II The Crisis in Emerging Markets—and Other Issues in the Current Conjuncture

International Monetary Fund
Published Date:
November 1998
  • ShareShare
Show Summary Details

Since the finalization of the May 1998 World Economic Outlook, the recessions in the east Asian emerging market economies suffering from financial crisis and in Japan have deepened. And the August crisis in Russia has brought a new wave of financial market pressure to many emerging market economies, especially in Latin America, and helped to trigger a sell-off in equity markets worldwide. Economic conditions in most industrial countries except Japan, however, have remained favorable, with growth moderating in some cases and strengthening somewhat in others: the adverse effects on overall activity of developments in Asia and emerging markets elsewhere have been relatively muted. The volume of world trade has decelerated sharply in 1998. Prices of oil have remained very weak, and nonfuel commodity prices have weakened further. Reflecting these developments, world output growth this year is now projected at 2 percent—more than 1 percentage point lower than projected in May (Table 2.1). Global growth is projected to recover only moderately in 1999, and the risks appear to be predominantly on the downside.

Table 2.1.Overview of the World Economic Outlook Projections(Annual percent change unless otherwise noted)


from May 1998

World output4.
Advanced economies3.
Major industrial countries2.
United States3.
United Kingdom2.
Other advanced economies3.
Industrial countries2.
European Union1.
Euro area1.
Newly industrialized Asian economies6.36.0-2.90.7-4.7-3.8
Developing countries6.
Middle East and Europe4.
Western Hemisphere3.
Countries in transition-1.02.0-0.2-0.2-3.1-3.6
Central and eastern Europe1.
Excluding Belarus and Ukraine3.
Transcaucasus and central Asia1.
World trade volume (goods and services)
Advanced economies6.
Developing countries9.
Countries in transition10.
Advanced economies6.
Developing countries8.810.93.95.5-3.5-1.2
Countries in transition7.
Commodity prices
In SDRs23.7-0.2-29.210.4-6.41.3
In U.S. dollars18.4-5.4-31.19.3-7.20.1
In SDRs3.32.0-11.61.4-5.60.6
In U.S. dollars-1.2-3.3-13.90.4-6.5-0.5
Consumer prices
Advanced economies2.
Developing countries14.
Countries in transition41.427.929.534.615.725.9
Six-month LIBOR (in percent)3
On U.S. dollar deposits5.
On Japanese yen deposits0.
On deutsche mark deposits3.33.43.7-0.2
On Euro deposits3.7
Note: Real effective exchange rates are assumed to remain constant at the levels prevailing during July 27–August 24, 1998, except for the bilateral rates among ERM currencies, which are assumed to remain constant in nominal terms.

Simple average of spot prices of U.K. Brent, Dubai, and West Texas Intermediate crude oil. The average price of oil in U.S. dollars a barrel was $19.27 in 1997; the assumed price is $13.28 in 1998 and $14.51 in 1999.

Average, based on world commodity export weights.

London interbank offered rate.

Note: Real effective exchange rates are assumed to remain constant at the levels prevailing during July 27–August 24, 1998, except for the bilateral rates among ERM currencies, which are assumed to remain constant in nominal terms.

Simple average of spot prices of U.K. Brent, Dubai, and West Texas Intermediate crude oil. The average price of oil in U.S. dollars a barrel was $19.27 in 1997; the assumed price is $13.28 in 1998 and $14.51 in 1999.

Average, based on world commodity export weights.

London interbank offered rate.

The following sections discuss the global growth and inflation outlook, and some salient features of the background to recent developments and the projections.

Global Growth and Inflation Outlook

In Asia, while exchange rates have stabilized and, in some cases, recovered significantly from their lows, the economic downturns in the crisis countries and Japan have worsened beyond expectations early in the year, with particularly sharp declines in activity in the first half of 1998 in Indonesia, Korea, Malaysia, and Thailand. The weakness of the Japanese economy and the yen has put additional pressures on financial markets both in the crisis countries and in other emerging market economies, especially in Asia.

Growth has continued to be well sustained overall in the advanced economies of North America and Europe, but with significant changes in relative growth performance. In the United States, growth eased to below its potential rate in the second quarter of 1998 after registering its highest rate for almost two years in the first quarter, but final domestic demand remained robust. Growth has also slowed in the United Kingdom and Canada. In continental Europe, the recovery has gained momentum since early last year, underpinned by strengthening domestic demand. Recent indicators, including continued strong business and consumer confidence (Figure 2.1) and improvements in labor market conditions, suggest that near-term growth prospects in the EU remain favorable. As in North America, growth in the European advanced economies appears thus far to have been relatively little affected, on net, by the crisis in emerging markets; but in a number of cases the composition of demand has changed significantly. While external positions vis-à-vis Asia have worsened, domestic spending has been stimulated by further declines in long-term interest rates since mid-1997 and the strength until recently of equity markets, as well as improvements in the terms of trade—developments partly attributable to the Asian crisis. However, with the recent corrections in equity prices and the further weakening of net exports as a result of the continuing adjustments in Asia and emerging markets more generally, a widespread slowing of output growth is likely over the coming year.

Figure 2.1.Selected European Union Countries, Japan, and the United States: Indicators of Consumer and Business Confidence1

Consumer confidence remains very high in the United Suites and has continued to improve in France and Germany.

Sources: Consumer confidence—fur the United States, the Conference Board: for European Union countries, the European Commission. Business confidence—for the United States, the U.S. Department of Commerce, Purchasing Managers Composite Diffusion Index; for European Union countries, the European Commission; for Japan, Bank of Japan.

1Indicators are not comparable across countries.

2 Percent or respondents expecting an improvement in their situation minus percent expecting a deterioration.

Deteriorating economic conditions in Asia were a key factor behind the further falls in commodity prices in the first half of 1998. Successive agreements among oil-producing countries in the spring to cut production led to only temporary increases in oil prices, with market skepticism persisting about the effectiveness of the cuts. Prices fell to a 1998 low in the middle of June, in real terms the lowest level since 1973 (Figure 2.2). Prices rebounded somewhat following fresh pledges of production cutbacks by oil-exporting countries in late June, but declined again in August. On the basis of futures markets quotations for the rest of the year, oil prices in 1998 as a whole are projected to be about 30 percent lower, on average, than in 1997, with only a partial recovery projected for 1999. The prices of many nonfuel commodities weakened further between May and August, by around 9 percent on average, following some stabilization in early 1998, as the situation in Asia and other emerging markets deteriorated. They are now projected to be about 12 percent lower on average in 1998 than in 1997, and to show little upward movement next year. With the weakness of primary commodity prices adding to the downward pressures on the broader price level arising from enhanced competition from Asia, most of the advanced economies have moved closer to price stability (Figures 2.3 and 2.4). At the same time, commodity price declines have boosted real incomes and domestic demand in these economies. However, commodity price weakness has also lowered growth prospects of the many developing and transition economies that are mainly commodity-based.

Figure 2.2.Prices of Crude Petroleum and Nonfuel Commodities

(Quarterly average; 1990 = 100)

Prices of oil and nonfuel commodities declined further in the first half of 1998, in both nominal (U.S. dollar) and real terms.

Figure 2.3.Selected Advanced Economies: Inflation

(Annual percent change)

Inflation in the advanced economics has stabilized at low levels.

Figure 2.4.Advanced Economies: Inflation and Commodity Prices

(Percent change from a year earlier)

The decline in commodity prices since 1997 has helped lo keep inflation low in the advanced economies.

Reflecting these various developments, growth prospects have been marked down significantly for Asia, Russia, and to a lesser extent for many emerging market countries in other regions, including oil and other commodity exporters, while growth forecasts for 1998 have been revised upward for the United States, and, to a lesser extent, for some continental European countries. (Policy assumptions underlying the projections are summarized in Box 2.1, on page 26.) With the downward revisions dominating, growth in the world economy is now projected at 2 percent in 1998, 2¼ percentage points lower than projected a year ago. In 1999, global growth is projected to pick up moderately to 2½ percent; a modest turnaround of growth in Asia, including Japan, is projected to be offset by a slowing of expansions in North America and, to a lesser extent, in Europe, with the developing countries outside Asia, and the transition countries, showing little or no improvement. The impact of the emerging market crisis is the main factor contributing to the projected slowdown in world trade growth to 3¾ percent in 1998, markedly lower than the exceptionally high 1997 growth rate, with only a moderate pickup predicted in 1999. The trade growth now projected for 1998 and 1999 is similar to that experienced during 1992–93.

Among the advanced economies of Asia, there have been large downward revisions to the growth projections for Japan and also, as discussed below, for Hong Kong SAR, Korea, and Singapore (Table 2.2).

Table 2.2.Advanced Economies: Real GDP, Consumer Prices, and Unemployment Rates(Annual percent change and percent of labor force)
Real GDPConsumer PricesUnemployment Rates
Advanced Economies3.
Major industrial countries2.
United States3.
United Kingdom12.
Other advanced economies3.
Taiwan Province of China5.
Hong Kong SAR4.65.3-
New Zealand23.12.3-
European Union1.
Euro area1.

Consumer prices are based on the retail price index excluding mortgage interest.

Consumer prices excluding interest rate components; for Australia, also excluding other volatile items.

Consumer prices are based on the retail price index excluding mortgage interest.

Consumer prices excluding interest rate components; for Australia, also excluding other volatile items.

In Japan, following the modest rebound in activity in the third quarter of 1997, output declined for three consecutive quarters, with the economic situation deteriorating significantly in the first half of 1998 (see Chapter IV). Factors underlying the disappointing performance include increasing financial strains, particularly in the banking system: the large withdrawal of fiscal stimulus initiated in April 1997 (Table 2.3); the fallout from the crises in neighboring economies: and a general weakening of confidence. In response to the downturn, the government announced in April 1998 a fiscal stimulus package including tax and spending measures worth 2½ percent of GDP, mainly concentrated in the second half of the year. A further fiscal package, including tax reductions, was announced in August, and in early September the Bank of Japan cut the overnight call rate by 25 basis points to ¼ of 1 percent while leaving the discount rate unchanged. The authorities have also taken steps to address balance sheet problems in the banking sector and improve the availability of credit. Following the failure of several major financial institutions in late 1997, the Bank of Japan provided ample liquidity to ease money market conditions. In addition, in early 1998, public funds totaling around 6 percent of GDP were made available to bolster the deposit insurance system and recapitalize banks, and in July the authorities unveiled plans for restructuring the banking system (see Chapter I). The April and August fiscal stimulus packages should ensure a pickup in activity late this year and prevent a withdrawal of fiscal stimulus in 1999 as a whole. But even with the additional stimulus. GDP in Japan is now projected to fall by 2½ percent in 1998 as a whole, and to grow by just ½ of 1 percent in 1999—significantly down from the projections in the May 1998 World Economic Outlook. Moreover, this projection is subject to downside risks, particularly relating to possible further deteriorations in the banking system and in private sector confidence.

Table 2.3.Major Industrial Countries: General Government Fiscal Balances and Debt1(Percent of GDP)
Major industrial countries
Actual balance-2.9-3.8-4.2-3.5-3.4-2.7-1.2-1.1-0.90.5
Output gap-0.7-1.5-2.7-2.3-2.3-1.7-1.1-1.3-1.5-0.3
Structural balance-2.6-3.0-2.9-2.4-2.3-1.8-0.6-0.5-0.10.6
United States
Actual balance-2.8-4.4-3.6-2.3-1.9-
Output gap-1.6-3.2-3.2-2.1-2.2-
Structural balance-2.2-3.1-2.3-1.4-1.1-
Net debt37.450.352.453.353.152.950.547.742.633.1
Gross debt51.664.866.866.266.666.662.759.352.941.2
Actual balance-0.41.5-1.6-2.3-3.6-4.3-3.1-5.7-6.2-2.7
Output gap0.71.7-0.6-2.3-2.9-1.5-3.0-7.4-8.4-1.3
Structural balance-
Net debt19.
Gross debt67.870.
Actual balance excluding social security-3.5-2.0-4.8-5.1-6.5-6.8-5.2-7.3-7.8-4.4
Structural balance excluding social security-3.6-2.4-4.6-4.6-5.8-6.5-4.5-5.5-5.8-4.1
Actual balance-2.1-2.8-3.2-2.4-3.3-3.4-2.7-2.6-1.9-1.0
Output gap-1.12.0-2.0-2.2-2.5-3.2-3.1-2.6-1.7-0.2
Structural balance-1.6-4.0-2.2-1.2-2.0-1.5-0.7-1.0-0.9-0.8
Net debt21.127.735.140.548.951.552.452.652.349.1
Gross debt41.044.147.950.258.060.461.361.461.158.0
Actual balance-2.2-3.8-5.6-5.7-4.9-4.3-3.0-2.9-2.0-0.6
Output gap0.4-0.5-3.8-3.0-2.7-3.3-3.2-2.5-1.4
Structural balance-2.4-3.4-3.2-3.7-3.1-1.9-0.9-1.3-1.1-0.6
Net debt322.530.234.440.243.646.348.049.249.947.0
Gross debt30.839.245.248.352.555.457.858.959.756.8
Actual balance-11.1-9.6-9.5-9.2-7.7-6.7-2.7-2.6-1.7-1.3
Output gap0.3-0.2-2.6-2.5-1.1-2.0-2.3-2.1-0.7
Structural balance-11.3-9.5-8.2-7.9-7.1-5.7-1.6-1.7-1.3-1.2
Net debt75.5103.0112.8118.3117.6117.4115.2112.2107.798.8
Gross debt83.0108.7119.1124.9124.2124.0121.6118.5113.7104.4
United Kingdom
Actual balance-1.9-6.3-7.9-6.9-5.6-4.6-1.9-
Output gap-0.8-4.5-4.5-2.4-1.6-
Structural balance-1.3-3.7-4.4-4.3-4.2-3.6-1.8-
Net debt42.129.033.839.442.345.747.943.338.931.2
Gross debt50.136.142.548.450.553.854.551.547.039.0
Actual balance-4.2-7.5-7.3-5.3-1.0-
Output gap-0.6-5.1-3.6-1.8-1.9-2.7-1.6-0.9-0.5
Structural balance-3.8-4.0-4.6-4.0-2.9-
Net debt32.756.063.367.268.570.066.563.355.443.0
Gross debt65.388.296.998.7100.2100.596.291.781.667.0
Note: The budget projections are based on information available through September 1998. The specific assumptions for each country are set out in Box 2.1.

