Although initial internal and external conditions varied considerably across the eight countries, in most of them the need for adjustment stemmed from an excess of domestic demand over output that had led to an unsustainable external imbalance. So the task before policymakers was to bring domestic absorption down to more sustainable levels in a way that minimized contractionary effects on output, and, at the same time, to foster conditions for raising long-term growth. This section highlights several broad themes of stabilization policies pursued by these countries, drawing where appropriate upon particular country examples, as well as the econometric analysis of private investment behavior discussed in Section V.
The output response to stabilization policies depends on a complex set of interactions among a variety of factors, including the mix of policy measures, structural features of the economy, the conditions prevailing prior to adjustment, and the impact of exogenous shocks. The three main macroeconomic policy tools used to achieve stabilization have been fiscal policy, exchange rate policy, and monetary or credit policy. These policies influence output and long-term growth through several channels: by acting on the level of aggregate demand (expenditure reduction); by influencing the ease with which demand and supply shift between the tradable and nontradable goods sectors (resource switching); and by affecting intertemporal consumption choices, that is, the composition of demand between investment and consumption.
Within this broad framework, this section centers on three main themes: how the timeliness of policy responses to emerging imbalances can greatly influence the initial impact of stabilization policies on output and investment; the importance of the consistency and sustainability of policies; and the need to take careful account of key structural rigidities in designing macroeconomic policies.
Costs of Delayed Adjustment
In most of the eight countries, adjustment had to be implemented in the context of some form of macroeconomic crisis. Thus, in all of them except Thailand and Bangladesh, the authorities were faced with uncontrolled inflation, or an abrupt deterioration in access to foreign financing, or both (see Table 3 and Chart 5). The experience of the eight countries suggests that stabilization policies undertaken because of a macroeconomic crisis will generally have more substantial, and more prolonged, contractionary effects than when policies are implemented in a more timely manner. This arises not simply because the size of initial imbalances requiring adjustment is larger, but—what is equally important—because crisis tends to spawn second-best policies and sap the private sector’s confidence in government actions.
Abrupt Reduction in Real Domestic Absorption
Delays in adjustment that led to the buildup of large initial imbalances, the depletion of foreign exchange reserves, and the emergence of severe external financing constraints were often associated with a more abrupt reduction in real domestic absorption. This can be illustrated by comparing the magnitude of corrections to domestic absorption and the contemporaneous slowdown in output growth—as measured by output gaps (Table 6).22 The largest reductions in domestic demand generally occurred in countries that had the greatest initial internal and external imbalances (notably Chile and Mexico) or where existing macroeconomic problems made economies especially vulnerable to adverse external shocks. The latter were particularly large in several cases—for example, Chile I and Ghana prior to the Economic Recovery Program, as well as in Thailand (see Chart 6). For most of the countries with severe initial macroeconomic imbalances, there were few realistic alternatives to rapid and substantial stabilization; such cases were usually associated with the largest widening of output gaps and the most prolonged subsequent recessions—unless, as in Morocco, favorable supply shocks intervened.
Changes in Domestic Absorption and Output Gaps During Particular Contractionary Episodes
(Cumulative changes in percent over indicated period; in constant prices, unless indicated otherwise)
Excluding transfers.
Excluding grants. Positive change indicates a decrease in the deficit or an increase in the surplus.
Based on the contemporaneous rate of inflation of the GDP deflator. For Bangladesh, Morocco, and Senegal, includes credit to public enterprises.
The change in nonagricultural GDP is given in parentheses, to exclude the impact of weather-related supply shocks.
