X Conclusions and Lessons for Program Design
- Kalpana Kochhar, Erik Offerdal, Louis Dicks-Mireaux, Mauro Mecagni, Jianping Zhou, Balázs Horváth, David Goldsbrough, and Sharmini Coorey
- Published Date:
- August 1996
Anumber of questions stem from the central issue of why the response of investment and growth to adjustment policies was often slow. This section poses those questions, summarizes the principal messages emerging from the country studies, and outlines the lessons for the design of adjustment policies. These messages not only are relevant for IMF-supported adjustment programs but also point to issues that are central to the IMF’s surveillance role. In discussing these questions, however, the limitations of the methodological approaches that are utilized, which are discussed in more detail in the introduction, should be recognized. In particular, while the use of case studies provides an opportunity for a more in-depth examination of the effects of policies, it is difficult to establish the robustness and generality of conclusions based on evidence from only a small group of countries.
The experience of the eight countries supports the broad consensus that follows and that is also derived from a wide variety of other studies that are summarized in Appendix I: (1) the objectives of a sustainable external position, moderate inflation, and faster economic growth are likely to be mutually reinforcing over the medium term; and (2) getting certain fundamental policies right—prudent fiscal and monetary management, creating effective and secure financial systems, investing heavily in human capital (especially basic education), limiting domestic price distortions (especially to ensure that the domestic relative prices of traded goods remain close to international levels), and promoting flexible labor markets—builds the foundations for, but does not guarantee, sustained growth.
Going beyond this broad consensus to consider the problems of a transition to a policy environment conducive to growth, two central themes emerge. First, because of the forward-looking nature of investment and saving decisions, market assessments of the internal consistency and sustainability of policies greatly influence the size and speed with which private investment and saving respond to adjustment measures. Consequently, the indirect effects of second-best policy choices—through reduced confidence inducing investors to wait and see or through disruptions to intertemporal consumption choices because new policies are expected to be temporary—may be greater, although harder to predict, than their direct macroeconomic effects. Second, in each country there appear to be close links between particular aspects of macroeconomic and structural reforms that are likely to be mutually supporting. Therefore, the notion of a “critical mass” of reforms does appear to be important, even though the key linkages are likely to be country-specific and difficult to establish robustly. This judgment is reinforced by one of the conclusions of the companion fiscal study, which emphasizes that systemic reforms of the tax and public expenditure management structures are likely to be essential components of a growth-oriented fiscal policy, whatever the pattern of expenditure reduction or tax increases initially adopted as part of the stabilization effort.99
Were the episodes of sharp declines in investment and growth and the subsequent slow recovery inevitable?
The severity of the decline in output and investment as well as the speed of recovery depended upon how promptly adjustment was initiated. No country in the study that postponed adjustment until faced with severe macroeconomic instability and external crisis avoided a substantial decline in investment and a slowdown in growth. Indeed, the evidence from the case studies and from cross-country analysis suggests that periods of severe macroeconomic instability are especially harmful to investment and growth and that some of these harmful effects are likely to linger even after adjustment begins. The reasons, not all of which are easily quantifiable, appear to include the rational “wait and see” attitude of private investors when faced with increased uncertainty and the difficulty of establishing stable relative price signals, including those for the exchange rate and interest rates. Policies undertaken in crisis conditions sometimes included ad hoc or emergency measures that were not conducive to growth. Therefore, in countries in which a withdrawal of external financing, and a consequent sharp turnaround in the direction of net resource transfers, forced an abrupt contraction in domestic absorption, the impact fell heavily on investment, both private and public, and typically gave insufficient time for resources to switch from the nontradable to tradable goods sectors. Private investment generally fell more steeply in such episodes than can be explained by traditional macroeconomic determinants.
