviewers asked a number of questions to make their determination. These questions and the results are shown in Table 4.4. Reviewers found that for all SALs a Fund program was in place at the time of Board presentation. For SECALs, Fund programs existed in only two- thirds of the FY89 AL countries. The practice which currently exists in Operations is for the Bank to reach its own independent judgment of the consistency of the macroeconomic program which is being supported by the Fund. Reviewers found that in most cases, this appeared to have been done (about 80?). Reviewers then turned to the quality of the analytical work done by staff to assess the consistency of the macroeconomic framework. For all of the Bank's borrowers, a Revised Minimum Standard Model (RMSM) had been prepared. The standard RMSM, however, does not ensure consistency between the budget, the public enterprises, the monetary sector, the external accounts, the private sector and the national accounts. It is not possible to check the macroeconomic consiotency which underpins an adjustment loan with the RMSM. To ensure the consistency of short-term and medium-term macroeconomic policies a more sophisticated model is required ("RMSM-X" or "XXT). Reviewers determined whether a flow of funds consistency model (or other model which incorporates behavioral structure) had been used by the Bank to evaluate the macroeconomic framework. The results indicate that in about 60? of the AL sample countries this had not been done. For the second part of the precondition (that there be a recorded agreement between the Bank and the borrower regarding the adjustment program) reviewers looked at Letters of Development Policy which had been prepared for each adjustment operation (Table 4.5). While the Letter of Development Policy has no legal standing, it is the principal vehicle for recording understandings which are reached between the Bank and the borrower. The findings were that in mar.y cases (73Z) the Letter of Development Policy does contain a general understanding on overall macroeconomic policy. However, in only about one-third of the cases do these letters record understandings on short-term stabilization policies. Finally, reviewers evaluated the extent to which the third precondition had been met by assessing the adequacy of external financing and the realism of the analysis. In about half of the cases the analysis was judged to be unrealistic. While this judgment might be questioned as being excessively demanding of the country projections, it is more favorable than the results of a similar assessment of realism recently carried out by central staff in Finance which concluded that only 25? of the country projections are realistic. - 29 - Table 4.4: AOIACY Of MACROICONIC FRAMEWOK IN FY*9 AL COIMTIES S1AL. SECAL. Total Was there a Fund Proorm In plicet Yen t00 ,.7 76.2 me 0 $8.8 22.8 Did the lank evaluate thil Proram? Y" 68.8 n.6 73.6 me 16.7 22.2 21.4 Did the Book use a cosnistency model to Yes 88.8 41.2 38.1 check the saero coamic frameeork? me ".7 f,.3 60.9 Was the analysis done by the Bank Yee $0.0 l@.@ U.5 rsslistlc? me 30.0 41.2 48.5 - 30 - Table 4.t fYS ABJJSTMOiT LOAKS: ACtIMYff ON MACROECONOMIC POLICY CONTAD DiN LETTES OF DEVELAPMS4T POLICY GCneral Agreement on Macroseonomic Agreoont on Short-Tore Country Loa Now Conslstency of th Progroa !/ StabiIlzation Policy q Argentlno Trde Peolicy N N hga I adi Eertgy Seteor 4 N S1Irt SAL t y y Burundi Agricultural N N Comre.r SAL I Y Y Chad Transport See N N Financial Rehab Y N Coe"t RIcO SAL It Y Y Gamble SAL 11 Y N Ghana SAC It v Y Gulaoi elsi u SAL ItI Y Y Guinea SAL Y Y lmebduras SAL I Y Y Indonesia Private Sector Developent Y N Konya PSNIL Y N Lao P.D.1. SAL I Y y Mexico Ind. Pol. Y N Pt N N PINSL Y N Ind. Reotructuring y N Morocco SAL I Y N Mo 'mbique Rehab. Cr. III Y "epal SAC Kt y y Nigerla Trode/Inv. Pol Y Y Pokilton Energy Sec N N Agric. Sec N N FtNIL Y N Philpplnes FPNSL N Soeoli Agricultural Y N Tansania