Changes in Domestic Absorption and Output Gaps During Particular Contractionary Episodes
(Cumulative changes in percent over indicated period; in constant prices, unless indicated otherwise)
| Initial External Current Account Balance (In percent Of GDP)1 | Change in Domestic Absorption (In percent of GDP) | Change in Estimated Output Gap (In percent of trend GDP) | Change in Real GDP | Number of Years for Output to Return to Previous Peak | Change in Primary Fiscal Balance2 | Sector Credit Growth3 (Annual average) | ||
|---|---|---|---|---|---|---|---|---|
| Most severe initial macroeconomic problems | ||||||||
| Chile I(1974–76) | −3 | −14 | −13 | −10 | 4 | 8 | 55 | |
| Chile II (1981–83) | −14 | −20 | −24 | −16 | 6 | −4 | 16 | |
| Ghana (1980–82) | — | −8 | −13 | −10 | 6 | 3 | −10 | |
| India (1990/91–1992/93) | −3 | −7 | −4 | 9 | — | 2 | −1 | |
| Mexico (1981–83) | −7 | −13 | −11 | −5 | 4 | 12 | −3 | |
| Morocco (1981–83) | −10 | −8 | 1 | 9(7)4 | — | 3 | 8 | |
| Other countries | ||||||||
| Bangladesh (1979/80–1981/82) | −6 | −4 | −1 | 5 | — | 1 | 14 | |
| Thailand (1984–86) | −5 | −8 | −6 | 10 | — | — | 9 | |
Excluding transfers.
Excluding grants. Positive change indicates a decrease in the deficit or an increase in the surplus.
Based on the contemporaneous rate of inflation of the GDP deflator. For Bangladesh, Morocco, and Senegal, includes credit to public enterprises.
The change in nonagricultural GDP is given in parentheses, to exclude the impact of weather-related supply shocks.
Changes in Domestic Absorption and Output Gaps During Particular Contractionary Episodes
(Cumulative changes in percent over indicated period; in constant prices, unless indicated otherwise)
| Initial External Current Account Balance (In percent Of GDP)1 | Change in Domestic Absorption (In percent of GDP) | Change in Estimated Output Gap (In percent of trend GDP) | Change in Real GDP | Number of Years for Output to Return to Previous Peak | Change in Primary Fiscal Balance2 | Sector Credit Growth3 (Annual average) | ||
|---|---|---|---|---|---|---|---|---|
| Most severe initial macroeconomic problems | ||||||||
| Chile I(1974–76) | −3 | −14 | −13 | −10 | 4 | 8 | 55 | |
| Chile II (1981–83) | −14 | −20 | −24 | −16 | 6 | −4 | 16 | |
| Ghana (1980–82) | — | −8 | −13 | −10 | 6 | 3 | −10 | |
| India (1990/91–1992/93) | −3 | −7 | −4 | 9 | — | 2 | −1 | |
| Mexico (1981–83) | −7 | −13 | −11 | −5 | 4 | 12 | −3 | |
| Morocco (1981–83) | −10 | −8 | 1 | 9(7)4 | — | 3 | 8 | |
| Other countries | ||||||||
| Bangladesh (1979/80–1981/82) | −6 | −4 | −1 | 5 | — | 1 | 14 | |
| Thailand (1984–86) | −5 | −8 | −6 | 10 | — | — | 9 | |
Excluding transfers.
Excluding grants. Positive change indicates a decrease in the deficit or an increase in the surplus.
Based on the contemporaneous rate of inflation of the GDP deflator. For Bangladesh, Morocco, and Senegal, includes credit to public enterprises.
The change in nonagricultural GDP is given in parentheses, to exclude the impact of weather-related supply shocks.
But initial conditions and policy delays are not the only reasons for contraction. Even countries that are generally regarded as having acted in a timely manner, such as Thailand, did not escape fully from contractionary influences: although not weak by international standards, per capita growth slowed to the lowest rate in a decade during 1985-86, and the lag in the response of output and investment to adjustment policies was of concern to the authorities at the time.