The output gap is actual less potential output, as a percent of potential output. Structural balances are expressed as a percent of potential output. The structural budget balance is the budgetary position that would be observed if the level of actual output coincided with potential output. Changes in the structural budget balance consequently include effects of temporary fiscal measures, the impact of fluctuations in interest rates and debt-service costs, and other noncyclical fluctuations in the budget balance. The computations of structural budget balances are based on IMF staff estimates of potential GDP and revenue and expenditure elasticities (see the October 1993 World Economic Outlook, Annex I). Net debt is defined as gross debt less financial assets of the general government, which include assets held by the social security insurance system. Debt data refer to end of year; for the United Kingdom they refer to end of March. Estimates of the output gap and of the structural budget balance are subject to significant margins of uncertainty.

Data before 1990 refer to west Germany. For net debt, the first column refers to 1986–91, Beginning in 1995, the debt and debt-service obligations of the Treuhandanstalt (and of various other agencies) were taken over by the general government. This debt is equivalent to 8 percent of GDP and the associated debt service to ½ of 1 percent of GDP.

Figure for 1981–91 is average of 1983–91.

Note: The budget projections are based on information available through September 1998. The specific assumptions for each country are set out in Box 2.1.

The output gap is actual less potential output, as a percent of potential output. Structural balances are expressed as a percent of potential output. The structural budget balance is the budgetary position that would be observed if the level of actual output coincided with potential output. Changes in the structural budget balance consequently include effects of temporary fiscal measures, the impact of fluctuations in interest rates and debt-service costs, and other noncyclical fluctuations in the budget balance. The computations of structural budget balances are based on IMF staff estimates of potential GDP and revenue and expenditure elasticities (see the October 1993 World Economic Outlook, Annex I). Net debt is defined as gross debt less financial assets of the general government, which include assets held by the social security insurance system. Debt data refer to end of year; for the United Kingdom they refer to end of March. Estimates of the output gap and of the structural budget balance are subject to significant margins of uncertainty.

Data before 1990 refer to west Germany. For net debt, the first column refers to 1986–91, Beginning in 1995, the debt and debt-service obligations of the Treuhandanstalt (and of various other agencies) were taken over by the general government. This debt is equivalent to 8 percent of GDP and the associated debt service to ½ of 1 percent of GDP.

Figure for 1981–91 is average of 1983–91.

The U.S. economy has continued to perform strongly, with real GDP accelerating to 5½ percent growth at an annual rate in the first quarter of 1998 before slowing in the second quarter to 1¾ percent growth. The widening current account deficit, attributable mainly to the Asian slump and the strength of the dollar, has been acting as an increasing drag on growth, but the expansion of final domestic demand has remained robust. Above-potential growth has further reduced the unemployment rate (to 4½ percent, close to a 28-year low) and put upward pressure on the growth of labor earnings. But core price inflation has shown little rise: it was 2½ percent at an annual rate in the first half of 1998, compared with 2¼ percent in 1997. After the slowing of the expansion in the second quarter, growth is projected to be close to its potential rate (around 2¼ percent a year) in the second half of 1998, giving growth of 3½ percent for the calendar year, compared with 3 percent projected previously. GDP growth next year is expected to remain close to potential, as domestic demand softens in response to the recent decline in equity prices and external demand remains weak. A further deterioration of economic conditions in Asia, continuing global financial turmoil, or a significant further stock market decline would tend to slow growth more than projected.

In the future euro area as a whole, economic recovery has gained momentum since early 1997, with consumption and fixed investment taking over from foreign demand and stockbuilding as the main forces sustaining the expansion. The contractionary effects of the Asian crisis on exports have been largely offset by the expansionary effects of improved terms of trade, lower long-term interest rates, and continued strong growth in many non-Asian markets. Inflation in the area has remained subdued, at around 1½ percent, and there are few signs of general inflationary pressures despite high capacity utilization in some of the smaller countries and strong asset markets. Growth is projected to reach 3 percent in 1998 and to moderate to 2¾ percent in 1999, with strong business and consumer confidence, recent declines in interest rates, and an almost uniform shift to expansionary fiscal stances following the efforts to meet the Maastricht criteria in 1997 all tending to boost activity. Among the main risks to this outlook are the possibilities of a sharper-than-projected slowdown in the United States or a further deterioration in growth prospects in Asia, Russia, and emerging market economies in other regions.

Developments in the future euro area as a whole mask important differences among member countries. In the three largest economies—Germany, France, and Italy—recoveries are still at relatively early stages, and output gaps remain significant (Figure 2.5). Indicators for 1998 show some acceleration of growth in France and Germany—albeit less certain in the latter case—as a result of upturns in investment and consumer spending (Figure 2.6), while inflation has fallen below 1 percent in both cases. The momentum of growth is projected to be fairly well-maintained in 1999 in both countries. With output gaps remaining significant and falls in commodity prices still feeding through, inflation is expected to remain subdued. In Italy, recent indicators, including modest GDP growth in the first half of 1998, indicate that the recovery is still tentative, with growth in private consumption remaining weak. The economy’s upturn is expected to strengthen, however, with the further convergence of short-term interest rates toward the euro-area average; in addition, consumer confidence remains high, business investment is picking up, and export orders have been growing strongly, although at a somewhat reduced pace recently, apparently reflecting reduced demand from Asia, In Austria and Belgium—the other two euro-area countries where the recovery lagged until 1997—growth has firmed recently, with domestic demand gaining momentum and net exports remaining strong.

Figure 2.5.Major Industrial Countries: Output Gaps1

(Actual less potential output, as percent of potential)

Output in the United States has recently exceeded its potential level, while in the other major industrial countries, except Japan, it is approaching potential.

1Shaded areas indicate IMF stuff projections. The estimates of output gaps are subject to a significant margin of uncertainty. For a discussion of approaches to calculating potential output, see Paula R. De Masi, “IMF Estimates of Potential Output: Theory and Practice,” in Staff Studies for the World Economic Outlook (Washington: IMF, December 1997), pp. 40–46.

2Data through 1991 apply to west Germany only.

Figure 2.6.Selected European Countries: Real Total Domestic Demand1

(Percent change from four quarters earlier)

Domestic demand has been strengthening in France and Germany but weakening in the United Kingdom.

1Shaded area indicates IMF staff projections.

The other euro-area countries are at more advanced stages of the cycle, with resource utilization relatively high, growth generally vigorous, and in some cases rising inflationary pressures. In Spain, the economy has been growing by 3½–4 percent at an annual rate since the middle of 1997 and is projected to continue to do so through 1999. Wage moderation is helping to keep inflation under control in Spain, and this is also the case in Finland, where growth is projected to moderate to 3½ percent in 1999. Inflation risks are particularly high in Ireland, where growth has remained the fastest in Europe, capacity constraints have been increasingly binding, the impact of the 1997 decline in the effective exchange rate is still being felt, and a significant reduction of short-term interest rates is in prospect by end-1998 with pre-EMU convergence. Inflationary pressures have also been rising in the Netherlands and Portugal, where economic growth accelerated in the first quarter of 1998 and labor markets have tightened further.

In the United Kingdom, after five years of solid expansion, growth has slowed significantly since late 1997, with a marked weakening in net exports resulting mainly from the prolonged strength of the pound sterling, and domestic demand growth abating since early 1998 as a result of past fiscal and monetary tightening. Taking into account the appreciation of sterling since late 1996 as well as the rise in short-term interest rates since May 1997, monetary conditions have tightened very sharply in the past two years (Figure 2.7). At the same time, a fiscal tightening of some 3 percent of GDP is being implemented over 1997–98 as a whole, in early June, with inflation above its 2½ percent target and labor earnings accelerating, the Bank of England raised official interest rates by 25 basis points to 7½ percent—the first change in seven months. While labor markets remain tight—with unemployment down to 4½ percent—earnings growth appears to have moderated during the summer, and inflation has edged down to meet its target. Real GDP growth is projected to moderate to 2¼ percent in 1998, broadly in line with potential, and to decline further to 1¼ percent in 1999.

Figure 2.7.Major Industrial Countries: Indices of Monetary Conditions1

Monetary conditions—reflecting changes in both interest rates and exchange rates—have tightened in the United States and, especially, in the United Kingdom.

1For each country, the index is defined as a weighted average of the percentage point change in the real short-term interest rate and the percentage change in the real effective exchange rate relative to a base period (January 1990 to December 1997). Relative weights of 3 to 1 are used for Canada, France, Italy, and the United Kingdom: 4 to 1 for Germany; and 10 to 1 for Japan and the United States. The weights are intended to represent the relative impact of interest rates and exchange rates on aggregate demand; they should be regarded as indicative rather than precise estimates. For instance, a 3-to-1 ratio indicates that a 1 percentage point change in the real short-term interest rate has about the same effect on aggregate demand over time as a 3 percent change in the real effective exchange rate. Movements in the index are thus equivalent to percentage point changes in the real interest rates. The lag with which a change in the index may be expected to affect aggregate demand depends on the extent to which the change stems from a change in the interest rate or the exchange rate, and varies depending on the cyclical position: the lag also differs across countries. No meaning is to be attached to the absolute value of the index; rather, the index is intended to show the degree of lightening or easing in monetary conditions from the (arbitrarily chosen) base period. Small changes in the relative weights may affect the value of the index but not the qualitative picture.

In Norway, currency depreciation associated with falling oil prices, together with rising inflationary pressures as a result of continued strong growth outside the oil sector and accelerating wages, prompted increases in interest rates amounting to 3¾ percentage points between late June and late August. Projected growth in 1998 has been revised down by 2¼ percentage points to 3 percent, mainly on account of weak oil prices and their effects on oil production and the terms of trade. In Denmark, where signs of overheating have emerged after five years of steady growth, a fiscal tightening was announced in June 1998, and official interest rates were raised by a full percentage point in September. These measures should help to slow growth to a more sustainable pace in 1999. In Switzerland, the recovery has only recently begun to gather pace; growth in 1998 is now projected to be somewhat higher than previously, with domestic demand strengthening while inflation should remain subdued.

In Australia, Canada, and New Zealand, exchange rate depreciations against the U.S. dollar and the European currencies, and robust growth in the United States and Europe, have moderated the contractionary effects of the downturn in exports to Asia and lower commodity prices. In Canada and Australia, buoyant domestic demand has also been cushioning the impact of the Asian crisis. In Canada, growth is projected to moderate to 3 percent in 1998 as a whole and to slow further in 1999. In Australia and New Zealand—advanced economies with particularly large trade exposure to Asia—the Asian crisis is having a more significant negative impact. Growth has slowed particularly sharply in New Zealand: in fact, activity declined in the first quarter, and the economy is projected to contract slightly in 1998 as a whole. A moderate recovery is projected for 1999, partly in response to the recent easing of monetary conditions.

In Israel, following a tightening of macroeconomic policies designed to reduce overheating pressures that emerged in 1996, growth has slowed, unemployment has increased, and inflation has declined below 5 percent. The abatement of demand pressures has allowed monetary policy to be eased recently, and financial markets have been relatively little affected by the turmoil in other emerging markets. Following the liberalization of the capital account in May, continuing structural reform, combined with additional fiscal consolidation, would establish conditions for further interest rate reductions and a resumption of sustainable growth.

Among the developing countries, in the Middle East and Europe region output growth is projected to weaken from 4¾ percent in 1997 to 2¼ percent this year but to pick up to 2¾ percent in 1999. The projections represent significant downward revisions from the May 1998 World Economic Outlook. The sharp decline in oil prices has adversely affected activity not only in the oil-producing countries in the region but also in other countries, owing to lower workers’ remittances and reduced regional demand. Reductions in fiscal outlays in the Islamic Republic of Iran, Saudi Arabia, and other oil-producing countries, while containing the deterioration of budgetary and external balances, are likely to restrain growth, and growth projections have been marked down accordingly (Table 2.4). Growth in Turkey, following a very strong first-quarter performance, is projected to slow in the second half of 1998, in part owing to the short-term impact of policies being implemented to curtail inflation. In Jordan, growth data for 1996–97 have been revised down, and growth prospects for 1998 and the medium term now seem less favorable in light of an apparently slower underlying growth trend and the limited progress with the regional peace process.