Moreover, the case studies generally support the judgment that the response of the economy to policy measures may differ, and be harder to predict, during episodes of macroeconomic crisis; for example, “crowding-in” effects, which would tend to offset the direct contractionary impact of fiscal consolidation, are likely to be weaker and slower to operate in conditions of economic uncertainty. In these circumstances, policies that minimize the damage to growth are likely to be those that begin an early restoration of internal and external stability and avoid actions—such as a widespread intensification of controls or measures of a confiscatory nature—that further undermine private sector confidence. Typically, the recovery in investment following such a crisis took several years; such investment “pauses” have also been identified in a number of other studies, discussed in Appendix I. The contrasting experience of Thailand, which also encountered substantial external shocks prior to its adjustment, reinforces the view that a strategy of early corrective action to emerging macroeconomic imbalances enhances growth.
Two important lessons for IMF operations stem from these observations.
• As an economy emerges from such a crisis, there are limits to what adjustment policies can achieve in the short term. Evidence on IMF-supported programs in the eight countries suggests some tendency to overestimate the speed of private investment recovery—although unanticipated exogenous factors and policy slippages probably also played a role.
• Since a good “track record” of macroeconomic adjustment appears to make adjustment less costly, the role of surveillance is central to the IMF’s contribution to improving growth prospects—through detecting, and encouraging authorities to address, imbalances at an early stage.
How did macroeconomic policies enhance or impede the recovery in investment and growth?
The individual country studies suggest that the sustainability and consistency of policies are likely to be central in preventing expectations of policy reversal from potentially locking an economy into a low-investment, low-growth equilibrium. In a number of cases, potential inconsistencies between different components of the adjustment effort (for example, between exchange rate, fiscal, and wage policies, or between the goals of fiscal consolidation and certain structural reforms) appear to have contributed to weakening the supply response and sometimes led to policy reversals.
The fiscal position is usually critical. The fiscal adjustment has to be strong enough both to minimize the burden placed on interest rates and private credit, and hence private investment, and also to support any needed adjustment in the real exchange rate. However, expectations about the future course of fiscal policy are likely to be critical to the transmission process. The importance of considering the budget in a multiperiod framework is also supported by the results of the companion study on the composition of fiscal adjustment, since certain reforms may entail a trade-off between deficit reduction now and deficit reduction in the future.100 One possible benchmark of fiscal sustainability that is considered in the present study is the primary balance that would be consistent with maintaining a constant public debt-to-GDP ratio in the context of low inflation and no financial repression. The results suggest that judgments about the sustainability of fiscal policy based on such a debt dynamics criterion do not necessarily imply that fiscal policy is consistent with other macroeconomic objectives, notably for growth and the external current account. In particular, the case studies suggest that the linkages between fiscal and external imbalances can be difficult to quantify in view of possible changes in private sector saving and investment behavior at a time of substantial changes in macroeconomic and structural policies.
In most of the eight countries, the evidence from econometric estimates of the determinants of investment suggests that efforts to rationalize public investment programs did not, on balance, discourage private investment. Indeed, withdrawal from sectors where public investment competed with private sector opportunities often appear to have generated a positive investment response. However, this does not mean that public investment in basic infrastructure, health, and education was not crucial to growth. The findings also suggest that ensuring an adequate flow of credit to the private sector, including through the removal of credit constraints, is likely to promote a faster recovery of investment.
The country case studies also reaffirm the message emerging from previous empirical work that raising public saving is likely to be the most effective means of raising total saving in the economy.
Policies to promote resource switching are necessary to minimize the initial output contraction following stabilization and to provide adequate incentives for investment in export and import-competing activities. The speed with which such switching occurs depends upon the coordination of exchange rate, fiscal, and labor market policies and their impact on the real exchange rate. The experiences of Chile (in 1978-82) and Mexico (after 1988) suggest that consistency among policies is especially important in influencing the trade-off with external competitiveness when the exchange rate is used as the main nominal anchor in a rapid disinflation strategy. Fiscal correction and removal of backward wage indexation both appear to be important in breaking inflation inertia, but even then the short-term consequences for aggregate demand and the external current account are hard to predict, especially when trade and financial sector reforms are under way at the same time. The likelihood of an unsustainable real appreciation and consumption boom, followed by a late recession, can be reduced if fiscal policy remains geared toward controlling demand pressures and the anchoring strategy gives way to greater exchange rate flexibility, perhaps in the form of a one-time devaluation, once disinflation objectives have largely been achieved.