Second-Best Crisis Measures
Policies undertaken in crisis conditions sometimes included ad hoc or emergency measures that were not necessarily conducive to investment and growth. Fiscal adjustment tended to be greatest in cases where the initial imbalances were largest—particularly Chile I, Mexico, and Morocco, and where loss of access to external financing forced an especially abrupt adjustment (Chart 10).23 It is difficult to establish precise links between particular fiscal consolidation measures and growth. The reduction in public investment in these countries reversed previous booms and may well have eliminated many projects with low rates of return. However, the discussion in the companion study on the composition of fiscal adjustment does illustrate that fiscal adjustment in such circumstances was usually not based, at least initially, on well-designed changes in tax and expenditure systems of the type thought to be conducive to growth:
Fiscal Indicators1
(In percent of GDP)
Source: IMF staff estimates.1 t is the initial year (given in parentheses) of adjustment period.2 Excluding grants, except Mexico, India and Bangladesh. No data on primary deficit for Bangladesh.3 At constant prices; for Chile and Senegal excludes public enterprise investment.4 t–l is the average of 1970–73.5 Excluding central bank losses.-
Often there is a sequencing problem, where deficiencies in public expenditure management and tax administration initially require a rather crude approach to expenditure restraint and prevent the speedy reform of the tax system. Some expenditure cuts can, however, be damaging, if not reversed in due course. To facilitate their reversal and to improve expenditure allocation, immediate action to reform both tax administration and public expenditure management is necessary.24
Moreover, in a number of cases (Ghana, Morocco, Mexico, and India), initial efforts to protect the balance of payments involved the imposition or intensification of import and exchange controls, which squeezed imports of intermediate and capital goods and probably added an adverse supply shock to the contractionary effects stemming from the demand side. Also, in a few countries (for example, Mexico and in Ghana on several occasions prior to the Economic Recovery Program), measures of a confiscatory nature, such as mandatory conversion of foreign currency bank deposits on nonmarket terms, further eroded private confidence.
Uncertainty and the Effectiveness of Policies
The uncertainty and loss of credibility associated with a crisis can influence the effectiveness of particular policy measures. A deterioration in private confidence, indicated by an acceleration of capital flight, is likely to have adverse consequences for growth. Beyond this, the response of the economy to policy measures may differ in several important ways, and be harder to predict, during episodes of macroeconomic crisis. These factors help to explain why a good track record of macroeconomic policies appears to make adjustment less costly.
Lower fiscal deficits affect the composition of demand by lowering the cost and increasing the availability of credit to the private sector. But these “crowding-in” effects, which would tend to offset the direct contractionary effects of fiscal consolidation, are likely to be weaker and slower to operate in conditions of economic uncertainty. In such circumstances, the rational response of private investors would be to increase the wedge they require between the cost of capital and their expected rate of return.25 As a result, the effects of interest rate and credit policies on private sector behavior may be especially difficult to predict in the short term. In cases where adjustment was undertaken in the context of a crisis (Chile, Ghana, and Mexico), it appears that typically it took several years for credibility to be established even when substantial fiscal adjustment was undertaken up front, which contributed to the generally longer duration of the output gap. The empirical importance of such influences is hard to test, but some partial evidence is discussed in the next section.
Macroeconomic instability and heavy-handed use of government regulations prior to, and during periods of crisis, can trigger the emergence of institutional arrangements that will influence the impact of particular policy measures. For example, in countries in which a legacy of high inflation had led to widespread wage indexation (for example, Chile in the 1970s), the impact of nominal devaluations on the real exchange rate was typically greatly diminished. By contrast, a tradition of low inflation in some other countries (for example, Morocco and Thailand) allowed a more effective use of nominal depreciations to facilitate necessary changes in the real exchange rate (Chart 11).26 Similarly, attempts to reduce inflation through a tightening of monetary policy are likely to have a more substantial contractionary effect when inflation inertia is high (for example, Chile during the money-based stabilization of 1974-75).
Nominal and Real Effective Exchange Rates and Inflation in Various Indexation Regimes1
Source: IMF staff estimates.1 Rate of end-period consumer price inflation (in percent) is plotted on the right-hand scale. For Morocco, period average consumer price inflation. Nominal and real effective exchange rate indices are plotted on left-hand scale with base 1990 = 100. Increase indicates an appreciation.Less Time for Resource-Switching Policies to Take Hold
If aggregate demand declines rapidly, there is generally less time for resource-switching policies to operate sufficiently on supply, particularly of tradables, so as to minimize the contractionary effects on output. Although it is difficult to observe directly the extent of resource switching from the nontradable to the tradable goods sectors, developments in exports and imports provide some indirect indications. In most cases, the export response, although strong, typically occurred with a lag and was not sufficient to offset the initial contractionary impact on output of reductions in domestic absorption (Chart 12). Indicators of the extent and timing of resource switching in the eight countries—the strength of export and import volume growth and estimates of deviations from historical trends of the import intensity of GDP and of export market share—suggest, first, that export performance strengthened during the adjustment episodes in all countries except Senegal, but typically with a lag, so that it did not cushion the initial contractionary impact on output of reductions in domestic absorption. Second, beyond the initial period of demand squeeze, substantial increases in export volumes supported the strongest recoveries in output and investment in Chile (1985 onward), Ghana (1984 onward), and Thailand (1986 onward). Third, the intensification of import restrictions, which fell heavily on intermediate inputs, caused growth to suffer—in Ghana (1976-82), India (1991), and Mexico (1982). The longer-term export response is discussed in Section VIII, in the context of trade reform.