Table 2.4.Selected Developing Countries: Real GDP and Consumer Prices(Annual percent change)
Real GDPConsumer Prices
Developing countries6.65.82.314.19.110.3
Côte d’Ivoire6.
South Africa3.
SAF/ESAF countries25.84.64.615.68.67.1
CFA countries5.
Middle East and Europe4.74.72.324.622.622.6
Iran, Islamic Republic of4.
Saudi Arabia1.
Western Hemisphere3.55.12.820.813.910.8
Dominican Republic7.

The inflation figures published in the May 1998 World Economic Outlook were end-of-period data.

African countries that had arrangements, as of the end of 1997, under the IMF’s Structural Adjustment Facility (SAF) or Enhanced Structural Adjustment Facility (ESAF).

The inflation figures published in the May 1998 World Economic Outlook were end-of-period data.

African countries that had arrangements, as of the end of 1997, under the IMF’s Structural Adjustment Facility (SAF) or Enhanced Structural Adjustment Facility (ESAF).

Box 2.1.Policy Assumptions Underlying the Projections for Selected Advanced Economies

Fiscal policy assumptions for the short term are based on official budgets, adjusted for any deviations in outturns as estimated by IMF staff and also for differences in economic assumptions between IMF staff and national authorities. The assumptions for the medium term take into account future policy measures that are judged likely to be implemented. Both projections and policy assumptions are generally based on information available through September 1998. In cases where future budget intentions have not been announced with sufficient specificity to permit a judgment about the feasibility of their implementation, an unchanged structural primary balance is assumed, unless otherwise indicated. For selected advanced economies, the specific assumptions adopted are as follows (see Table 2.3, and Tables 1416 in the Statistical Appendix for the projected implications of these assumptions).

United States. The fiscal projections are based on the administration’s May 1998 Mid-Session Review of the FY1999 Budget, adjusted for differences between the IMF staff’s and the administration’s macroeconomic assumptions. State and local government fiscal balances are assumed to remain constant as a percent of GDP.

Japan. The projections take account of recent policy initiatives, including the ¥16 trillion supplementary budget announced in April 1998, and the August 1998 spending guidelines for central government spending. The projections also assume that reductions in marginal personal and corporate income tax rates worth ¥7 trillion are implemented in 1999, which is consistent with plans announced in August 1998, and that the increase in the public pension contribution rate planned for October 1999 occurs on schedule. Over the medium term, the structural deficit, excluding social security, is assumed to be reduced gradually to 3 percent of GDP by 2005/06, broadly in line with the objective set in the revised Fiscal Structural Reform Act.

Germany. The projection for 1998 is based on the 1998 federal budget and Financial Planning Council projections for the other levels of government. It incorporates the unwinding of temporary measures implemented in 1997 and takes account of the tax changes that took effect at the beginning of 1998 (such as the 2 percentage point reduction in the solidarity income tax surcharge and the increase in the basic income tax allowance) and the April increase of 1 percentage point in the value-added tax (VAT) rate. The projections for 1999 and the medium term assume unchanged policies of the territorial authorities—approximated by a constant structural primary balance. The surplus of the social security funds (estimated at 0.3 percent of GDP in 1998) is projected to diminish gradually as the legally mandated liquidity reserve of the pension fund is restored.

France. The projection for 1998 reflects the staff’s assessment of the state and social security budget plans for 1998. The difference from the government’s deficit goal reflects a better revenue performance (mainly on VAT and social securities receipts) partially offset by overruns in health care spending. For 1999, the projection is based on announced official fiscal targets, implying a slight decline in the ratio of expenditure to GDP. Beyond 1999, it is assumed that the ratio of revenue to GDP and the structural primary balance remain broadly unchanged.

Italy. The projection for 1998 is based on the official budget, allowing for the impact of the IMF staff’s macroeconomic forecast and the announced change in the timing of pension payments, which results in a onetime saving in the current year. For 1999–2001, the projections are based on IMF staff estimates of the “current services” budget (tendenziale), corrected for the measures announced in the three-year plan for those years. It is assumed that the plan’s measures are fully implemented and yield the officially estimated amounts. Projections beyond 2001 assume an unchanged structural balance.

United Kingdom. The budgeted spending ceilings for FY 1998/99 are assumed to be observed. Thereafter, non-cyclical spending is assumed to grow in line with the new expenditure plans announced following the recent Comprehensive Expenditure Review. For revenues, the projections incorporate, for FY 1998/99, the announced commitment to raise excises on tobacco and road fuels each year in real terms; thereafter, real tax rates are assumed to remain constant.

Canada. Federal government outlays for departmental spending and business subsidies are assumed to conform to the commitments announced in the February 1998 budget. The employment insurance premium is assumed to be cut by 10 cents a year during 2000–2003. Other outlays and revenues are assumed to evolve in line with the IMF staff’s macroeconomic projections. After 1998/99, it is assumed that the federal government will maintain a small budget surplus, which implies some small cuts in taxes and increases in program spending. The fiscal position of the provinces is assumed to be consistent with their stated medium-term targets.

Australia. Projections are based on the Australian Treasury’s September 1998 projections of ratios of expenditure and revenue to GDP until 2001/02, with these ratios assumed to remain constant thereafter. Unchanged policies are assumed for the state and local government sector from 1998.

Belgium. The fiscal deficit for 1998 reflects the budget and a subsequent revision of the expected outturn owing to lower-than-projected short-term interest rates and the favorable impact of strong economic activity on tax receipts. For 1999 and the medium term, the projections incorporate a slight decline in the structural primary surplus, as a result of planned tax cuts.

Denmark. Fiscal projections for 1998 are based on the implementation of the 1998 Finance Act, which tightened policies to slow down domestic demand growth. For 1999 and beyond, no major changes in tax and expenditure policies are assumed; fiscal projections are adjusted for differences between the economic projections of the IMF staff and the authorities.

Greece. The projection for 1998 is based on the official budget, adjusted to reflect the corrective measures announced following ERM entry and the higher-than-budgeted average interest rate projected for the year. Projections beyond 1998 incorporate the authorities’ intention to increase public capital formation (although by less than the amount indicated in the revised convergence program), an unchanged ratio of current primary spending in structural terms, and a trend decline in the current revenue ratio owing to a projected decline in the share of consumption in GDP. The projections, including those for inflation and other macroeconomic variables, do not incorporate the effects of possible indirect tax reductions or other measures under consideration.

Israel. Projections are based on the 1998 budget and the government’s medium-term fiscal plan, which establishes annual deficit targets for the central government until 2001. In the years thereafter, the projections assume an unchanged fiscal deficit as a percent of GDP.

Korea. The medium-term projections assume that the central and general government budgets will be broadly in balance. In the short term, however, unemployment is expected to pick up sharply, and there are costs of restructuring the banking system. The associated increase in social safety net and bank restructuring expenditures will imply moderate fiscal deficits over the next few years.

Netherlands. The 1998 projections are based on the 1998 budget and the IMF staff’s macroeconomic projections. Projections for 1999–2003 reflect the new government’s expenditure norm and take into account planned tax cuts.

New Zealand. The projections are based on the New Zealand Treasury’s September 1998 projections of ratios of expenditure and revenue to GDP until 2000/01, and IMF staff estimates thereafter.

Portugal. The projection for 1998 is based on the budget, allowing for the impact of recent macroeconomic developments. Projections for 1999 and beyond are based on estimates of revenue and expenditure trends on the basis of unchanged policies.

Singapore. Projections for 1998 are based on budget data adjusted to take account of a fiscal stimulus package worth 1½ percent of GDP announced in end-June 1998; the package comprises tax breaks and public spending on infrastructure projects. The projections also assume lower capital receipts as a result of a freeze of state land sales in 1998 and 1999.

Spain. The projections for 1998 assume that the budget is implemented as passed, allowing for small differences in macroeconomic assumptions. For 1999 and beyond, it is assumed that there is no major change in tax policy. (Notably, the projections do not yet incorporate the impact of the proposed personal income tax reform.)

Sweden. The fiscal projections are based on the authorities’ policies and projections as presented in the 1998 Spring Budget Bill, which includes the objective of achieving a fiscal surplus of 2 percent of GDP on average over the cycle.

Switzerland. The projection for 1998 is based on the 1998 budget plans, adjusted for the estimated fiscal impact of recent macroeconomic developments. For 1999–2001, the projections are in line with official projections and take account of the constitutionally mandated annual maximum deficit ceilings for the federal budget deficits during 1999–2001. Beyond 2001, the general government’s structural primary balance is assumed to remain unchanged.

Monetary policy assumptions are based on the established framework for monetary policy in each country. In most cases this implies a nonaccommodative stance over the business cycle, so that official interest rates will firm when economic indicators suggest that inflation will rise above its acceptable rate or range, and ease when indicators suggest that prospective inflation will not exceed the acceptable rate or range; that prospective output growth is below its potential rate; and that the margin of slack in the economy is significant. It is assumed that Economic and Monetary Union (EMU) in Europe will be implemented from the start of 1999, with short-term interest rates converging close to the lower end of the current range observed across the member countries of the prospective euro area. On this basis, it is assumed that the London interbank offered rate (LIBOR) on six-month U.S. dollar deposits will average 5.7 percent in 1998 and 5.7 percent in 1999 (40 basis points less than projected in the May 1998 World Economic Outlook); on six-month Japanese yen deposits will average 0.7 percent in 1998 and 0.6 percent in 1999 (60 basis points less than projected in the May 1998 World Economic Outlook); on six-month deutsche mark deposits will average 3.7 percent in 1998; and on six-month euro deposits will average 3.7 percent in 1999. Changes in interest rate assumptions compared with the May 1998 World Economic Outlook are summarized in Table 2.1.

In the emerging market economies of Latin America, the weakness of oil prices has contributed to downward revisions in growth projections for Colombia, Mexico, and especially Venezuela, on account of oil production cutbacks and fiscal restraint in response to the weakening of budgetary revenues. Increases in interest rates associated with financial market pressure, and policy measures taken to reduce countries’ vulnerability to such pressure, have affected growth prospects more widely In Mexico, following two previous fiscal adjustments in January and March 1998, the government in July announced an additional round of measures aimed at offsetting the effect of lower oil prices on the budget. Projected growth has also been marked down for Chile, mainly reflecting a tightening of credit policy and weak demand in Asian markets—which account for about one-third of Chile’s exports, about three times the average for the region. In Brazil, where financial market pressures associated with the Asian and Russian crises have been felt more acutely, growth in 1998 and 1999 is projected to slow down considerably from the 3¼ percent rate achieved in 1997. Growth projections for Argentina have also been reduced; unemployment, having fallen below 14 percent, could increase in the period ahead.

Economic developments in Africa have remained uneven. A number of countries in the north have benefited in 1998 from improved agricultural production following last year’s drought and from the strengthening of growth in Europe, and many other countries have been net beneficiaries of declines in oil and non-oil commodity prices. But other countries have been adversely affected by the weakness of commodity prices, and in some cases economic activity has been disrupted by civil and military unrest. The impact of El Niño was less significant than previously feared. South Africa is the only country in the region to have suffered significant financial contagion from the Asian crisis (see below). All these factors have combined to lower the growth projection for Africa in 1998, with significant risks of a weaker outcome.

In Algeria, the strong recovery in agricultural output following the severe drought last year is expected to help growth to rebound to 4 percent this year, despite weak oil prices, and the expansion is projected to be maintained in 1999. A recovery in agricultural production also underlies the rebound of growth in Morocco. In Tunisia, the expansion of manufacturing production is projected to be sustained at relatively high rates in 1998–99, driven partly by the recovery in import demand in European markets. In Nigeria, reflecting the fall in oil prices and political instability, growth is projected to weaken to 2 percent in 1998. In Kenya, growth is projected to strengthen somewhat following the slowdown in 1997, which reflected a deterioration in macroeconomic imbalances. Torrential rains in Uganda in early 1998 may adversely affect this year’s production and exports of coffee.

Activity in the transition countries as a group is now expected to stagnate in both 1998 and 1999, but with substantial differences in performance across countries (Table 2.5), The three Baltic countries and Poland are all expected to record further rapid growth in 1998 and 1999, underpinned by fast-expanding domestic demand, business investment in particular. In Hungary, the adjustment policies implemented since 1995 have set the foundations for sustained growth, projected to accelerate to about 5¼ percent this year and to moderate slightly in 1999. In contrast, growth is projected to slow somewhat in Croatia and the Slovak Republic, where current account deficits of close to 10 percent of GDP, banking sector problems, and other weaknesses need to be addressed. In the Czech Republic, where growth slowed to 1 percent in 1997 as a result of the May currency crisis and associated tightening of policies, activity has remained subdued: real GDP in the second quarter of 1998 was 2½ percent lower than in the corresponding quarter of 1997. In Bulgaria, which saw renewed macroeconomic instability and output contraction last year, activity has rebounded following the fresh impetus given to stabilization and restructuring, while Romania, which was in a similar situation in 1997 but where policies in both areas have lagged after some initial progress, seems likely to see a further decline in output in 1998.