Ind. Trode Adjustmnt Y N Tunisia Agricultural Sector It N N Uruguay SAL tI Y Y VeOH uola Trode y N SAL I Y Y j/ ThIe eelso. ldleates whether thore Is (Y) or Is net (N) a genral agreement on meroeeonomic polley. The asr seemt or maroeoneo ic policy cowr te following areas: esehogoe rate, toa, w*o debt m sent, budget deficit, public currest ar capitol expeniturco, subeldloe and me_y "uply PO lles. ki This colum indicates wbether them Is (Y) so I mt (N) a specific sgrsm a Short-term stobilzartioe policies. Sleurce: President's Report, Letters of Developomet Pol icy - 31 - Macroeconomic Policy in Program Design The analysis next turned to the macroeconomic content of Bank- supported adjustment programs and to an assessment of the extent to which key macroeconomic policy elements of the program were incorporated into the conditionality of the adjustment loan. Two-thirds of the loans were found not to have introduced (or to have been introduced only partially) the key elements of the macroeconomic policy framework. Table 4.6 lists for all of the adjustment loans committed in FY89 all of the legally binding macroeconomic conditionality. Eleven loans, including one SAL had no such conditionality. Only five loans had three or more legally binding conditions. One loan, the Cameroon SAL, had a total of 275 conditions of which only three were legally binding macroeconomic conditions. These three conditions were general ("adopt the public investment program in FY89 and FY90' and "provide adequate funding for the execution of the program"). The Bank frequently has included macroeconomic policies in President's Reports as a "desired action' (i.e., not legally binding). Reviewers focused on program content to identify which macroeconomic policies have most frequently been part of overall program design. As discussed in Section II and applied in Section III, policies were classified as expenditure reduction, switching and, supply-side policies. Fiscal deficits are frequently the driving force behind inflation and balance of payments deficits. They contribute to high real rates of interest, lower levels of investment and declining growth. To what extent did the Bank's FY89 adjustment loans contain fiscal deficit reduction as a condition for second tranche release? It was found that about two- thirds of the operations reviewed had such conditionality; in most cases, this was not legally binding. Switching policies were given less importance in the loans reviewed. For 602 of all SECALs there was no mention of exchange rate policy. While a large number of SALs included exchange rate policy in the program, there were no instances where it was a clearly specified legally binding condition. The most important focus of the Bank's FY89 adjustment lending continues to be supply-side policies aimed at increasing output (Table 4.7). Legally binding conditions are much more important in areas related to supply- side policies. About half of the FY89 loans reviewed contained legally binding conditionality related to public sector and fina:..ial sector reform. Such conditionality was also found to be important in policy areas dealing with trade and pricing policy reforms; this was included in about one-third of the loans. Many of the conditions included in an adjustment loan are complementary; some may be potentially conflicting. For example, liberalization of the trade regime which includes tariff reduction may conflict with budget and current account deficit reduction. Reviewers - 32 - Tab I, 111 L.v I@Vo 101m1 71MWrT PO *oJAnTIy 60 Cami4py LA" 1b4s CdetiOm LI gI CeiS;el VI /s of c..oltidws A,gent,fte T'ade 22 7gAei 46. e. fio;w_ of tie .eel .s"*ng. ' . T eed*4 21 Cee toeeapy of Omo"a"w*g C0g1 ga4 m_ t"py 0e.,e00 *eth oel.y f.e.me. e' enol Enry b..w E yletr.. S... WAL I Sn -. Iv.AEji AgP. 103. * 7. Ce._... Ul I 27 U Adow, the euh 4 bi* A.04t4 free.r. for P'H. SAL 1 275 23 SAge A4, sebi l6 e.1eetmoI e.gess, fop '1". SAL 275 2 Pise. dem.B46 f..egA4 for she o,"t4es, *f th.. aprgoe. Chad *'.. AjM" . j. pi.