Changes in Domestic Absorption, Exports, and Imports During Periods of Demand Contraction
(In percentage points)
Source: IMF staff estimates.1 Compares change in domestic demand in relation to real GDP with contribution of exports over the contraction periods in Table 6.2 Compares change in real domestic demand to change in import I volumes over the contraction periods in Table 6.3 Domestic absorption excludes inventories.Sustainability and Consistency of Policies
Expectations of policy reversal may induce private investors to wait for uncertainty to be resolved, potentially locking the economy into a low-investment, low-growth equilibrium. In this context, the individual country studies suggest that two aspects of policies are likely to be especially important: progress toward a sustainable fiscal position and the consistency of macroeconomic policies.
Sustainability of Fiscal Policies
The effects of fiscal adjustment on output, investment, and saving depend on the composition of expenditure and revenue measures and the mix of financing, which in turn influences the pattern of real interest rates and private sector access to credit. However, expectations about the future course of fiscal policy are critical to the transmission process. If private agents expect the fiscal adjustment program to succeed in reducing the deficit and therefore the need for inflationary or debt financing, crowding-in effects will help to offset the initial contractionary effects and set in motion a virtuous circle of investment and growth.
Did private sector concerns about the medium-term sustainability of fiscal policy dampen the investment response during adjustment? There is no unique standard of fiscal sustainability, but one possible benchmark is the level of 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 existence of a potential credibility problem could be signaled by a path of the actual primary deficit that greatly exceeded this “sustainable” one, implying either a continued increase in the public sector debt-to-GDP ratio (and therefore a rising servicing burden) or the need for continued reliance on the inflation tax or financial repression (Chart 13).27 In the latter case, the fiscal stance would be inconsistent with a low inflation objective or financial sector reform. The consistency with financial liberalization is important because a number of countries in the study (for example, India, Mexico, and Morocco) initially financed a large part of their deficits by requiring banks and other financial institutions to hold domestic government securities at below-market interest rates.28 Therefore, the partial or complete liberalization of interest rates implemented in these countries had a substantial impact on the fiscal deficit. Details of the exercise are presented in Appendix III.
Difference Between Actual and “Sustainable” Fiscal Primary Balances1, 2
(In percent of GOP)
Source: Appendix III.1 Difference between actual primary balance and the estimated primary balance that would be consistent with no increase in the government debt-to-GDP ratio, in the context of low inflation and no financial repression. A positive figure implies that the actual balance is above the sustainable level (see Appendix III).2 Primary balance includes grants.3 Figures for Chile include the quasi-fiscal deficit of the central bank.The results suggest that Thailand’s fiscal stance was broadly sustainable from the very outset of adjustment, which probably contributed to the relatively rapid pace of the private investment response. Among the countries that began their adjustment with large or intermediate fiscal imbalances, all made some progress toward setting public debt on a more sustainable path, although in several cases (notably Morocco among those with large initial imbalances) such a position was not attained until several years into the adjustment period. Moreover, in a few countries some of the progress toward fiscal sustainability was subsequently reversed (notably in Ghana and India), which may have contributed to the delayed crowding in of private investment in these countries. Mexico is of particular interest because it achieved a decisive up-front shift to a more sustainable fiscal position, in the narrow sense used here, and maintained that position throughout the entire adjustment period; yet the crowding-in effects in the form of higher private investment were slow to materialize and were relatively weak, thereby exacerbating the initial recession. The reasons for this will be explored more fully in the next section, but the debt crisis itself—and the associated shift in net resource transfers—involved a major switch in public sector financing from external to domestic sources, which constrained the availability of financing for the private sector at least through the mid-1980s.