Table 2.5.Countries in Transition: Real GDP and Consumer Prices(Annual percent change)
Real GDPConsumer Prices
Countries in transition-1.02.0-0.2412830
Central and eastern Europe1.62.83.4323818
Excluding Belarus and Ukraine3.73.23.7234117
Czech Republic3.91.01.09811
Macedonia, former Yugoslav Rep. of0.81.55.0222
Slovak Republic6.66.54.0667
Transcaucasus and central Asia1.62.14.1693121
Kyrgyz Republic7.16.56.0302612

Major downside risks for the countries of central and eastern Europe and the Baltics arise from the uncertainties surrounding the effects of the financial crisis in Russia. A weakening of exports to Russia will adversely affect growth in the Baltic countries especially, since their exports to Russia still accounted for up to 20 percent of total exports before the crisis and even a limited deterioration in the trade balance with Russia could have significant effects, as their current account deficits are already very high. In the central and eastern European countries, where the proportion of exports to Russia is less than 7 percent in Hungary and Poland and less than 4 percent in the Czech Republic, the impact of the Russian crisis is expected to be felt mainly through international and domestic financial market conditions, although the relatively strong macroeconomic fundamentals of these countries have helped to limit these effects.

In Russia, and to a lesser extent Ukraine, growth performance and prospects have been damaged by the intense financial market pressures since May, which in Russia’s case culminated in the August crisis, discussed below. Growth in Russia is projected to be severely affected by a sharp fall in domestic demand and disruptions in the financial and payments systems, and output declines of 6 percent are projected for both 1998 and 1999—downward revisions of 7 percent and 8 percent respectively. In Ukraine, the spillover and contagion effects from the Russian crisis and a monetary policy tightening to support the new exchange rate band are expected to result in zero growth this year and only a very moderate recovery in 1999.

A number of smaller countries of the former Soviet Union, including Armenia, Azerbaijan, and Georgia, are projected to enjoy continued strong growth of 5 percent or more in 1998–99. Their trade exposure to Russia is relatively small, and the early stage of development of their financial markets also limits their susceptibility to external disturbances. Other countries of the former Soviet Union that have stronger trade links to Russia or more integrated financial markets are expected to be more affected by the Russian crisis. In Kazakhstan, for instance, and also on account of the decline in oil prices, growth projections have been marked down to 1½ percent in 1998 and zero in 1999. In Belarus, growth projections are subject to substantial downside risk, due to that economy’s heavy reliance on preferential credit and trade with Russia.

The Emerging Markets Crisis: Its Evolution and Spread

Why Has Economic Activity Contracted More Than Expected in the East Asian Crisis Countries?

Since the May 1998 World Economic Outlook, projected growth in 1998 and 1999 for the crisis-afflicted economies of east Asia has been revised down by significant margins. For the countries at the center of the crisis—Indonesia, Korea, Malaysia, the Philippines, and Thailand—downward revisions in projected 1998 growth range from 3 to 10 percentage points, the largest revisions being for Indonesia and Malaysia, where output is now projected to contract by 15 percent and 6½ percent, respectively.

The downward revisions have been prompted partly by new data on activity and trade in the first half of 1998, which have shown the initial impact of the crisis to have been considerably worse than previously thought. In addition, a number of developments have indicated forces making for deeper downturns than projected earlier, including further declines in equity markets and asset values more broadly; the political turmoil in Indonesia; mounting problems in financial and corporate sectors more serious than previously expected; the effects of the deteriorating situation in Japan on confidence and activity throughout the region; and spillovers from the crisis in Russia. Moreover, adverse developments in individual countries of the region have clearly fed on each other, especially through a sharp contraction of regional trade.

The deterioration became increasingly clear in May and June as data for activity in the first quarter became available for several countries. Following signs of recovery in the region’s exchange markets during February–April, a number of the region’s currencies suffered further depreciations at this time, although, except for the Indonesian rupiah, they remained well above the lows reached in January, and upward pressure on the Korean won, in particular, quickly resumed (Figure 2.8). Since June, although exchange markets have continued to experience bouts of volatility, the rupiah also has recovered significantly, while the other currencies have remained fairly stable. Larger and more widespread declines were registered in equity markets (Figure 2.9). GDP data for the second quarter have confirmed the substantial declines in activity (Table 2.6).

Figure 2.8.Selected Emerging Market Countries: Bilateral U.S. Dollar Exchange Rates

(U.S. dollars per currency unit; January 5, 1996 = 100)

Exchange rates in the Asian crisis countries have recovered somewhat since early 1998, but downward pressures have continued in many emerging markets.

Sources: WEFA, Inc.; and Bloomberg Financial Markets, LP.

1Pegged to U.S. dollar.

Figure 2.9.Selected Emerging Market Countries: Equity Prices

(U.S. dollars; logarithmic scale; January 5, 1996 = 100)

Declines in the equity markets of the Asian crisis countries and Russia since late 1997 have been particularly large.

Source: International Finance Corporation, Emerging Markets Database.

1On September 25, 1998 equity prices for Thailand, Indonesia, and Russia reached lows of 9.3, 8.9, and 52.3, respectively, in terms of the indices shown.

Table 2.6.Selected Asian Economies: Macroeconomic Indicators(Percent change from four quarters earlier unless otherwise noted)
Hong Kong SAR
Output growth15.
Trade balance2-6.2-6.2-4.0-4.1-4.2-4.5
Import value growth34.
Export value growth32.
Export volume growth4.
Output growth18.
Trade balance21.
Import value growth311.0-7.8-2.6-10.1-31.3-26.7
Export value growth310.
Export volume growth26.220.433.533.032.819.1
Output growth15.
Trade balance2-7.3-1.8-
Import value growth33.90.8-3.8-14.8-35.4-36.6
Export value growth3-
Export volume growth17.324.135.323.235.024.1
Output growth18.
Trade balance20.8-
Import value growth3-
Export value growth36.10.12.7-5.1-10.6-9.2
Export volume growth
Output growth15.
Trade balance2-2.9-2.8-3.1-2.4-1.1-0.3
Import value growth314.28.320.912.9-5.9-17.4
Export value growth317.526.524.722.223.814.4
Export volume growth
Output growth14.08.510.
Trade balance2-2.1-1.4-2.5-
Import value growth3-
Export value growth3-
Export volume growth-0.38.810.57.87.6-2.0
Taiwan Province of China
Output growth16.
Trade balance21.
Import value growth35.
Export value growth35.84.917.17.1-0.3-7.8
Export volume growth9.35.69.711.43.80.8
Output growth17.07.5-4.2-11.5-16.8-15.8
Trade balance2-3.2-3.1-
Import value growth3-7.6-7.5-11.4-27.5-39.8-38.2
Export value growth3-
Export volume growth-1.74.311.716.314.312.5
Sources: Country authorities; IMF, International Financial Statistics (IFS), and IMF staff estimates.

GDP growth except for Thailand, where growth of manufacturing production is shown.

In billions of U.S. dollars, on national accounts basis (calculated as exports [f.o.b.] less imports [c.i.f.]), except balance of payments basis for the Philippines, Singapore, and Thailand.

In U.S. dollars, on a national accounts basis, except balance of payments basis for the Philippines, Singapore, and Thailand.

Sources: Country authorities; IMF, International Financial Statistics (IFS), and IMF staff estimates.

GDP growth except for Thailand, where growth of manufacturing production is shown.

In billions of U.S. dollars, on national accounts basis (calculated as exports [f.o.b.] less imports [c.i.f.]), except balance of payments basis for the Philippines, Singapore, and Thailand.

In U.S. dollars, on a national accounts basis, except balance of payments basis for the Philippines, Singapore, and Thailand.

The previous projection of early recovery in the crisis-hit countries was based partly on the assumption of rapid improvements in trade balances, including through growth in exports generated by improved competitiveness. The crisis countries have indeed registered substantially improved trade balances in the first half of 1998, with a combined annualized merchandise surplus of over $80 billion, compared with an annualized deficit of around $40 billion in the same period of 1997 (Box 2.2, on page 36). But the improvements have resulted mainly from import compression: total imports in U.S. dollar terms were over 30 percent lower in the first half of 1998 than in the first half of 1997, with declines ranging from about 10 percent in the Philippines to over 25 percent in Malaysia, and over 35 percent in Indonesia, Korea, and Thailand. The total exports of the five countries in U.S. dollar terms were roughly unchanged as a marked increase in export volumes virtually offset declines in prices in U.S. dollar terms, reflecting, in part, declines in primary commodity prices. Changes in export revenues varied considerably among the countries, however, with growth of about 20 percent in the Philippines and close to 5 percent in Korea but declines of 5–10 percent in Indonesia, Malaysia, and Thailand. The overall performance of exports from the five countries masks a sharp contraction of regional trade, especially with Japan, and a strong rise in export revenues from outside Asia. Thus the dollar value of Korea’s exports to the EU and the United States in the first five months of 1998 were, respectively, 6 percent and 11 percent higher than a year earlier.1

In all the crisis countries, notwithstanding the sizable improvements in real trade balances, indicators of overall activity point to deep recessions in 1998. Thus in the second quarter, real GDP was 6½ percent lower than a year earlier in Korea, 16½ percent lower in Indonesia, and 6¾ percent lower in Malaysia. Other indicators of economic activity, such as unemployment, confirm the sharp contractions. Thus in Korea, the unemployment rate climbed from 2 ¼ percent before the crisis to 7 percent in June, its highest level in 12 years (Figure 2.10). Activity has weakened more moderately in the Philippines, with real GDP in the second quarter 1¼ percent lower than a year earlier. Meanwhile, except in Indonesia, where currency depreciation has been by far the steepest, inflation has been generally well-contained, reflecting the flexibility of wages and prices as well as the weakness of demand.

Figure 2.10.Selected East Asian Countries: Unemployment

(Percent of labor force)

Unemployment rates have increased since the beginning of the crisis.

Sources: For Korea, WEFA, Inc.; for Hong Kong SAR, Hong Kong SAR Government Information Center.

1Seasonally adjusted quarterly data.

In Indonesia, Korea, and Thailand, the key priority for monetary policy in the programs supported by IMF arrangements has been exchange rate stabilization. In Korea and Thailand, monetary tightening effectively stanched depreciation early in the year (Box 2.3, on page 40); in both cases, the monetary authorities have allowed short-term interest rates to decline cautiously since January as exchange market pressures have eased (Figure 2.11). In fact, in both cases the currencies have appreciated significantly from their January lows. In Indonesia, exchange rate stabilization appears to have been achieved more recently, although at an extremely depreciated rate despite some strengthening of the rupiah in July and August. It is notable that among the Asian crisis countries, exchange rates held up relatively well in late August and early September, when stock markets in the region, as elsewhere, experienced spillovers from the Russian financial crisis. Data for private sector credit indicate that Indonesia, the Philippines, and Thailand have recently experienced significant declines in real terms, but these are far from matching the deep decline seen in Mexico in 1995 (Figure 2.12). Fiscal policies, meanwhile, have sought to strike a balance between, on the one hand, supporting external adjustment and avoiding an excessive accumulation of public debt that would adversely affect financial stabilization and, on the other, allowing automatic fiscal stabilizers to operate to help stabilize activity, and strengthening social safety nets (Box 2.4, on page 46). Fiscal deficits have widened considerably in all the crisis countries, with adjustment measures offsetting only part of the deterioration resulting from the weakening of activity (Box 2.5, on page 50).

Figure 2.11.Selected Emerging Market Countries: Short-Term Interest Rates


Short-term interest rates have risen in many emerging market economies to support exchange rates.

Sources: Bloomberg Financial Markets, LP; and IMF, International Financial Statistics. Three-month interbank rate or, if unavailable, comparable market-determined short-term rate.

1The Indonesian short-term rate in the first half of September averaged 70.7 percent.

2The Russian three-month interbank rate reached 127.8 on August 12, 1998, after which transactions were suspended.

Figure 2.12.Mexico and Selected Asian Economies: Real Private Sector Credit Growth During Crisis Periods1

Real private sector credit growth has turned negative in the Asian crisis countries, but not by as much as in Mexico in 1995.

Source: IMF, International Financial Statistics.

1Shaded areas indicate the postcrisis period—that is, from July 1997, except after December 1994 for Mexico. Three-month annualized percent change of nominal private sector credit deflated by consumer price index during January 1994 to December 1995 for Mexico and January 1996 onward for the selected Asian economies.

In Malaysia, the authorities in August imposed capital controls to support a repegging of the exchange rate and to contain the effects on external balance of a strategy to revive the domestic economy that involves expansionary fiscal and monetary policies. In the immediate aftermath of these announcements, while equity markets strengthened, in part because of large purchases by domestic institutional investors, yield spreads on foreign Malaysian bonds widened by over 300 basis points; by early September these spreads had widened to about 1,200 basis points.