^s. Po46b. 23 2 I A04 its, thCOoeyOe PIP. F;Mee. llemob. 25 2 l_f supplemntary budget. P1..... Rehab. 25 Z! AiSo .4AS 14 budget fo, eereost g~evoo14usoe. Coo" Rise SAL II ea 2 :a ceo. _ss , legislation OgSaSioAi"e o.rebed tins pewainv durinug PM. SL M a ZU I psoo to I I s sin an I saft wd" I do I peliey of mee1 A4ivegw 14. MAL III 15 U2 Rooegh lovei of .blee sevoine, of 5.71 of MP by 1Lou. SAL :11 a Z Ratak I,01 of oe1 s* i se Mt St of MP by 1119 At. z a U leeI b I i di budO46fy f6.305t S0 be" ovl i U.-AO OS I Oi them a of th. MP fop low. SAL a 22 tod th. OvoiSOd 1966-15 iA0t op *S 0 6o. SAL II a 2 On nt sworo say dieSS C.o-o ombeedee 46 O60.6609. Casoe MAL Ui so 25 b.in46ag.g of tAo bgg lib edis eve46wse .16 0,y4sg. SAL it 1 Pgster o modIow-kof Public 0aoo-nd.tu.. oPoo ... Cho-* 4AC it 21 ,3 IIIS 1I66 pmeog. of oaf cofee, foe 4 .i,I SI it 1*1 f ee ish UP .6 11O-0111 subt is in.enmo. opeO,.. C it 121 22 1 A.P** i04 U on tow P0.699.64 04 t.eww ee SM t 121 U I Ag..oa Ieoh MA 6460 Nom rwomo in IO beedoos. Ceeneeooisemr WSA I 4 U I AdUWftmAm% of the OffSil .obgg.o e.ft. 2U; I . t e Uio of th figogI sad master e.ge... for VI.. C..eege ~SAL It II1 2 Admb PI for the tapied 196 I.41 SAL II lot 23 I lebI . s -ateel sI epes t. 46..gpo' f I .o b"..a of eamma at ee fto, Svgpgyer*. SAL tt 1ol 2 *sft he 166t b;;i_o1. UL I s o 'w I 0te..s ihe g,g.g30 os,*o i 46.in for IWO by n. i n ,ee 40 gilIeIA f iIeP". MAL I so 23 Aiugeo th de;faitA of the ob I I gseeSe by IIU ; gi ain of Isinse.. SAL t so U Zm_s, 40s 01104i_ OS"i. SA. I s0 a A.oet nI GOgee_*e4 sd the sbo .6 ane inetmn oe.g.. fee VIU. Ze4Eeeim P,;.. Se. O,. 70M Kyl e a PT6L 44 U Sesiefget"u .ortafeeme ;e P"* swings ogre" 46reoto fee budge" and domestic fifg.oe.O.S Lee P.*.*. SAL. I U2 22 1 Ade.4te ad efeer ,Ou.emtee %&ig4eoewe the Owegpgti. of son PgYww0. " e tee Ind. Pei. 90 2 C.6iee.e Segepeo ob46i"ifood644.654 flooe_,ieeg foe "oise pe1tpeof to6 few t1 9, Pe 4 U Ppggg.e4 gewdol leo abos eoSee 6.0eepe4eb de.olg0ORSe PNL m e". Ined. ftesloPeseg4uM9 me" Moseges SIAL I a 21 1 Reetrusivpo inegtime pggoeeble fe gepes gobt. SAL t A U Isetbi abi oeiyoe O, tofret i".Vestm" se...s for to". M8906bi4ou 366b. Ce. III UT go P rom n.hi 0.0 on the .oi"g of %he ltr_ep ope =ibaogoditme.e e.e Mogul SAC S U 8 I Ad*otise Of the bewdbe fe, PM. SMC US 15 25 Op.I g s stGauf1;,*4- p dc . o ;v* o inel t osi _is . i rig toey*/Ipr. Psi lop Z 1 Osoesto .1 mt beS ig .edie t4 m.gtios OP eaGe 100. T,94eiIe.. PeiI eV U UPp " semetes,e4 sPeisiglis foe Me6 g.46.eol 7.sdoIee. P- l0 U 1 fiemle yit 2w"i *| _i_ _*fit mo bglomong 4efetipn.%* Yprggievo. Vol I p 22 U Peinoloe 1.oTlgi-botegeWSW b.ege,e fith 1mov. P.4.5.6 be. Soot.. 96 h~~~~~~~~~elem, Soote' ~~~22 21rnIe. "' 15 a PUP Pst odiese,g boguemtA dofiolit/ etlie foe P190. Tegel lad. 1.4. Adj. 15 a NgdWeo 16PIff md emlem SON es0mOI%e. lad. T.4. Adj. 241 U Pegegeg moftio re.g.m of Mesle 46too ogb Tete ee, . it 40 100r. UPeepia, II t5 U Pemogoe sotli plan 46, 1s64t too .ofoe. end i0geree Ad too l eft i m. SMA. ISt 22 I Rgduoti of gebig is 46e 4Ofisil 46t 1.1111 of CP. SML II1 U dgof "ibi la. 6aseee., eeiro" fee ,ooro 1w96.15. A/ 6SflI fi. 14 T1 . if te mdist. ile I. thre I tM __e m ga ds Woliooe 4 egWr_d. LfLeonga effgetiseem U2 a somem teud Ps e i m Third _elookd, vg4ees1 a * "Neal isnd"bism A" pgg"4d _6 Wwe P alawan. V I * lAd10.50 itf the ndetiSs ism bop omisgt of th, go.. - 33 - Table 4.7s SUPPLY-SIDE POLICIES IN FY89 ADJUSTMENT LOANS SALS (1) SECALs (2) Total (X) Public Soctor Reform Legally binding 66.7 35.3 43.5 Desired 33.3 29.4 30.4 None 0.0 35.3 26.1 Trade Policy Reform Legally binding 66.7 23.5 34.8 Desired 16.7 35.3 30.4 None 16.7 41.2 34.8 Financial Sector Reform Legally binding 33.3 52.9 47.8 Desired 50.0 11.8 21.7 None 16.7 35.3 30.5 Pricing Policy Reform Legally binding 66.7 17.6 30.4 Desired 0.0 52.9 39.1 None 33.3 29.' 