Judgments about the sustainability of fiscal policy based on a debt dynamics criterion however, do not necessarily imply that fiscal policy is consistent with other macroeconomic objectives, notably the avoidance of excessive external current account deficits. The linkages between the fiscal and external imbalances are difficult to quantify on an ex-ante basis, in view of possible shifts in private sector saving and investment behavior at a time of substantial changes in macroeconomic and structural policies. Consequently, the fiscal adjustment framework may need to include contingency provisions in case the private sector response diverges substantially from that expected. For example, the different fiscal policy responses to the surges in capital inflows that occurred late in the adjustment period of several of the countries illustrate that a narrow debt dynamics criterion may not be a sufficient guide to the stance of fiscal policy. For example, in Mexico, fiscal policy did not respond to the surge in capital inflows of the early 1990s and the associated sharp decline in private saving; the fiscal position (excluding privatization revenues) was, however, in surplus. In contrast, Thailand’s fiscal policy was more explicitly countercyclical, despite its comfortable public debt position. These issues are taken up again in Section VII.
Consistency of Macroeconomic Policies
Even during the transition toward a more sustainable medium-term fiscal position, an inconsistent mix of macroeconomic policies may have adverse consequences for investment and output growth. The experience of the eight countries suggests that two difficult areas are the mix of monetary and fiscal policies and the coordination of fiscal, exchange rate, and wage policies.
The Mix of Monetary and Fiscal Policies
When faced with a macroeconomic crisis and constraints on the immediate scope for fiscal action, policymakers often tend to rely excessively on contractionary monetary policies. Such tendencies are exacerbated when adjustment is delayed, and access to external financing is sharply curtailed. In such circumstances, even quite strong fiscal adjustment may involve higher public sector domestic borrowing and a consequent squeeze on private sector credit. In this section, the yardsticks used to assess the stringency of private sector credit conditions are (1) growth rates of real bank credit to the private sector, which will reflect both changes in the demand for, and the supply of, credit29 and (2) the level of real domestic interest rates compared with the real SDR interest rate, as an indicator of international financial market conditions (Chart 14).30
Real Interest Rates1
(In percent per annum)
Sources: International Monetary Fund, International Financial Statistics; and IMF staff estimates.1 National nominal rates are deflated by the contemporaneous inflation rate of the GDP deflator. The real SDR rate is calculated as the weighted average interest rate on three-month instruments in the five countries whose currencies comprise the SDR deflated by the contemporaneous annual inflation rate of the GDP deflator. The weights are the currency units used in calculating the value of the SDR.2 Real deposit (not lending) rate.If interest rates are not market determined, excessive fiscal borrowing would tend to crowd out the private sector through a direct reduction in the amount of credit allocated by the banking system. Ghana, in the period prior to the Economic Recovery Program, is such a case when access to external financing was reduced sharply (Chart 15). Although some modest fiscal adjustment took place during this time, the main burden seems to have fallen on private sector credit, which was squeezed substantially in real terms (Table 7). Another example is Mexico during 1983-87, when, despite substantial fiscal adjustment, the share of private sector credit in total credit and in GDP declined, because of portfolio constraints designed to channel a substantial proportion of bank lending to the public sector, which had virtually no access to external financing after the debt crisis. Similar influences appear to have been at work in Morocco in 1986-88. Of course, in all these cases, the high degree of macroeconomic uncertainty probably also dampened the demand for credit.
Stock of Private Sector Credit1
(In percent of total domestic credit)
Sources: International Monetary Fund, International Financial Statistics; and IMF staff estimates.1 Fiscal year data for Bangladesh and India. Private sector includes public enterprises for Bangladesh, Morocco, and Senegal.2 Break in coverage of monetary accounts in 1984.If interest rates are somewhat more flexible, public sector borrowing requirements may absorb a large share of loanable funds from domestic credit markets and exert upward pressure on real interest rates or reduce private sector access to bank credit in still segmented credit markets. This appears to have been an important factor in the early stages of the 1990-91 crisis in India, when the authorities had to rely primarily on monetary policy to restrain demand because of the political constraints on early fiscal action, which resulted in a sharp contraction in the volume of credit and a subsequent rise in interest rates. By contrast, the major fiscal consolidation in Thailand since 1987 contributed to a decline in real interest rates and the rapid rise in the share of private sector credit in total credit and in GDP.