The unexpectedly deep contractions in activity in the Asian crisis countries have led some observers to question whether the downturn will end in the foreseeable future. As illustrated by experience from earlier financial crises, however, such crises can end and turn around quite quickly. A possible recovery scenario from the present crisis, embodied in the current baseline projections, is discussed next.

Box 2.2.Trade Adjustment in East Asian Crisis Countries

The trade balances of the east Asian crisis countries, which had generally deteriorated in the years leading up to the recent financial crisis, are expected to turn around rapidly this year and next.1 Sharp declines in exchange rates have improved external competitiveness, while contractions in domestic spending have lowered import demand. How large is the overall adjustment in their trade balances likely to be, and how quickly will it be felt? What is the expected composition of the adjustment in terms of movements in export and import values? This box looks at these issues from three perspectives: the staff’s projections, the recent merchandise trade data for these countries, and econometric estimates of trade adjustment.

The May 1998 World Economic Outlook projected an improvement in the combined balance on trade in goods and nonfactor services of the east Asian crisis countries—Indonesia, Korea, Malaysia, the Philippines, and Thailand (henceforth referred to as the Asian-5 countries)—of some $73 billion between 1996 and 1998. It was projected that over half of this adjustment would come from higher export values, with a rise of $43 billion, while imports were forecast to fall by $30 billion. Since that time, forecasts for domestic demand in these countries have been marked down further, reinforcing the projected decline in imports. These changes have more than offset downward revisions to exports, owing in part to weaker activity in Japan and elsewhere in Asia. As a result, the overall adjustment in the Asian-5 trade position has risen to $110 billion in the current projections, with about $100 billion reflecting lower imports and $10 billion attributable to higher exports.

The figure illustrates the recent merchandise trade performance of the Asian-5 region. The first striking feature is the magnitude and speed of the adjustment in the combined trade position since early 1997: from an annualized deficit of roughly $40 billion (4 percent of Asian-5 GDP in 1997) in the first half of 1997, the balance swung to a surplus of over $80 billion in the first half of 1998 (equivalent to 12 percent of projected 1998 aggregate GDP). Even this large shift of $120 billion probably understates the true magnitude of the swing in the external balance because these data reflect only trade in goods—to the extent that the balance on nonfactor services has also improved, the adjustment in the overall trade position would be larger.2 The second important feature of the monthly data is the marked asymmetry in the division of the adjustment between exports and imports. In particular, merchandise exports (expressed in U.S. dollars) were roughly unchanged in the first half of 1998 relative to their level a year ago, while the import bill plunged by about $120 billion, or about one-third of the precrisis level.

Asian-5 Economies: Trade Developments1

(Billions of U.S. dollars, annualized monthly data)

Source: WEFA, Inc.

1The Asian-5 economies are Korea, Indonesia, Malaysia, the Philippines, and Thailand.

To assess the consistency of the trade adjustment indicated by these data with the historical experience of the Asian-5 region, an equation was estimated explaining the movements in trade balances (as a ratio to GDP) as a function of domestic demand relative to foreign activity and the real effective exchange rate. Panel estimation was performed using annual data for the five countries over the 1987–96 period. The results indicate that, in the short run, a 1 percent rise in domestic demand relative to foreign activity weakens the trade balance by about ¾ of 1 percent of GDP, whereas an 8 percent appreciation of the real effective exchange rate reduces the balance by about 1 percent of GDP.

The current projections suggest that domestic demand in the Asian-5 countries will fall by a cumulative 20 percent between 1996 and 1998, compared with an increase in foreign activity of 5 percent. Based on the estimated parameters, the difference in domestic versus foreign growth would improve the trade balance by about 15 percent of GDP in the short run. In addition to this activity effect, the real effective exchange rate of the Asian-5 countries, on a weighted average basis, has depreciated by almost 40 percent since 1996. On the basis of historically observed elasticities, this would generate an additional improvement in the trade position of 5 percent of GDP. Taken together, the activity and exchange rate effects imply an overall swing in the trade balance of 20 percent of GDP, or $140–150 billion—somewhat larger than the adjustment indicated by the recent merchandise trade data. This difference would be consistent with some improvement in the services trade balance, which is not reflected in the merchandise data. On this basis, it appears that the dramatic improvement in the combined Asian-5 external position observed so far in the incoming trade data is broadly in line with historical relationships.

Why has the adjustment reflected a sharp decline in import bills, while export revenues have shown virtually no growth through early 1998? The stagnation of export revenues appears to reflect two factors: lower export prices (in U.S. dollars), which have masked the effect of higher volumes; and weaker domestic demand in other Asian markets, especially Japan. Assessing the precise size of the export price decline is difficult given the lack of timely data on trade prices and volumes for these countries. The exception is Korea, where monthly data for early 1998 indicate that export prices in dollar terms fell by some 15–20 percent as the won depreciated by about 40 percent against the dollar. Assuming that a decline of this size is characteristic of the experience of the Asian-5 countries as a group, the stability of export values indicates that volumes have increased substantially—by well in excess of the 5 percent rise in partner-country activity during 1997–98. These calculations suggest that export volumes have indeed responded to exchange rate depreciation, but that the effect on dollar values has been offset by lower prices.3

As regards the effect of partner-country activity on exports, a comparison of the behavior of growth in U.S. dollar revenue from exports to Japan and the United States is instructive. Data for the Asian-5 countries (excluding Indonesia) indicate a drop in exports to Japan in the first quarter of 1998 of about 14 percent relative to the same period in 1997, versus a rise in exports to the United States of 16 percent. Over this period, Japanese domestic demand fell by about 5 percent, while U.S. domestic demand rose by almost 5 percent. In addition, the yen weakened against the dollar by about 10 percent, reducing the competitiveness gains of Asian-5 exporters in Japanese markets. Although complete data are not yet available on the geographic breakdown of Asian-5 trade or demand growth in partner countries, this comparison suggests that Asian-5 export revenue in U.S. dollar terms has continued to grow strongly in markets where domestic demand has been robust and competitiveness gains large, but that weaker activity and exchange rates elsewhere in Asia have reduced revenue from exports to regional trading partners.

On the import side, the fall in dollar import bills of about one-third observed through early 1998 can be attributed partly to the decline in Asian-5 domestic demand of 20 percent projected for 1998, and partly to sharp increases in the relative price of imports. Again, comprehensive price and volume data are not yet available that would allow a full analysis of import adjustment. But the observed magnitude of the decline seems consistent with plausible responses to standard determinants. Assume, for instance, that import prices measured in dollars have remained unchanged, while Asian-5 real exchange rates have depreciated by some 40 percent. Then a short-run elasticity of imports with respect to domestic demand of unity and to relative import prices of -⅓ would generate a 34 percent drop in import volumes over this period, similar to the decline observed.4

On balance, then, the picture provided by both observed trade data for the first half of 1998 and the econometric estimates suggests that the ultimate adjustment in the Asian-5 trade position could be even larger than that embodied in the latest forecast. It also seems possible, at least in the short run, that more of the adjustment could come through lower import values, and less through higher exports. Of course, this situation could well change over time, since the response of export volumes to declines in relative prices may well increase over the longer run and constraints on the availability of trade credit to exporters ease. The prospects for Asian-5 exports also depend on the course of the Japanese economy, as well as future developments in the exchange rates of these countries.

1The countries included in this group are Indonesia, Korea, Malaysia, the Philippines, and Thailand.2Data on trade in nonfactor services are not available on a timely basis for most of the Asian-5 countries. An exception is Korea, where data through May indicate an improvement of about $7 billion (annual rate) in the first five months of 1998 relative to the same period in 1997.3Another perspective on valuation effects is given by expressing export revenue in terms of the local currencies of the Asian-5 countries—on this basis, export revenue rose by about 65 percent in the first months of 1998 over the equivalent period in 1997, implying substantial increases relative to GDP for these countries.4These elasticities are, of course, subject to considerable uncertainty, as is the assumption that import prices in dollars have remained constant as Asian-5 exchange rates have depreciated. To the extent that import prices have in fact declined, the implied drop in import volumes would be smaller, reinforcing the conclusion that the drop in import values does not seem atypically large.

What Are the Forces That Will Determine the Recovery Path in East Asia?

A key uncertainty in the baseline scenario concerns the timing and vigor of the eventual turnaround and recovery in the Asian crisis countries. On current projections, output in each of the crisis countries is expected to bottom out during the first half of 1999, with some pickup in activity by the end of the year. Little growth is projected for 1999 as a whole, compared with 1998, but the pickup in activity in the course of next year is projected to be followed by a period of sustained recovery beginning in 2000 (Table 2.7, on page 45). This scenario, however, is subject to both upside and downside risks that differ among the affected countries. In Korea and Thailand, financial market confidence seems to have begun tentatively to return, and there are reasonable prospects of some pickup in activity in 1999. In Indonesia, however, where the economic situation has remained particularly difficult, recovery seems likely to come later than in the rest of the region.

Table 2.7.Baseline Recovery Path of the Asian Crisis Countries1(Annual averages; in percent of GDP unless otherwise noted)
Contributions to growth
Foreign balance-
Domestic demand7.51.1-
Public consumption0.60.20.9-1.1-0.40.2
Private consumption4.32.4-4.1-
Gross investment2.7-1.5-
Central government fiscal balance1.0-4.9-3.9-2.2-0.3
Current account balance-4.9-
Net private capital flows6.8-1.6-18.5-
Consumer prices (median)

Comprises Indonesia, Korea, Malaysia, and Thailand. IMF staff estimates from 1998 through 2003.

Annual percent change.

Comprises Indonesia, Korea, Malaysia, and Thailand. IMF staff estimates from 1998 through 2003.

Annual percent change.

The current projections are based on the assumption that stabilization and reform policies being implemented in the crisis countries will put in motion a set of mutually reinforcing economic forces that will stem the contraction in activity and pave the way for initial recovery. Chief among these forces are the impact on financial market confidence of the turnaround in external balances and progress in the reform process. As discussed in the preceding subsection, in all the crisis countries trade and current account balances have already shifted into large surplus. These surpluses will help countries to reconstitute their depleted foreign exchange reserves and thereby restore the confidence of investors in the ability of the authorities to meet normal demands for foreign exchange. The restoration of confidence, in turn, along with the liberalization of regulations governing foreign ownership of assets in many of the crisis countries, will promote the recovery of capital inflows, provide further support for the stabilization of exchange rates, and permit gradual easing of interest rates. This process has already begun—in fact, it seems quite well advanced in Korea and Thailand—and with continued implementation of appropriate policies it may be expected to continue in the period ahead. The return of financial market confidence and the cautious easing of interest rates will reinforce the stimulative effects of increased net exports by providing support for a pickup of private consumption and business investment.

Although these positive forces will eventually bring an end to the current downturn in activity and provide some initial stimulus for recovery, other factors may act as a drag on recovery. Two major issues in this context are the related questions of how quickly and effectively financial sectors are restructured, and the effects and eventual resolution of the problems associated with high corporate debt burdens.2 These problems, which are discussed in more detail in Chapter III, have led to large-scale insolvency in the financial and corporate sectors. Moreover, as a consequence of the debt burdens borne by the private sector, creditors have been reluctant to extend further credit even to viable enterprises, which are suffering from a lack of working capital, trade credit, and export guarantees. This has been one cause of the sharp contraction in activity in the first half of this year in these countries.

In the past few months, considerable efforts have been made to find solutions to financial and corporate sector balance sheet problems. In Thailand, significant progress has been made in implementing financial sector reforms. A large number of failed financial institutions have been closed, a strategy is being implemented for the restructuring and return to private ownership of institutions in which authorities have intervened, and bank regulation and supervision are being strengthened. There has also been significant progress in reforming and strengthening the banking system in the Philippines. In Korea, there has been significant progress in financial and corporate sector restructuring: the merchant banking sector has been rationalized and consolidated; 5 out of 12 undercapitalized commercial banks have been closed, and recapitalization plans for the 7 others are being evaluated; a procedure for voluntary debt workouts between corporations and banks has been established; and a number of chaebols have been pursuing restructuring. In Indonesia, a separate agency has been established to deal with failed financial institutions. In addition, in Korea a debt-conversion agreement among international creditors, domestic banks, and the government, covering part of outstanding bank debt, has been reached. There has been less progress in addressing corporate sector debt problems in the crisis countries, although, in Indonesia, international creditors and the authorities have agreed on a framework to restructure the country’s private external debt.3 Given the complexity and size of the indebtedness of the corporate sectors in these countries, it is unlikely that comprehensive solutions can be quickly decided and implemented. In the near term, the more pressing need is to put in place mechanisms that will, on the one hand, assure creditors that their interests associated with existing claims will be safeguarded while a solution to the problems is being worked out and, on the other hand, will enable viable enterprises to carry out productive activities.