Lessons for the design of IMF-supported programs include the following:
• Programs should contain an explicit evaluation of fiscal sustainability in the context of a forward-looking medium-term framework, and such assessments need to take account of private sector responses to the adjustment effort; since these responses cannot be predicted with accuracy, the trade-offs involved need to be kept under close review.
• More generally, a recurrent theme is the difficulty of predicting accurately the response of some key macroeconomic variables, such as private saving and investment and the external current account, to adjustment policies. This underscores the importance of program reviews, and of surveillance activities during and after the program, not only to assess whether announced policies are on track but also to provide the opportunity for a more fundamental reappraisal of whether assumptions on key relationships underlying the macroeconomic framework remain valid.
• As for saving, some IMF-supported programs in the eight countries appear to have taken insufficient account of the partial offset in private saving that typically occurs when public saving is raised and therefore tended to be overoptimistic about saving developments. Program projections for private saving should take careful account of the available empirical evidence on the determinants of saving in each country, but such projections are inherently subject to wide margins of error at times of considerable structural change. Consequently, the medium-term framework should be revamped if the outcome for private saving is significantly lower than assumed; in these circumstances, a larger increase in public saving may need to be targeted to achieve the authorities’ growth and external objectives.
• The mix of fiscal and monetary policies should ensure an adequate supply of credit to the private sector since this is crucial to a rapid investment response. Moreover, consistency of fiscal policies and disinflation objectives is necessary to reduce uncertainty, irrespective of the nominal anchor being used. Although these issues are already a central concern of IMF-supported programs, it would be useful to address more explicitly in program documents the assumed links between credit availability, interest rates, and private investment that underlie program design. Furthermore, program design needs to deal explicitly with the problems of weak bank portfolios.
How did the design of structural reforms improve or impede growth prospects?
Although the complex nature of structural reforms makes it difficult to establish precise quantitative links with productivity and investment, the weight of evidence from the case studies supports the judgment that the elimination of major distortions fosters higher growth. It is not possible to extract from the case studies a single comprehensive blueprint on the design and sequencing of structural reforms; indeed, in countries where the supply response was strongest governments did not always follow what are generally regarded as “best practices” (for example, in trade and financial sector reforms). Nevertheless, in each country there were particular aspects of structural reform where strong complementarities existed. This suggests that implementation of an internally consistent core set of reforms is likely to minimize the risks of policy reversals and, by signaling their credibility and sustainability, to maximize the beneficial impact on growth. However, the nature of these critical links can vary from country to country, so program design needs to take careful account of individual country circumstances.
Eliciting a strong supply response to reforms also depends importantly upon overall consistency with supporting financial, exchange rate, and wage policies. For example, even in Chile, where reforms began early and were the most wide-ranging, the full benefits for growth were delayed by policies that contributed to an unsustainable exchange rate appreciation and external current account deficits during the early 1980s.
The “core” set of reforms is inevitably country specific, but key complementarities that emerged in a number of the country cases include the following:
• Trade reforms that were accompanied by broader tax reforms were more likely to be sustained. Where quantitative trade and exchange restrictions were extensive, their early removal appears to have helped the supply response as well as easing the potential trade-off between fiscal consolidation and tariff reduction.