A related issue is the problem of weak bank balance sheets and its implications for monetary policy. A high proportion of effectively nonperforming assets—although often not explicitly recognized as such—contributed to high interest margins and inefficient intermediation in a number of countries and appears to have been a major impediment to private investment in some (for example, Bangladesh). Beyond this, the continued extension of new credit to service bad loans, in a context of ineffective bank supervision, in some cases undermined the effectiveness of monetary policy in controlling aggregate demand and thereby contributed to the severity of the eventual crisis and recession. The starkest example of this is Chile prior to 1982 (see Section VIII).
Coordination of Fiscal, Exchange Rate, and Wage Policies
Perceived inconsistencies between exchange rate, wage, and fiscal policies may undermine the credibility of an adjustment program, increasing uncertainty and expectations of impending policy reversals. In such circumstances, adverse effects on investment and growth may materialize through up-ward pressures on real interest rates or on the real exchange rate. For example, given a predetermined path for the nominal exchange rate, a fiscal policy stance that fuels excess demand pressures risks leading to higher nontradable goods’ prices and thereby an appreciated real exchange rate, with potentially penalizing effects on the contribution of net exports to growth. Equally important, continued backward-looking indexation practices limit the effectiveness of discrete exchange rate changes in bringing about a reduction in real wages and improvements in external competitiveness.
The influence of inconsistent fiscal and exchange rate policies appear to have been important in Mexico (1988-89) and Senegal (post-1987). In Mexico, interest rates rose sharply in the first year of the exchange-rate-based stabilization, reflecting incomplete credibility of the exchange rate anchor and large capital outflows. These pressures appear to have originated in part because of uncertainties surrounding the fiscal policy stance prior to the 1988 elections, as well as the shift to domestic debt financing of public sector borrowing requirements discussed earlier.31 Moreover, in the later stages of the exchange-rate-based stabilization program (after 1992), the primary fiscal balance and public saving deteriorated. While it is doubtful that a strengthening of public finances sufficient to fully offset the impact of the massive capital inflows could have been feasible, a tighter fiscal stance could have helped to reduce the risks to external sustainability posed by the appreciation of the real exchange rate.
In Senegal, concern about the feasibility of the adjustment strategy led to a rise in the premiums on domestic over French franc interest rates after 1987. With a large persistent decline in the terms of trade, fiscal and public sector wage policy (especially in the early 1990s when the fiscal deficit increased and arrears were accumulated) became inconsistent with the fixed parity between the CF A franc and the French franc.32
Key Macroeconomic Indicators and Summary of Exchange Rate Policy
(In percent, unless otherwise specifiedt refers to year in parentheses)
Excluding grants.
Average annual change over the three-year period (t through t + 2); deflated by GDP deflator. For Bangladesh, Morocco, and Senegal, includes credit to public enterprises.
Percentage change in real effective exchange rate between (t – 1) and (t + 2). A minus sign indicates a depreciation.
Revisions to national income accounts imply that ratios to GDP prior to 1975 may not be strictly comparable to later estimates.
Fiscal deficits excluding central bank losses.