Box 2.3.The Role of Monetary Policy in Responding to Currency Crises

In the course of the Asian crisis, a number of emerging market economies have experienced severe currency pressures. In several instances—most notably, of course, in Asia itself—this has led to the abandonment of currency pegs and to large depreciations well beyond what could reasonably have been expected prior to the turbulence (see figure). But in other cases, authorities have successfully maintained exchange rate pegs (for example, in Argentina and Hong Kong SAR) or other arrangements with limited flexibility (for example, Brazil), or contained depreciation under more flexible arrangements (for example, Mexico), or reestablished exchange rate stability fairly quickly after an initial depreciation (for example, the Czech Republic).

Pros and Cons of Monetary Tightening

The pressure that arose in each case depended on the perceived sustainability of the exchange rate, the closeness of the economy’s links with the crisis economies, and the health of the financial system, as well as on the credibility accumulated by the authorities as a result of the past conduct of policy. But a major element determining the outcome in each case was the response of the authorities to the pressure that arose.1 Currencies were stabilized when monetary policy was tightened with ample force and persistence and when broader measures were implemented where necessary to address underlying macroeconomic imbalances and structural weaknesses. Currency stabilization and a return of market confidence in turn provided room for a gradual lowering of interest rates, helping to mitigate and reverse the adverse effects on economic activity of the initial tightening. But when monetary policy was not tightened forcefully, in some cases because vulnerability of the financial system formed an obstacle, instability persisted.

Monetary policy tightening is generally necessary to defend a currency that is under severe pressure. Higher interest rates raise the nominal return to investors from assets denominated in the currency and make speculation more expensive by increasing the cost of shorting the currency. Tighter monetary policy will tend to support the exchange rate also by reducing expectations of future inflation and therefore of future currency depreciation, and by lowering domestic demand and improving the current account. A forceful response to currency pressure also reduces default risk for domestic residents who have borrowed in foreign currency.

This is not to say that monetary tightening as a response to currency pressure does not have drawbacks or limitations, apart from its negative impact on domestic activity. Tight monetary policies increase the debt burden of domestic currency borrowers, which may put downward pressure on the exchange rate when investors perceive greater credit risks and thus put a higher risk premium on the currency itself.2 In this regard, the adverse effects of tighter monetary policy will be greater in economies with relatively large domestic debt burdens or when, because of credibility problems, high interest rates have to be maintained for a prolonged period.3 In addition, interest rate increases are unlikely to be sufficient to provide lasting support for the exchange rate when there are perceived to be substantial inconsistencies between the exchange rate and economic fundamentals, or when economic circumstances are such that the higher interest rates are not considered sustainable for any length of time. It should also be noted that high interest rate policies have distributional implications, since they will lead to redistribution from domestic debtors to domestic creditors for debts denominated in domestic currency, while lessening income losses for (mainly corporate) agents with net liabilities in foreign currency. But despite these caveats, in a currency crisis, when the first priority must be to secure macroeconomic stability and prevent a pervasive loss of confidence, a monetary stance that demonstrates the authorities’ determination to maintain monetary discipline cannot be avoided, and it is usually the least costly in terms of forgone output.

Policy Under Pressure

Among the emerging market economies that in the fall of 1997 were successful in supporting exchange rate pegs through interest rate increases along with other macroeconomic and financial adjustments, Argentina, under its currency board arrangement, let market-driven interest rate increases absorb the pressure that spilled over from Asia. But such spillovers were limited, and overnight interest rates peaked at around 13 percent in early November before falling back by early 1998. Subsequently, in the aftermath of recent events in Russia and spillovers to Latin America, interest rates increased to over 10 percent by early September. Similarly, in Hong Kong SAR, short-term rates have on a number of occasions in the past year risen into the middle teens as they have absorbed pressure on the Hong Kong dollar, and they briefly touched higher levels in late 1997. As the currency market came under renewed speculative attack in August 1998, however, the Hong Kong authorities responded by intervening in the currency and stock markets, amid concerns about the state of the domestic economy, and subsequently announced measures to discourage speculative trading, boost liquidity, and lower interest rates. Among countries having arrangements with limited flexibility, Brazil tightened monetary conditions forcefully—with official rates rising from around 20 percent to more than 43 percent in late October 1997—as part of the measures taken to maintain the real within the official currency band. When exchange rate pressures again arose in August, the authorities initially responded with intervention in the currency market and only raised interest rates from around 20 to 50 percent after some delay in early September when international reserves were being expended at an unsustainable pace. In part, the delayed response on monetary policy may be attributable to concerns about a sluggish domestic economy, political uncertainties related to an upcoming general election, and a large stock of mostly domestically held short-term public debt, which had been increasingly indexed to overnight interest rates or the U.S. dollar during 1998.

Selected Economies: Currency Market and Monetary Indicators

Source: Bloomberg Financial Markets, LP.

1In U.S. dollars per national currency; for Czech Republic, a basket of currencies (sum of 0.65 times deutsche mark exchange rate and 0.35 times U.S. dollar exchange rate) per national currency. The Czech koruna was pegged to this basket until May 1997.

2Monthly average of daily overnight interbank rates.

The experience of these countries also illustrates that, for an interest rate defense to be successful, policies addressing macroeconomic imbalances or structural weaknesses need to be in place. Argentina, for example, had significantly strengthened its financial system following the 1995 banking crisis, established a track record in defending its currency during the “tequila crisis” of 1994–95, and has tightened its fiscal stance further since late 1997 and again in August 1998, while Brazil announced several packages of fiscal measures to accompany the interest rate hikes, including to in part offset some of the fiscal costs of those hikes.

The east Asian crisis countries, in contrast, did not raise interest rates promptly and significantly, or engineer appropriate and orderly exchange rate adjustments, when exchange rate pressures emerged in the course of 1997. As a result, they were forced to abandon their pegged or relatively inflexible exchange rate arrangements as international reserves were depleted. In Thailand, the authorities responded to periodic pressures on the baht in early 1997 by intervening in the spot and forward markets, but overnight interest rates generally remained below 10 percent until May. As the crisis intensified in the second half of May and in June, interest rates were allowed to rise; overnight rates peaked at around 30 percent at the beginning of July, when the peg to the U.S. dollar was abandoned. In both Malaysia and the Philippines in that same period, the authorities only briefly defended the values of their currencies relative to the U.S. dollar through interest rate hikes and intervention and subsequently allowed their currencies to depreciate substantially. A more forceful but delayed interest rate response in Indonesia in the second half of July and the first half of August 1997 drove up overnight interest rates to close to 100 percent, but this proved insufficient to stem the pressure, and the rupiah was allowed to float independently on August 14. When downward pressure on the Korean won intensified starting in the last week of October 1997, the authorities first responded by intervening heavily. They raised interest rates and floated the won only in December, when a currency crisis was already well under way.

Given the structural problems, such as corporate overindebtedness, present in the east Asian countries, a more rapid and forceful monetary policy response alone would not, of course, have been sufficient to stem the pressure on their currencies. The unresolved structural problems thus not only contributed to the pressure but made an interest rate defense less credible and more costly. A combination of structural measures and tight monetary policies, together with appropriate fiscal policies, was needed. In cases where there were grounds for considering currencies to be overvalued and exchange rate adjustment to be called for, monetary policy tightening would have needed to have been accompanied by timely exchange rate action.4

The importance of adequately addressing fundamental imbalances is underscored by Russia’s experience.5 It succeeded in defending the ruble and maintaining its currency band arrangement during successive periods of market turbulence in late 1997 and the first half of 1998, but did not manage to restore investor confidence while fiscal problems remained unresolved. As a result, interest rates had to be kept substantially above pre-October 1997 levels, exacerbating the fiscal problems given the large amount of short-term government debt being rolled over, and international reserves were gradually depleted. By early July 1998, international reserves had fallen to $13.5 billion, down from $24.5 billion in mid-1997, and the authorities were forced to seek additional financial assistance from the IMF and other official sources for continued defense of the currency band. When pressure on the exchange rate resumed in early August, as confidence in the implementation of measures to address the fiscal problems weakened, monetary authorities were not any longer willing to tighten monetary policy. The Central Bank of Russia kept the refinance rate, which in late July had been lowered from 80 percent to 60 percent, unchanged and extended large amounts of liquidity support to banks in an attempt to sterilize the effects of its foreign exchange market interventions. Within weeks, the exchange rate band vis-à-vis the U.S. dollar had to be abandoned.

With independently floating exchange rates, interest rate policies still have an important role in preventing self-reinforcing cycles of depreciation and losses of confidence, and periods of relatively high interest rates may be needed to stabilize a floating exchange rate when confidence is lacking. For instance, in the Czech Republic, when in late May 1997 the monetary authorities abandoned the currency peg against the deutsche mark-U.S. dollar basket and switched to a float, the exchange rate against the dollar fell by some 10 percent relative to the previous parity. Interest rates were initially kept high following the float and as market pressures abated in the following weeks they were then gradually lowered, with the two-week repo rate being brought down from a peak of 75 percent at the end of May to under 20 percent at the end of June. In addition, the authorities introduced measures to reduce the fiscal deficit, restrain wage growth, and strengthen the structural reform program. This combination of policies prevented the currency from depreciating further: its value in terms of the deutsche mark was broadly unchanged in the remainder of 1997, and by July 1998 had returned to levels prevailing before the May 1997 crisis. Following the financial turmoil in Russia in August, the currency weakened somewhat, alleviating concerns about the appreciation in previous months.

In the east Asian crisis countries, by contrast, in the period immediately following the switch to independent floating, the monetary authorities allowed interest rates to fall back quickly, and confidence and stability were not regained. It is also noteworthy that in these countries domestic credit growth remained strong throughout 1997. In contrast, domestic credit growth was sharply curtailed in the other countries facing exchange market pressures during the same period, as well as in other countries in previous crises (see Chapter II). Because of a lack of sustained commitment to higher interest rates and delays in the adoption of structural measures, not only could currencies not be stabilized for several months after they were floated, but also the subsequent hike in interest rates lacked credibility because of the earlier hesitation and the deterioration in macroeconomic and financial conditions in the meantime. Nevertheless, in Thailand, tighter monetary policies in the last quarter of 1997, along with other policy adjustments, finally began to stabilize the currency by early 1998. In Korea, by January 1998, the overnight interest rate had been raised to around 30 percent, and these high rates together with other factors—including debt restructuring, mobilization of additional financial support, and structural measures—managed to halt the decline of the currency. The exchange rate has also stabilized in Indonesia since June 1998.

Breathing Room

Once currencies have stabilized and confidence returns, authorities have room to begin lowering interest rates again, thereby limiting the negative output effects of the initial tightening. This changing stance of monetary policy following stabilization is illustrated by the recent experience of the east Asian crisis countries and some other emerging markets. Argentina and Brazil, for example, which in the wake of the Asian crisis were successful in defending their currency arrangements, and the Czech Republic, which avoided an excessive depreciation when it abandoned its peg, were in a position to bring down interest rates again relatively quickly. In Argentina, the increased level of overnight interest rates during late 1997 was maintained for less than a month, while in Brazil, following sizable capital inflows in the first quarter of 1998, the official rate was by mid-April brought down close to levels observed in early October 1997. In the Czech Republic, the benchmark two-week repo rate had been reduced to around 200 basis points above its precrisis lows within two months after the introduction of the float. Similarly, in Korea and Thailand, where sustained monetary policy tightening managed to restore confidence in early 1998, interest rates fell significantly in the second quarter, and by the end of July overnight interest rates in Korea and Thailand had returned to levels observed at the beginning of 1997, despite ongoing turbulence in other emerging markets. At the same time, with greater confidence, part of the sharp depreciation of the initial crisis months was corrected, as the Korean won and the Thai baht recovered by about 20 percent during the first eight months of 1998.

1See Box 5, “Policy Responses to Exchange Market Crises,” in the December 1997 World Economic Outlook: Interim Assessment, pp. 42–43, for a discussion of policy responses to exchange market crises prior to 1997 and initial policy responses to the east Asian financial crises.2See, for example, Steven Radelet and Jeffrey Sachs, “The East Asian Financial Crisis: Diagnosis, Remedies, Prospects” (Cambridge, Massachusetts: Harvard Institute for International Development Working Paper, April 1998).3Among the east Asian countries, Korea and Malaysia were on this account more susceptible, prima facie, to tighter monetary policy because most of the debt in these countries was owed to domestic creditors prior to the crises, while Indonesia and Thailand seemed more susceptible to the risk of excessive currency depreciation. Korea, however, was also vulnerable to a currency depreciation because a relatively large portion of its foreign debt was short-term.4In contrast with the Thai experience, the Czech Republic, when faced with increasing pressure in May 1997, raised interest rates and abandoned its fixed peg well before international reserves were depleted.5During the first bout of turbulence, in the fall of 1997, the initial interest rate response also proved insufficient. Reluctant to raise interest rates further, the Central Bank of Russia in the course of November allowed international reserves to fall from around $23 billion to less than $17 billion, while offsetting the liquidity impact by letting net domestic assets (NDA) expand. Only at the beginning of December was more decisive interest rate action taken, and the expansion in NDA halted.