30.4 -34 - assessed the degree to which there were potentially conflicting conditions contained in an operation. Numerous examples of such conflicts could be identified. Conflicts among financial reform and budget deficit reduction were present in Brazil and Nigeria. Trade liberalization conflicted with budget deficit reduction in Argentina and Cameroon. Finally, exchange rate policy and the decision to maintain ceilings on interest rates in Venezuela seem to be inconsistent. Overall, about one quarter of FY89 loans reviewed were found to contain such conflicts. Related to potential conflicts in the program is the issue of appropriate sequencing. Sectoral reforms may not be effective unless they are preceded by adequate progress in stabilization. In addition, adjustment costs can be minimized if the program is designed in a way that avoids giving contradictory signals to producers by overshooting of exchange rate or price adjustments, if this is achieved, the movement to the new equilibrium will be as smooth as possible. Reviewers found that about one-third of the FY89 SECALs had been made before there was adequate progress in stabilizing the economy. In about half of all adjustment loans reviewed they had questions about the sequencing of the measures proposed which were not addressed in the loan documents. Table 4.8 indicates those countries which reviewers considered to be examples of best practice in ensuring an adequate macroeconomic policy framework and in resolving key issues in the use of adjustment loans. Conclusions and Issues Identified in the Case Studies A substantial amount of case study material was generated as an input for the paper. The following is a summary of issues and conclusions identified in the case studies: (a) Analytical Framework. Adjustment lending should fit into a medium-term macroeconomic framework which ensures consistent demand management, switching, supply-oriented growth policies, and adequate external financing packages. This framework should at a minimum be based on a quantitative analytical structure which links stock and flow variables for the non-financial public sector, the monetary sector, the balance of payments, the national accounts and the private sector. The emphasis should be on financing the government budget and the external sector. Resources, time and data permitting, this quantitative framework would be extended to include a behavioral structure which would permit disaggregated simulation of the effects of adjustment policies. Analysis carried out by the IMF will be useful in elaborating the macroeconomic framework. However, the Bank and the Fund differ with regard to objectives and priorities. For these reasons the Bank must form its own judgment of the - 35 - Tabl. 4.: ASISSCt OF MACORCONOUC uwRIINO OF PYSO AOJUSThD LE110: '0U5T PAACtICISO Z* Pr cond1tt;na far Adiustt i.-dtrna Oweverut a"enohip *f bprer Uruguy. Mseat., Ind.ewc, Ghana Camero Exatstenco f am ad"uate ma.re.co*i@ framework (a) Cnoneitatt 0e4raeCAeIC c pe Pfi Uray, Wec (b) ReOGrded agrieit en short and _1um- term meareanmi Policies Urgy C_mren, Guiww t"Hlim of prejetione and ad"uacy of osternol financing Rugua s1ndeoei, Ghana. Ssngal. h 2S. ,Le eLoa C.lnt Maroerano c Conditionalltv Reity of Bank's .ereeemate analyels Malme*. venesala. Pakistan P"eu of cedittl.nality en key mcroeeeninle prbim w/ Meat l", Pakista*, Uruguay Pollcy s"tie prier toe nftwcness aeavela. Indonesia Appropriate 1 WInl n of deand onnagmnnt policies Veaumela, Uganda, Seiliis, Mexee Appropelote Il a luele of eacheg rate Policy Vasogula, Urgaty, Ghana Focus ao supply-ailde eroe Ghana, Meteeo, Philippine SS. Ke, lose" lin Prears De_lm Adeuate t n of fItet smeg" abJetiyso babe, bro"ec 1Reelstie %timin of Mefer &lel1a, sates, Snindmet Correct e nof refre Bolivia, Ghana Appropriate am of 6CGL. bel, lbee, Pabtltan, qhans i/ Poro le** offeelvnea or leplly bilding fee ..ead-transhe rleases. - 36 - consistency of the short-term stabilization component as well as the medium-term structural policy reforms. Given that the Fund has an increasingly