Key Macroeconomic Indicators and Summary of Exchange Rate Policy
(In percent, unless otherwise specifiedt refers to year in parentheses)
| Fiscal Deficit/GDP1 | Primary Fiscal Deficit/GDP1 | Average Change in Real Credit to Private Sector2 | Change in Real Effective Exchange Rate3 | ||||
|---|---|---|---|---|---|---|---|
| t – 1 | t + 2 | t – 1 | t + 2 | Exchange Rate Policy | |||
| Bangladesh (1980/81) | −12 | −11 | … | … | 12 | −15 | Adjustable peg. Dual exchange rate in effect. |
| Bangladesh (1985/86) | −7 | −7 | … | … | 5 | −19 | Adjustable peg. Dual exchange rate unified in 1992. |
| Chile (1974) | −254 | — | … | 2 | 55 | — | Floating during 1974–77. During 1978–82, exchangerate-based nominal anchor strategy, with a preannounced crawling rate at less than rate of inflation, followed by peg in 1980–82. |
| Chile5 (1983) | −2 | −2 | −1 | — | −3 | −36 | Substantial depreciation in 1983–84. Crawling band introduced in 1985, geared to ward maintaining export competitiveness. |
| Ghana (1983) | −5 | −3 | −4 | −2 | 30 | −81 | Series of large discrete devaluations to eliminate massive overvaluation. Parallel market premium reduced from 700 percent to 100 percent in 1986. |
| Ghana (1987) | −4 | −4 | −2 | −2 | −12 | −30 | Dual exchange rate system during 1986–90, with gradual Unification since 1988. Managed floating since 1990. |
| India (1991/92) | −8 | −8 | −5 | −3 | — | −24 | Peg with 19 percent devaluation in mid–1991. Managed floating since March 1993, with central bank intervention to dampen appreciation. |
| Mexico (1983) | −15 | −8 | −3 | 4 | −5 | 5 | Series of large discrete devaluations followed after 1986 by managed floating geared toward a real depreciation. |
| Mexico (1988) | −14 | −2 | 6 | 8 | 16 | 36 | Exchange-rate-based nominal anchor strategy: initial pegging (1988), followed by preannounced crawling rate at less than Inflation rate. In November 1981, greater, albeit limited, flexibility with progressively wider intervention band. Floating rate after December 1994. |
| Morocco (1981) | −10 | −12 | −8 | −7 | 5 | −16 | Adjustable peg with a series of devaluations through 1986 aimed at improving competitiveness. |
| Morocco (1986) | −10 | −5 | −3 | 2 | −1 | −11 | Managed floating during late 1980s. Peg to a basket of currencies in 1990, followingan initial devaluation. Small one-step devaluation in 1992. |
| Senegal (1985) | −6 | −2 | −2 | — | 1 | 12 | Peg (CFA franc zone). |
| Senegal (1989) | −2 | — | 1 | 3 | −1 | −12 | Peg (CFA franc zone); 50 percent devaluation in early 1994. |
| Thailand (1984) | −5 | −5 | −2 | −2 | 9 | −22 | Adjustable peg with devaluations in 1981 (9 percent) and 1984 (15 percent). |
| Thailand (1987) | −5 | 3 | −2 | 5 | 21 | −7 | Peg to an undisclosed basket of currencies. Weight of U.S. dollar increased in basket in late 1980s, resulting in nominal depreciation vis-a-vismany trading partners. |
Excluding grants.
Average annual change over the three-year period (t through t + 2); deflated by GDP deflator. For Bangladesh, Morocco, and Senegal, includes credit to public enterprises.
Percentage change in real effective exchange rate between (t – 1) and (t + 2). A minus sign indicates a depreciation.
Revisions to national income accounts imply that ratios to GDP prior to 1975 may not be strictly comparable to later estimates.
Fiscal deficits excluding central bank losses.
Key Macroeconomic Indicators and Summary of Exchange Rate Policy
(In percent, unless otherwise specifiedt refers to year in parentheses)
| Fiscal Deficit/GDP1 | Primary Fiscal Deficit/GDP1 | Average Change in Real Credit to Private Sector2 | Change in Real Effective Exchange Rate3 | ||||
|---|---|---|---|---|---|---|---|
| t – 1 | t + 2 | t – 1 | t + 2 | Exchange Rate Policy | |||
| Bangladesh (1980/81) | −12 | −11 | … | … | 12 | −15 | Adjustable peg. Dual exchange rate in effect. |
| Bangladesh (1985/86) | −7 | −7 | … | … | 5 | −19 | Adjustable peg. Dual exchange rate unified in 1992. |
| Chile (1974) | −254 | — | … | 2 | 55 | — | Floating during 1974–77. During 1978–82, exchangerate-based nominal anchor strategy, with a preannounced crawling rate at less than rate of inflation, followed by peg in 1980–82. |
| Chile5 (1983) | −2 | −2 | −1 | — | −3 | −36 | Substantial depreciation in 1983–84. Crawling band introduced in 1985, geared to ward maintaining export competitiveness. |
| Ghana (1983) | −5 | −3 | −4 | −2 | 30 | −81 | Series of large discrete devaluations to eliminate massive overvaluation. Parallel market premium reduced from 700 percent to 100 percent in 1986. |