• Problems with the quality of bank balance sheets appear to have been a major impediment to the response of private investment in a number of cases—because of the consequent high intermediation costs—and sometimes gave rise to severe financial crises that exacerbated recessions. For these reasons, early attention to strengthening prudential regulation and supervision is crucial. Difficulties in effectively enforcing an announced hardening of budget constraints on public enterprises was a major cause of the problem of nonperforming loans (for example, in Bangladesh, Ghana, and Senegal), which suggests that efforts to restructure and recapitalize the banking system are unlikely to yield their full benefits unless accompanied by effective public enterprise reform and privatization.101
• In general, the design of structural policies should emphasize early technical preparation and implementation. Certain broader structural reforms need to be accompanied by and closely coordinated with sector-specific reforms. Important examples of such linkages are (1) agricultural pricing and marketing reforms to complement trade reform, and (2) establishment of a clear legal and institutional framework to accompany broader deregulation of the economy and a scaling back of the public enterprise sector. In a number of cases, insufficient emphasis on or delays in implementing sectoral-level and institutional measures appear to have dampened the supply response to structural reforms. Design and implementation of these reforms generally took a long time and often lagged behind the broader reforms, which suggests that they should be initiated early in the adjustment process. This evidence also underscores the importance of close coordination between the IMF and the World Bank on structural issues.
How has the structure of labor markets influenced the response of output and employment to adjustment?
Although the quality of information on labor market developments in the eight countries varies substantially, their experience generally suggests that the degree of flexibility in labor markets can have a major influence on the speed and size of the output and employment response to policies.
With the exception of Thailand, all of the countries began the adjustment process with significant rigidities involving, to varying degrees, one or more of the following features: segmented labor markets with regulated formal sectors that limit wage flexibility and discourage the reallocation of labor between the formal and informal sectors; the extensive role of the public sector in employment and wage-setting in the formal sector; and a high degree of wage indexation (most notably Chile until 1982). However, with the marked exception of Chile, features of labor markets in the eight countries have been very difficult to change. Moreover, in a few other countries (Ghana and Senegal) in which labor market reform was identified as a major element of adjustment, actions typically fell well short of expectations.
Nevertheless, in a number of countries, labor market reform appears to offer considerable potential to increase the flexibility of real wages and raise the elasticity of employment with respect to output. The limited evidence available suggests a fairly strong association between output and employment in countries where the formal sector is large or labor market segmentation is limited (as in Thailand). In countries with large informal sectors, institutional rigidities have typically meant that the main burden of adjustment has fallen on wages in the informal sector—although the patchy nature of the data makes this difficult to confirm.
Wage indexation arrangements can have an important influence on the short-term path of output following the implementation of disinflation strategies. A clear example is Chile, where the exceptional severity of the 1975 and 1982-83 recessions (following money-based and exchange-rate-based stabilization efforts, respectively) appears to have been due in part to the prevalence of backward-looking wage indexation. For example, the failure to change indexation arrangements during 1978-82, when the exchange rate was used as an anchor for the price level, contributed to a sharp increase in real wages, excess demand pressures, and an unsustainable real appreciation that led eventually to a severe contraction of output and employment. By contrast, following the elimination of mandatory indexation in 1982 and the shift to a flexible exchange rate regime, real wages were lowered sharply and Chile was able to effect a large real exchange rate depreciation without reigniting inflation, thereby helping to lay the foundation for a period of sustained growth in output and employment.102
The experience of the eight countries suggests two labor market issues deserving of particular attention in the design of adjustment programs:
• Institutional arrangements affecting the extent and nature of wage indexation should be an important consideration in the design of disinflation strategies. The extent of de facto or de jure wage indexation influences the effectiveness of policy instruments such as the nominal exchange rate and should be addressed explicitly in program design, especially when wage-price inertia is high, whatever nominal anchor strategy is pursued.
• Labor market rigidities in the eight countries were generally given limited attention in program design. Their experience suggests that programs would benefit from a greater integration of such issues into the overall macroeconomic and structural framework. However, such rigidities have typically been one of the most difficult areas in which to bring about fundamental change.
How important was the availability of external financing for growth?
It has generally been difficult to identify a robust association between external financing and growth because of the endogeneity of most financing and because policies and financing interact in complex ways. Nevertheless, the partial evidence available from the country studies suggests that the timely availability of additional financing can enhance growth prospects.