mixed record to its programs achieving their objectives, it is also not sufficient for the Bank to rely on a Fund program to ensure that stabilization policies are adequately implemented. (b) Realism of Projections. A frequent shortcoming of recent adjustment programs is excessive optimism about: (i) the level of resources needed and investment required for switching to occur and growth to be restored; and (ii) the lag between policy change and response of economic agents (i.e., even where adequate resources are available, investment and growth may not occur given uncertainties about the lag structure in a number of key areas). Lack of realism has frequently led to underfunded programs which are reflected in either excessive compression of absorption (most frequently of investment) or in substantial divergence between unrealistic ex ante program objectives and ex post macroeconomic performance in which external balance is achieved at the expense of internal balance and growth. (c) Focus on a Few Key Policy Conditions (Avoidance of "Christmas Tree' Effect). Macro conditionality in both SALs and SECALs should focus on a few key conditions. These conditions should, as appropriate, be time-bound, legally binding and as specific/precise/quantitative as possible so that there is no ambiguity about the Bank's commitment to implementation of the key r roeconomic policy conditions. An excessive number of conditions which dilute the key conditions may in practice make it more difficult for the Bank to use its leverage to support policy reform in key areas (e.g., if 75 of the 80 conditions of an Agricultural SECAL have been implemented, the Bank may appear to be unreasonable if it does not agree to second tranche release even though a few macroeconomic conditions -- albeit key ones -- have not been met). (d) Need to Record and Monitor Implementation of Macroeconomic Component of Adjustment Program. Agreement between the Bank and the AL country on specific elements of the macroeconomic program should be recorded for all SALs and SECALs so that there exists a point of reference for monitoring subsequent macroeconomic performance. This should be done in sufficient detail that key elements of the macroeconomic program are made explicit. Without this (i.e., that the Bank generally "is satisfied" with implementation progress), there is a danger that insufficient weight will be given to ensuring implementation of key macroeconomic conditions and excessive weight will be given to other objectives such as avoiding a negative net transfer and fulfilling lending targets. - 37 - (e) Proper Sequencing of Structural Adjustment Measures. Since some components of an adjustment program will be ineffective or will have perverse effects if they are not preceded by improvements in macroeconomic pe:formance, proper sequencing in the design of adjustment programs is important. For example, efforts to bring about shifts in relative prices are unlikely to be effective if high and volatile rates of inflation exist or are expected (relative price signals will be swamped by the general price increase and will not induce the desired switching response). Similarly some sectoral adjustment objectives will be Impossible to achieve if they are not preceded by adequate macroeconomic policies and institutional reforms. Efforts to liberalize the financial sector (reduction of reserve requirements, elimination of forced holding of government paper, liberalization of interest rates and relaxation of cumbersome banking regulations) are unlikely to be successful unless the government's fP cal deficit is under control, inflation (and inflationar; expectations) has been reduced to a manageable level and administration of the commercial bank regulation framework has been improved. (f) Sequencing of Stabilization Policies. Macroeconomic policies themselves must be properly sequenced if they are to have the desired effect. For example, freeing of interest rates before large fiscal