| Ghana (1987) | −4 | −4 | −2 | −2 | −12 | −30 | Dual exchange rate system during 1986–90, with gradual Unification since 1988. Managed floating since 1990. |
| India (1991/92) | −8 | −8 | −5 | −3 | — | −24 | Peg with 19 percent devaluation in mid–1991. Managed floating since March 1993, with central bank intervention to dampen appreciation. |
| Mexico (1983) | −15 | −8 | −3 | 4 | −5 | 5 | Series of large discrete devaluations followed after 1986 by managed floating geared toward a real depreciation. |
| Mexico (1988) | −14 | −2 | 6 | 8 | 16 | 36 | Exchange-rate-based nominal anchor strategy: initial pegging (1988), followed by preannounced crawling rate at less than Inflation rate. In November 1981, greater, albeit limited, flexibility with progressively wider intervention band. Floating rate after December 1994. |
| Morocco (1981) | −10 | −12 | −8 | −7 | 5 | −16 | Adjustable peg with a series of devaluations through 1986 aimed at improving competitiveness. |
| Morocco (1986) | −10 | −5 | −3 | 2 | −1 | −11 | Managed floating during late 1980s. Peg to a basket of currencies in 1990, followingan initial devaluation. Small one-step devaluation in 1992. |
| Senegal (1985) | −6 | −2 | −2 | — | 1 | 12 | Peg (CFA franc zone). |
| Senegal (1989) | −2 | — | 1 | 3 | −1 | −12 | Peg (CFA franc zone); 50 percent devaluation in early 1994. |
| Thailand (1984) | −5 | −5 | −2 | −2 | 9 | −22 | Adjustable peg with devaluations in 1981 (9 percent) and 1984 (15 percent). |
| Thailand (1987) | −5 | 3 | −2 | 5 | 21 | −7 | Peg to an undisclosed basket of currencies. Weight of U.S. dollar increased in basket in late 1980s, resulting in nominal depreciation vis-a-vismany trading partners. |
Excluding grants.
Average annual change over the three-year period (t through t + 2); deflated by GDP deflator. For Bangladesh, Morocco, and Senegal, includes credit to public enterprises.
Percentage change in real effective exchange rate between (t – 1) and (t + 2). A minus sign indicates a depreciation.
Revisions to national income accounts imply that ratios to GDP prior to 1975 may not be strictly comparable to later estimates.
Fiscal deficits excluding central bank losses.
Labor market and fiscal policies were also crucial in determining the trade-off in exchange-rate-based disinflation programs between inflation reduction and a loss of external competitiveness; both sides of this trade-off have potential implications for growth.33 This is illustrated by the examples of Mexico (post-1988) and Chile (1978-82). In Mexico, the exchange rate anchor was supported by successive forward-looking wage-price “pacts” agreed upon with the private sector, which succeeded, at least partially, in breaking inflation inertia.34 However, the real exchange rate did subsequently appreciate, under pressure from capital inflows and labor market policies that provided inadequate support for the exchange rate anchor (see also Section VIII). By contrast, in Chile (1978-82) there was no attempt to change the existing system of full backward-looking wage indexation during the preannounced nominal exchange rate period. This inconsistency contributed to excess demand pressures and a real exchange rate
In contrast, the potential benefits of coordinated exchange rate, fiscal, and labor market policies for providing adequate incentives to the tradable goods sector and eliciting a strong resource-switching response are well illustrated by the later adjustment episode in Chile and by Thailand. Chile’s strong recovery after 1984 owed much to the sustained real depreciation of the exchange rate that underpinned a sharp increase in the volume of exports other than of copper. The success with which nominal exchange rate depreciations were translated into a significant real depreciation reflected the suspension of compulsory wage indexation in 1982 and the large fiscal retrenchment. The persistence of considerable un-employment at least until 1987 was also an important factor in holding real wages in check. In Thai-land, depreciation of the nominal exchange rate contributed to a large sustained real depreciation after 1984 and induced large export market gains. The magnitude of the real depreciation closely matched the size of the nominal devaluation reflecting the flexibility of Thailand’s labor market, a tradition of relatively low inflation, and stable macroeconomic policies that kept inflationary expectations in check. The achievement of sizable fiscal surpluses helped contain inflationary pressures when economic activity surged and offset upward pressures on the exchange rate stemming from large capital inflows.