As discussed in the May 1998 World Economic Outlook, historical experience suggests that the duration and speed of recovery from financial crises vary considerably from case to case, often depending on how effectively financial sector problems and corporate sector difficulties—which are often an integral part of the financial sector weaknesses contributing to crises—are dealt with. In the current crisis, too, how deftly the financial and corporate sector problems are managed will be important—not only for the strength of the initial pickup in activity, but also or the prospects for sustained recovery. To a large extent this will determine whether the east Asian countries are in store for a protracted period of slow growth, as in Japan since the early 1990s, a U-shaped recovery, as in the case of Chile in the early 1980s, or a sharp rebound, as in Mexico in the mid-1990s.

In addition to the domestic conditions for recovery discussed above, economic prospects for the crisis countries are interdependent, given the importance of intraregional economic relations: on this account, recoveries will tend to be mutually reinforcing for the same reasons that the downturns have been mutually damaging. Economic prospects for the crisis countries will also depend critically on the external environment. Particularly important, in this regard, are the health of the Japanese economy, exchange rate developments among the major currencies and within the region, success in resolving the financial sector problems that afflict much of the region, and conditions in international capital markets, as discussed in Chapter I.

How Have Other Emerging Market Economies in Asia Been Affected?

Among the countries in Asia that have felt the crosscurrents from the crisis, growth projections have been revised the most for Hong Kong SAR. The economy is now projected to contract by 5 percent this year, compared with growth of 3 percent projected earlier, and to show no growth in 1999. Real GDP in the second quarter was 5 percent lower than a year earlier, with both investment and consumer spending depressed. Unemployment rose to 5 percent on average in June-August, its highest level in 15 years. Equity and real estate prices have declined sharply. The recession in Hong Kong SAR reflects the openness of the economy and the tightening of monetary conditions that has been needed to maintain the peg to the U.S. dollar under the currency board arrangement, in the face of the deterioration in competitiveness stemming from the general appreciation of the U.S. dollar and the declines of other currencies in the region, and several intensive speculative attacks. In response to the weakening of activity, the authorities in late June announced a number of measures to stem the decline in property prices and stimulate the economy, including a suspension of government land sales. The estimated fiscal cost of these measures is 2¼ percent of GDP. However, pressures resumed in July and August, with interbank interest rate rising and stock prices falling sharply. The Hong Kong Monetary Authority responded by intervening in the stock market. Whether this intervention bolstered confidence in equity values is unclear; it may have been counterproductive by suggesting a reduced commitment to the rules of the currency board. The intervention was halted by end-August, and new measures were adopted to discourage speculation in the futures market. In early September, to strengthen the currency board arrangement, the discount window facility was modified to buffer the effects of temporary market pressures on interest rates.

Projected growth in China in 1998 has been revised down to 5½ percent, partly owing to recent widespread flooding, with the weakening expansion having already contributed to rising unemployment. Industrial production growth has remained well below last year’s pace, imports have weakened further, and deflationary pressures appear to have intensified. Public investment in the first seven months of this year was 15½ percent higher than a year earlier, as the government stepped up efforts to boost demand, but it is unclear that planned fiscal measures alone will be successful in reversing the slowdown. Despite slowing growth elsewhere in Asia, China’s external position remains strong: its current account is projected to remain in surplus, and its foreign exchange reserves are large. Together with its relatively closed capital account, these factors have helped China to maintain its exchange rate vis-à-vis the U.S. dollar. However, there is some uncertainty surrounding the overall external position in view of recent increases in recorded and unrecorded outflows. There is also a considerable overhang of unused commercial property in major cities, exerting downward pressure on property prices, with adverse implications for the financial sector. Equity prices have remained under downward pressure, and yield spreads on Chinese bonds have increased. Furthermore, although there has been some progress in tackling the problems of inefficient state-owned enterprises, they continue to be a significant burden on the budget and the banking system, which is saddled with a significant proportion of nonperforming loans as a result of directed lending. The large accumulation of inventories by the industrial sector is also likely, sooner or later, to be a drag on growth.

Box 2.4.The Asian Crisis: Social Costs and Mitigating Policies1

The impact of the Asian crisis on poverty and human welfare is of considerable concern. In the three countries that have suffered the most acute crises and that are implementing programs supported by the IMF—Indonesia, Korea, and Thailand—poverty is likely to increase markedly. In Indonesia especially, the extent and depth of the social crisis is likely to be substantial. The programs being implemented incorporate many social protection measures aimed at mitigating the fallout of the crisis on the vulnerable as well as the short-term adverse effects of adjustment and reform.

Various factors are adversely affecting real household incomes in the three countries—including currency depreciation, higher interest rates, financial sector collapse, corporate bankruptcies, job losses, and supply bottlenecks, including national calamities. Although most households will be hurt by these developments, some could gain—for example, if they are engaged in the export sector or are net holders of foreign assets.2

The negative effects of the economic crisis and adjustment on living standards of low-income households will occur mostly through price increases and the associated cut in real consumption, and the loss of job opportunities.3 Of the three countries concerned, Indonesia is confronted with the largest reduction in real GDP, and the highest inflation and unemployment in 1998 (see first table).

The short-term effect of the economic crisis on the poor and vulnerable can be estimated by using data on household expenditures by income group, together with projected changes in real GDP and unemployment. This calculation only accounts for the direct, first-round impact of the crisis and neglects effects operating through channels other than real household income and job losses. For instance, the loss of purchasing power could be offset or reinforced by changes in saving behavior, informal safety nets (relatively important in Indonesia and Thailand), and safety nets provided by firms (important in Korea). In previous economic downturns in Korea (1980), Indonesia (1983), and Thailand (1984), changes in household saving behavior and informal safety nets absorbed a significant part of the downward pressure on consumption. However, there is evidence that households in Korea have raised their saving rates in the current crisis. The results of the analysis are therefore preliminary and should be interpreted with caution.

Real GDP, Unemployment, and Inflation in Indonesia, Korea, and Thailand(Annual average)
Real GDPIndonesia8.04.6-15.0
(percent change)Korea7.15.5-7.0
(percent of labor force)Korea2.02.77.0
Consumer price inflationIndonesia7.96.660.0
(percent change)Korea4.94.48.5
Food price inflation1Indonesia9.58.876.2
Sources: National employment statistics, International Labor Organization, and IMF staff estimates.

For 1998, price increases are averages for the first eight months.

Sources: National employment statistics, International Labor Organization, and IMF staff estimates.

For 1998, price increases are averages for the first eight months.

The impact of the decline in real GDP on poverty will depend on its effect on real household income. This can be estimated using the elasticity of the proportion of the population below the poverty line with respect to real mean consumption expenditures. The elasticity has been calculated at about -2.8 for Indonesia and -1.9 for Thailand during the period 1985–95.4 For Korea, the elasticity is assumed to be -1.5. Using these elasticities and the real GDP projections for 1998, and assuming that real household consumption falls at the same rate as real GDP, a rough estimate of the increase in poverty attributable to the fall in real incomes can be calculated. In Indonesia, the number of poor would increase by 9.2 million persons (5 percent of the population), in Korea 0.7 million persons (2 percent of the population), and in Thailand 1.3 million persons (2 percent of the population) (see second table). These estimates assume that income distributions in these countries remain unchanged, and that the elasticities—representing past conditions—are not affected.

Poverty in Indonesia, Korea, and Thailand(Percent of population)
Poor before the crisis11.315.715.1
Additional poor due to real income decline4.81.62.3
Additional poor due to unemployment0 to 6.40 to 10.50 to 9.3
Total additional poor4.8 to 11.21.6 to 12.12.3 to 11.6
Source: IMF staff estimates.
Source: IMF staff estimates.

The impact on real consumption could be reinforced by large price increases for particular commodities. These specific price increases affect households differently, depending on the shares of the goods concerned in households’ consumption baskets. In Indonesia and Thailand, food makes up a relatively large share of the poor’s consumption basket; for the average poor household, food accounts for 71 percent of household expenditure in Indonesia (with rice accounting for 20 percent) and about 55 percent in Thailand. Thus, the impact on poverty of the significant price increases for food would be relatively large in these two countries for the existing poor and those clustered just above the poverty line. In Indonesia, the latter group accounted for about 30 percent of the population before the crisis.

Job losses in the formal sector would push skilled workers into jobs in agriculture and the informal sector in Indonesia and Thailand, while they would mostly affect the middle class in Korea. Households in the top and bottom quintiles would not be affected by job losses as much as other income groups. Assuming that one person per household is affected by unemployment, an increase in unemployment in Indonesia of 9 million workers (equivalent to an increase in the unemployment rate by 10 percentage points) since before the crisis would affect 21 percent of all households. The projected unemployment increase in Korea and Thailand of 1.2 million workers (a 5.5 percentage point increase in the unemployment rate) and 1.4 million workers (a 4 percentage point increase in the unemployment rate), respectively, would affect 10 percent of Korean households and 9 percent of Thai households. The household size varies from an average of 3.7 persons in Korea to 4.5 persons in Indonesia. Both the projections of the number of unemployed and of the affected households should be treated with caution, given the lack of a robust relationship between output and unemployment in these countries.

The impact of rising unemployment will affect some households more than others and thus could alter the income distribution. Its effect on poverty will depend on the extent of the associated decline in household incomes; not all households affected by job losses will fall below the poverty line. For instance, in Indonesia, the average consumption of households affected by unemployment would decline by about 30 percent if these households were initially clustered in the middle of the household consumption distribution, and they would slide down one decile in the distribution after unemployment. The reduction in consumption of affected households in Korea and Thailand would be expected to be half as much under similar assumptions.5 The situation would be made worse in Indonesia if new labor market entrants, numbering about 2.5 million annually, do not find jobs.

The effects of increasing unemployment on poverty can be calculated for all three countries by using their respective consumption distribution data and the above assumptions about the number of middle-class households affected. In Indonesia the number of poor would increase by up to 12.3 million persons (6 percent of the population), by up to 4.7 million persons in Korea (10 percent of the population), and by up to 5.4 million persons in Thailand (9 percent of the population).

If the effects of the real income decline and unemployment on poverty are combined, the number of poor in Indonesia would increase by between 9.2 million persons and 21.5 million persons (5–11 percent of the population), in Korea by between 0.7 million persons and 5.4 million persons (2–12 percent of the population), and in Thailand by between 1.5 million persons and 6.7 million persons (3–12 percent of the population).6

The poverty estimates presented here are based on the assumption that the poverty line is not redefined in light of the large changes in real consumption levels. Such adjustments of the poverty line could take place, however, especially in Korea where the poverty line reflects social attitudes to minimum living standards. In Indonesia, in contrast, the scope for adjusting the poverty line is limited, because its level is already quite low.7

In all three countries, arrangements for social protection were already in place before the crisis. However, the coverage of these arrangements is limited. For instance, only Korea has a formal system of unemployment benefits. The experience of Mexico in dealing with the crisis in 1994 and 1995 illustrates the importance of labor market flexibility in mitigating adverse social effects. As in Mexico, it is likely that urban informal activities and migration from urban to rural areas will absorb some of the unemployed in Indonesia and Thailand.

The IMF-supported programs in Indonesia, Korea, and Thailand include many measures to strengthen social protection, and in particular to shelter the poor from the adverse effects of the economic crisis. The challenge has been to establish cost-effective and fiscally sustainable safety nets that do not create large labor market disincentives or discourage job creation. In Indonesia, subsidies on food, fuel, electricity, medicine, and other essential items have been increased to around 6 percent of GDP from about ½ of 1 percent of GDP in 1997, although the benefit of fuel subsidies accrues mostly to better-off population groups. In addition, employment-generating public works programs directed toward poor households, social programs targeted to children in poor regions, and credit programs targeted to rural areas and small and medium-sized enterprises, among other groups, have been expanded. In Korea, unemployment insurance coverage will be extended, in several stages, to workers in small firms, as well as part-time and temporary workers (daily workers will not be covered).8 Also in Korea, the allocations for social assistance and special loan programs for the unemployed have been increased, a temporary program of noncontributory income support for the unemployed has been introduced, and public works programs are being expanded significantly. The cost of these programs will exceed 2 percent of GDP in 1998. In Thailand, temporary labor-intensive civil works programs in construction and infrastructure rehabilitation and job training programs have been introduced, social spending has been strengthened, and government subsidies for urban bus and rail fares have been maintained. The budgetary cost of these programs will amount to around 6 percent of GDP in 1998.

If the crisis deepens, it will clearly be necessary to expand social protection measures further.