deficits are brought under control, the exchange rate regime is liberalized and inflationary expectations are broken, is likely to result in excessively high real rates of interest which are inconsistent with efforts to reduce the fiscal deficit and to reestablish an adequate rate of economic growth. Similarly, current account deficits in the balance of payments must be brought under control, interest rates liberalized and a more market determined exchange rate be put in place before the capital account can be opened. Otherwise, destabilizing capital flows (capital flight if real interest rates are negative in real terms or massive, destabilizing capital inflows if real interest rates are high and the exchange rate is fixed) which make it difficult for the government to achieve its other macroeconomic objectives are likely to occur. (g) Relationship between Stabilization and Financial Sector Reform. The financial sector is a primary conduit of macroeconomic policy with the rest of the economy. Conflict often arises when high real interest rates are part of the stabilization process since they can result in financial distress of borrowers and generalized deterioration of portfolio quality across the financial sector. If this problem becomes acute, restructuring and recapitalizing the financial sector becomes necessary and this can undermine the fiscal restraint necessary to achieve stabilization objectives. - 38 - (h) Reducing the Supply Response Lag. Productive sectors respond with lags to changed incentives; these lags vary between sectors because of their technology and institutions. Lags are apt to be especially long during the period of stabilization. A key issue is how to design adjustment programs so that institutional reform and sectoral policy are used to enhance sectoral supply response. (i) Use of Heterodox Macroeconomic Adjustment Policies. Use of "heterodox" measures in stabilization programs is an important consideration in program design. Heterodox approaches used by countries experiencing chronic inflation, such as Brazil (Plan Cruzado) and Argentina (Plan Austral), initially appeared to be successful in arresting the inertia of rising prices without sacrificing growth. Such approaches focus on use of price controls (wage, price and in some cases exchange policies), to slow inflation and to control expectations. The conclusion from these experiences is that while heterodox measures can help to overcome inertial inflation and can be an effective tool in managing inflationary expectations, they are unlikely to be successful unless they are supported by orthodox stabilization policies, especially by tight fiscal management. (j) Need for Contingency Mechanism for Responding to External Shocks. An important issue in program design is use of a contingency mechanism in targeting net international reserves. If this variable is to be targeted, there is need for a contingency mechanism to isolate the effect on net international reserves of temporary external shocks. In the absence of such a safeguard there is a danger that rolling adjustment program targets will be tightened excessively in response to external shocks and that this will derail efforts to reestablish growth. (k) Real vs. Nomincl Targets. If nominal targets are used and the inflation target is missed. all other nominal, targets become irrelevant. Since inflation is not controlled by any one policy instrument, but is the outcome of many simultaneous forces (i.e., the result of both policy intervention and exogenous shocks), including expectations about future inflation, forecasting of future inflation r tes is exceedingly imprecise. This is particularly true of the high and volatile inflationary situations in many of the middle income high debt countries at present. However, if real targets are used there remains the problem of selecting the appropriate nominal anchor. Currently there is no clear consensus within the economics profession or. this point. This