Confidence factors associated with the debt crisis and the authorities’ initial policy response appear to have been important factors behind the fall in private investment in countries that encountered severe debt-servicing difficulties. In some (for example, Mexico), the large switch from external to domestic financing by the public sector that followed the debt crisis appears to have squeezed credit available for the private sector, despite substantial fiscal adjustment. Although an independent effect of debt overhang indicators on investment could not be identified for most countries, this certainly does not imply that the debt crisis and the associated sudden and large shift in net resource transfers abroad did not have a major adverse influence on investment. Rather, such effects operate primarily through other variables, including interest rates and credit availability; indeed, there is some evidence to indicate that the resolution of debt and debt-servicing issues was associated with a reduction in interest rates. The beneficial effects on output arising from a relaxation of financing constraints also appear to have been important in those cases (most notably Ghana but also India) where previous delays in adjustment had led to severe constraints on imported intermediate inputs. However, higher growth did not occur in all countries in which inflows of official or private capital picked up; the most lasting effects appear to have occurred in countries in which fiscal and structural policies favored the channeling of the additional financing toward increased investment, especially in tradables.
Although the impact of external financing is manifold, the country cases point to the following general observation:
• Provided the right supporting policies are in place, the timely availability of additional financing is likely to strengthen growth, and the beneficial effects can be especially important following periods when financing was sharply constrained.
What distinguishes countries that appear to have been successful in achieving a transition to faster growth and what lessons do they suggest?
The linkages between policies and growth are often indirect and can operate with significant and variable lags. Two countries—Chile and Thailand—appear to have made the transition to a more rapid growth path. In addition, in Ghana output recovered sharply, although investment and private saving have yet to make the transition to a path consistent with sustained rapid growth. Many factors other than policies have an impact on growth. The initial rebound in Chile partly reflected a recovery from a massive recession; Ghana’s recovery undoubtedly arose from, in part, the reversal of disastrous policies prior to adjustment; and some of Thailand’s growth performance can be attributed to the benefits of being part of a rapidly growing, dynamic region. Consequently, short-term developments in output may not be a good criterion for judging the contribution of adjustment programs to longer-term growth.
Nevertheless, it is clear from the country case studies that macroeconomic and structural policies matter for growth and that a consistent track record of stable macroeconomic policies and nondistortionary structural policies aimed at fostering the private sector as the engine of growth offers the best prospects for sustained medium-term growth. Although there is no single comprehensive blueprint that links the cases of Chile and Thailand—indeed, important aspects of their economic histories have been strikingly different over the past two decades—some common elements are noteworthy.103
• Macroeconomic stabilization was generally achieved in a manner that was internally consistent and sustainable; in particular, both countries attained fiscal deficits that were consistent with moderate inflation and a sustainable external position.
• The combination of exchange rate and domestic financial and labor market policies yielded a real exchange rate that provided adequate incentives for the tradable goods sectors. In Thailand, these two elements were present almost from the outset, while in Chile they were achieved only after painful mistakes that delayed the transition to faster growth.
• Structural reforms were successful in establishing an overall institutional framework to foster the private sector as the engine of growth and in generating consistent and stable relative price signals; in Thailand, these basic elements were present from the beginning and were consistently maintained even though additional reforms were relatively modest.
• As indicated in the companion study,104 public sector reform also seems to have been the most conducive to growth in the two countries, although their experience also suggests that the path to high-quality fiscal adjustment can take different forms: Chile achieved a radical transformation of its budgetary, pension, and public enterprise systems while managing to protect education and health spending; Thailand began its adjustment with a relatively well-oriented public sector and, although there are some indications that cuts in capital spending may have contributed to infrastructural bottlenecks, the real resources devoted to (especially primary) education and health were increased substantially.
• The flexible operation of labor markets in Thailand and the eventual, albeit difficult and costly (in terms of unemployment), achievement of greater flexibility in Chile were also important factors.
• The stable macroeconomic and regulatory environment encouraged substantial inflows of foreign capital and policies were aimed at directing the inflows to investment rather than consumption.