subject is one of the major priorities of future macroeconomic policy research. - 39 - (1) Dealing with the Debt Overhang. There are important issues of program design which stem from a large debt overhang. Service payments on external debt require a twin transfer of resources (external and domestic). Because the public sector is ultimately responsible for servicing the external debt, but savings are usually being generated in the private sector, there is a need to transfer resources from the private to the public sector. Because there are limits on the size of transfer which can be achieved, many highly-indebted countries end up resorting to use of the inflation tax. In such circumstances the government may have to resort to use of high real interest rates to induce the public to hold government paper. As a consequence, many highly-indebted countries have experienced excessively high rates of interest for long periods of time which have the effect of crowding out private investment and impeding resumption of growth. This has frequently eroded the sustainability of adjustment programs. If not dealt with directly, the problem is likely to become more severe with time. There are solutions to these problems but their implementation is usually difficult. One solution is to cut Government expenditures. The difficulty with this is that in practice investment spending is usually the easiest to cut. While it is possible to identify some public sector investments which should not be undertaken, further investment cuts beyond this may make it more difficult to carry out switching and to resume sustained growth. The conclusion is that sustainability of adjustment programs will depend upon proper handling of the debt restructuring process. For some countries, debt reduction is a sine qua non of program design if the adjustment program is to succeed. - 40 - APPENDLX Coding, Conditionality and Implementation The tables in this paper are based on the Bank's Industry and Energy Department (IENIN) Adjustment Lending Conditionality and Implementation Database (ALCID). The sample covers 184 adjustment loans to 61 countries. Twenty-six of these countries had three or more adjustment loans during the 1980-89 period9. For each given country adjustment program, two sorts of information were considered: (1) the specific conditions attached to each Bank-supported program; and, where available, (2) the implementation status, or extent to which these conditions have been fulfilled. The conditionality and implementation were classified according to general information about a given loan and a given borrower (for example, the fiscal year and the dollar amount of the loan) and to more specific information about each condition. The sources of information on conditionality were legal agreements and President's Reports. Characteristics of conditions were coded on six dimensions: (1) firmness, whether a condition is precisely specific and perhaps quantifiable; (2) legality, whether the legal agreement requires the condition for disbursement; (3) timebound, whether a condition has to be fulfilled within a given timeframe; (4) importance, whether the ALCID coder considered a condition central to the adjustment process; (5) policy area; and (6) performance objective. Pakistan's 1989 Financial Sector Adjustment Loan provides an example of a condition which is specific, legally-binding for second tranche release, timebound and important: "reduce the government's budget deficit to 6.5Z of GDP in FY89.n The data on implementation indicates the extent to which a condition was fulfilled at last tranche release (during the lifetime of a loan) at loan completion report, and current status (Summer 1988 or Spring 1989). In the 184 country-program sample, only 97 loans have been in place long enough to have any information on implementation. The sources of informatior. on implementation were mainly supervision reports, tranche release documents, and where available, Project Completion Reports and Program Performance Audit Reports. These sources were supplemented by additional information obtained from Bank staff familiar with the country's adjustment program. While the coding resulted from subjective judgments, there were several checks at each stage to assure consistency of standards. 9 In Asia -- Indonesia, Korea, and the Philippines; in Europe, the Middle East and North Africa -- Morocco, Pakistan, Tunisia, Turkey and Yugoslavia; in Latin America and the Caribbean region -- Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Jamaica, Mexico and Uruguay; in Sub-Saharan Africa -- Code d'Ivoire, Ghana, Kenya, Malawi, Nigeria, Senegal, Tanzania and Zambia. - 41 - The implementation of each was graded: (1) little or no progress; (2) moderate progress (some steps taken but not much progress); (3) significant progress (substantial steps taken to move progress more than half-way); (4) implementation fulfilled; and (5) implementation more than fulfilled. Such a grading process has its limitations. First, it involves judgment; the appropriate "grade" was not always clear. Second, while full implementation is a desirable objective ex ante, it may not always remain desirable. Both economic and political circumstances can change unexpectedly, causing a shift in policy instruments to fit the changed economic environment. Third, it is worth noting that the grading process does not consider whether conditions differ. in their implementation difficulty and the strength of their impact on the economy1o. 10 In this regard, we have considered as well descriptive information drawn from PCRs and PPARs, which goes behind aggregate data. PRE Working Paper Series Contact ALhor for pa2e WPS496 Issues in Evaluating Tax and Robert Conrad August 1990 A. Bhalla Paymer.t Arrangements for Publicly Zmarak Shalizi 37699 Owned Minerals Janet Syme WPS497 The Measurement of Budgetary Carlos Elbirt August 1990 T. Gean Operations in Highly Distorted 34247 Economies: The Case of Angola WPS498 The Build, Operate, and Transfer Mark Augenblick August 1990 D. Schein ('BOT') Approach to Infrastructure B. Scott Custer, Jr. 70291 Projects in Developing Countries WPS499 Taxing Foreign Income in Capital- Chad Leechor September 1990 A. Bhalla Importing Countries: Thailand's Jack M. Mintz 37699 Perspective WPS500 Projecting Fertility for All Countries Eduard Bos September 1990 V. Altfeld Rodolfo A. Bulatao 31091 WPS501 Tax Systems in the Reforming Cheryl W. Gray September 1990 L. Lockyear Socialist Economies of Europe 36969 WPS502 Patents and Pharmaceutical Drugs: Julio Nogu6s September 1990 M. T. Sanchez Understanding the Pressures on 33731 Developing Countries WPS503 Household Production, Time John Dagsvik September 1990 M. Abundo Allocation, and Wehare in Peru Rolf Aaberge 36820 WPS504 Applying Tax Policy Models in Henrik Dahl September 1990 A. Bhalla Country Economic Work: Pradeep Mitra 37699 Bangladesh, China, and India WPS505 Creating the Reform-Resistant Arye L. Hillman September 1990 CECSE Staff Dependent Economy: The CMEA Adi Schnytzer 37176 International Trading Relationship WPS506 Changes in Food Consumption Merlinda D. Ingco September 1990 A. Daruwala Patterns in the Republic of Korea 33713 WPS507 Poverty in Poland, Hungary, and Branko Milanovic September 1990 A. Bretana Yugoslavia in the Years of Crisis, 37176 1978-87 WPS508 A RMSM-X Model for Chile Luis Serven September 1990 S. Jonnakuty 39074 WPSSO9 The Childbeanng Family in Odile Frank September 1990 B. Rosa Sub-Saharan Africa: Structure, 33751 Fertility, and the Future PRE Working Pager Seriea Contact Tidi AtAhor Da for paper WPS510 Public Expenditure Reviews for Antoine Schwartz October 1990 C. Cristobal Education: The Bank's Experience Gail Stevenson 33640 WPS511 The Macroeconomic Underpinnings Fred Jaspersen October 1990 A. Oropesa of Adjustment Lending Karim Shariff 39075 WPS512 WPS513 WPS514 Restrictive Labor Practices in Alan S. Harding October 1990 A. Joseph Seaports 33743