1Based on Sanjeev Gupta, C. McDonald, C. Schiller, M. Verhoeven, Ž. Bogetić, and G. Schwartz, “Mitigating the Social Costs of the Economic Crisis and the Reform Programs in Asia,” Paper on Policy Analysis and Assessment 98/7 (Washington: IMF, 1998).2Twenty-five percent of households in Thailand are net agricultural producers and are likely to see an increase in their net earnings as a result of an increase in food prices.3Ke-young Chu and Sanjeev Gupta, eds., Social Safety Nets: Issues and Recent Experiences (Washington: IMF, 1998), discuss the methodology used here for assessing the impact of macroeconomic developments on poverty.4This elasticity equals the slope of the Lorenz curve, which here represents the consumption distribution function, at the poverty line. Lorenz curves have been estimated for Indonesia and Thailand from their respective household consumption distributions.5The impact of rising unemployment on poverty is less in Indonesia than in the other two countries because households affected by job losses are relatively further away from the poverty line and can absorb a larger decrease in consumption before becoming poor.6A range of estimates is presented as there is bound to be an overlap between households affected by the across-the-board reduction in real income and those who become unemployed. However, the overlap is likely to be small because households affected by unemployment are assumed to be relatively distant from the poverty line, and therefore, they are not among the households who become poor as a consequence of the across-the-board real income decline.7Before the crisis, the monthly per capita poverty line was about $10 in Indonesia, $227 in Korea, and $28 in Thailand.8This extension in the coverage of the unemployment insurance scheme would raise eligibility to over 50 percent of the labor force (from 30 percent at present).

Spillovers from within the region have also affected Taiwan Province of China and Singapore, with growth slowing in both economies, especially Singapore. In Taiwan Province of China growth in 1998 is projected to remain positive on a calendar-year basis, while in Singapore output is expected to be flat. In both economies, strong macroeconomic and financial sector fundamentals have served to minimize adverse financial market reactions to developments in neighboring countries.

Since mid-1997, equity markets in Asia have fallen significantly relative to emerging markets in other regions (Figure 2.13). Not all of the decline in Asia, however, is attributable to the crises in southeast and east Asia. Thus, in India and Pakistan, following nuclear tests in May and June, exchange and equity markets reacted sharply to sanctions imposed by the United States and other countries, and to the prospect of cuts in bilateral and multilateral financial assistance. While the sanctions will have a larger impact on Pakistan owing to its greater dependence on official financing, they could also have an effect on the flow of private funds to India. Weakness in India’s exchange rate and equity markets intensified after the announcement of the government’s budget in early June, which, apart from increasing protection of the manufacturing sector, implied a continued loose fiscal stance. In reaction to the continued downward pressure on the Indian rupee, the central bank tightened monetary policy in late August. However, with the fiscal position remaining weak, there is a need for stronger measures, including reforms to strengthen the financial sector.

Figure 2.13.Developing Countries: Equity Prices

(U.S. dollar terms; logarithmic scale; January 1990 = 100)

Equity prices in Asia have fallen sharply relative to other emerging markets.

Source: International Finance Corporation, Emerging Markets Database.

In Vietnam, the regional crisis came at a time when domestic structural problems were already pointing to a slowing of economic growth in 1998. Reduced foreign direct investment from Asia, and eroding export competitiveness reflecting limited exchange rate flexibility, as well as emerging signs of stress in the banking system, have increased the vulnerability of the economy. In response, the authorities devalued the currency by 10 percent against the dollar in early August and have also begun cautiously to implement reforms in the banking sector and state-owned enterprises.

The Russian Crisis

In August 1998, Russia replaced Asia as the center of the financial crisis afflicting emerging markets. It had been subject to recurrent financial market pressures since the intensification of the Asian crisis in October 1997. These pressures reflected, in addition to financial contagion from Asia and the adverse implications of the decline in oil prices for Russia’s external balance, persistently large fiscal imbalances, heavy reliance on short-term foreign inflows, and delays in structural reform. The authorities responded through combinations of foreign exchange market intervention, interest rate hikes, and a widening of intervention bands. While these policies were successful in maintaining broad exchange rate stability, interest rates were left for long periods at extremely high levels that limited confidence in the sustainability of the response. Sharp declines in stock market values ensued. Following some easing of pressures in the spring, financial markets came under renewed intense pressures in mid-May. This new episode culminated in the full-scale crisis in August.

Unfolding of the August Crisis

The new wave of financial pressure in May 1998 was sparked by contagion from the worsening of the Asian crisis discussed above, but the placement of a major commercial bank under central bank administration and miners’ strikes over wage arrears were also contributory factors. More fundamentally, the pressure reflected growing perceptions among market participants that Russia’s fiscal position was unsustainable, and doubts about the implementation of required policies. In response, the central bank, in addition to intervening in the foreign exchange market, raised official interest rates in three steps, briefly to as high as 150 percent at the end of May. At the same time, the government announced revisions to the 1998 budget, including a cut in primary expenditures by 20 percent, and a number of initiatives to boost revenues. These policy actions and announcements temporarily eased tensions and allowed a partial reversal of the official interest rate hikes in early June.

In an acknowledgment that the actions taken in May and early June were insufficient to fully restore calm to financial markets, the authorities in late June adopted a more ambitious, three-pronged approach. First, the government unveiled a wide-ranging anti-crisis program, aimed at boosting tax revenues, cutting expenditures, and speeding up structural reforms. At the same time, the authorities intensified their campaign to strengthen tax collection, particularly from major companies in the energy sector, including by seizing assets. Finally, the government requested substantial additional official financial assistance to replenish international reserves and overcome liquidity problems stemming from the need to redeem short-term ruble-denominated debt held by foreign investors. However, market confidence did not return, and the central bank was forced to raise official interest rates again at the end of June.

To address the continuing difficulties, the authorities in mid-July introduced an additional policy package, in the context of an agreement with the IMF on an augmented Extended Fund Facility (EFF) arrangement. The package extended and strengthened the late-June anticrisis program and had three main elements: additional fiscal measures aimed at reducing the fiscal deficit; new structural reforms addressing the problem of arrears and promoting private sector development; and steps to reduce the vulnerability of the government debt position, including a voluntary restructuring of short-term treasury bills.

Box 2.5.Fiscal Balances in the Asian Crisis Countries: Effects of Changes in the Economic Environment Versus Policy Measures

The resolution of the crises in Asia has required monetary policy in the countries concerned to be geared largely toward stabilizing exchange rates. Thus, to a larger extent than usual, demand management has been delegated to fiscal policy, and actual and projected (or targeted) fiscal deficits have widened by much more than was envisaged before the scale of the downturns in activity and the structural problems needing to be addressed became apparent. To understand the role that fiscal policy has played in Indonesia, Korea, the Philippines, and Thailand during the recent economic turmoil, it is important to identify the factors that have influenced fiscal balances and, in particular, to distinguish the endogenous effects of the changes in the macroeconomic environment associated with the crisis from the effects of policy measures implemented in the context of programs supported by IMF financing.

In the course of the crisis, the economies of east and southeast Asia have experienced very large adjustments in their exchange rates, output, and price levels. Many facets of their economic and financial structures, both real and nominal, are likely to be very different once they have recovered. In this box, year-to-year changes in the fiscal balance are ascribed to various elements of the changed economic environment and to different categories of policy actions, all defined in a consistent way across countries.1

Changes in the economic environment comprise exchange rate movements, developments in international oil prices, and changes in the rate of real income growth. Some of these changes may have multiple effects on the budget balance, some of which may be offsetting. For example, a depreciation of the exchange rate will tend to increase expenditure on foreign-currency-denominated goods and services, but it will also tend to increase the local currency value of imports, thereby raising revenues from import duties. In some cases, however, such revenues may also be affected by changes in the composition of imports in favor of goods subject to lower duties. The figures reported here (see table) are net, aggregating across revenues and expenditures.

Determining what constitutes a policy action is judgmental. In part, this is because any form of policy inaction can alternatively be viewed as a deliberate policy action to maintain the status quo. This problem is particularly acute in the area of expenditure policy. An improvement in the fiscal balance that results from, say, a decline in the wage bill relative to GDP in the wake of a crisis-induced inflation could be regarded as a deliberate, contractionary policy to reduce the real wages of civil servants by denying them a cost of living adjustment. Alternatively, it could be labeled as policy inaction, on the grounds that nominal wages remain unchanged. The associated improvement in the fiscal position in the latter case would be attributed wholly to the changed economic environment. Similarly, a large increase in expenditure on subsidies that arises from a failure fully to adjust local prices of imported goods that are subject to price controls may be viewed either as a result of the changed economic environment (specifically, the depreciation of the exchange rate) or as a result of an explicit policy decision to expand social spending. Where possible, the calculations in this box use information about the authorities’ intentions to distinguish the results of policy decisions from the results of changes in the economic environment. For subsidies, however, increases in expenditure have been ascribed to changes in the economic environment.2 All of the analysis is based on program data or IMF staff projections available in mid-July 1998.

The analysis (summarized in the table), shows that, in all cases, the deterioration in the economic environment produced—or is expected (in the context of IMF programs) to contribute substantially to—deteriorations in fiscal positions. In Indonesia, changes in the economic environment are expected to cause the fiscal deficit to widen by as much as 11.1 percent of GDP in 1998/99. The corresponding effects in Thailand, the Philippines, and Korea are sizable but much smaller (3.1 percent of GDP, 1.5 percent of GDP, and 1.5 percent of GDP, respectively, in 1998) because the exchange rate depreciation and output decline have been more modest. The principal source of the deteriorations is the massive depreciations in exchange rates, particularly in the case of Indonesia (mainly through its impact on the cost of commodity subsidies). Weaker international oil prices are affecting the fiscal balance only of Indonesia, with an associated revenue shortfall of 0.7 percent of GDP in 1998/99. The decline or deceleration of real GDP worsened fiscal balances by 4 percent of GDP in Indonesia (1998/99), but by less than 1 percent of GDP in all the other cases.

Although all the countries took policy measures (or allowed policy inaction) to contain the deterioration in their fiscal balances, in some countries these have been outweighed by policy measures in support of the banking system and by additional social outlays that have increased budget deficits. In Indonesia, the country where the deterioration in the macroeconomic environment has had the largest impact on the budgetary position, the net impact of policy changes has been to offset part of the increase in the deficit resulting from the deterioration in the economic environment; in the Philippines, policy measures are projected to accommodate the effects of the deterioration in the economic environment.3 In contrast, in Korea and Thailand the net impact of policy measures has been to widen the fiscal deficit. Support for ailing financial systems accounts for a significant portion of discretionary policy measures.4 Measures to strengthen the social safety net have increased fiscal deficits the most in Korea. While Indonesia, the Philippines, and Thailand (in 1997/98) cut other budgetary outlays, Korea and Thailand (in 1996/97) actually increased them.

Changes in Fiscal Balances(Percent of GDP; a negative number indicates a deterioration in the fiscal balance)
Fiscal balance1-0.9-10.1-4.0-1.6-5.1-0.9-2.1
Change in fiscal balance-2.2-9.2-4.0-4.0-3.5-0.4-1.2
Change due to economic environment-4.2-11.1-1.5-0.3-3.10.7-1.5
Exchange rate2-3.5-6.4-0.9-0.2-2.0-0.9
GDP growth-0.5-4.0-0.6-0.1-0.90.6-0.6
Oil price-0.2-0.7
Policy changes2.71.7-2.5-2.6-0.6-0.61.6
Social safety net3-1.0-2.1-0.6-0.2
Bank restructuring-1.6-1.4-0.7-2.0
Statutory revenue change0.51.8-0.7-0.1-0.7
Residual (unexplained)0.70.2-1.10.1-0.5-1.3
Nominal GDP growth rate19.843.
Real GDP growth rate4.6-12.15.5-5.0-0.44-
Fiscal impulse6-2.1-

This measure of the balance excludes privatization proceeds but includes bank restructuring costs. Data are on a fiscal year basis. For Indonesia, the fiscal year runs from April through March; for Thailand, it runs from October through September. In Korea and the Philippines, fiscal years correspond to calendar years. Data for 1998, or 1998/99, are projections under IMF programs and are subject to revision.

Includes the impact on the fiscal balance of increases in domestic interest rates in response to currency depreciation.

This excludes the effect of exchange rate changes on subsidies arising from the failure to fully adjust commodity prices. These effects are included above as effects from “economic conditions.” In some cases, notably Indonesia, a strong case could be made for treating the increase in subsidies as a result of policy changes.

Refers to calendar year 1997.

Refers to calendar year 1998.

From Lorenzo Giorgianni, “The Fiscal Stance in Thailand and Other Countries: 1991–1998” (unpublished; Washington: IMF, April 1998). The fiscal impulse for Indonesia has been changed to reflect data available in mid-July 1998.

This measure of the balance excludes privatization proceeds but includes bank restructuring costs. Data are on a fiscal year basis. For Indonesia, the fiscal year runs from April through March; for Thailand, it runs from October through September. In Korea and the Philippines, fiscal years correspond to calendar years. Data for 1998, or 1998/99, are projections under IMF programs and are subject to revision.

Includes the impact on the fiscal balance of increases in domestic interest rates in response to currency depreciation.

This excludes the effect of exchange rate changes on subsidies arising from the failure to fully adjust commodity prices. These effects are included above as effects from “economic conditions.” In some cases, notably Indonesia, a strong case could be made for treating the increase in subsidies as a result of policy changes.

Refers to calendar year 1997.

Refers to calendar year 1998.

From Lorenzo Giorgianni, “The Fiscal Stance in Thailand and Other Countries: 1991–1998” (unpublished; Washington: IMF, April 1998). The fiscal impulse for Indonesia has been changed to reflect data available in mid-July 1998.

This analysis also illustrates that, even in the unlikely event that real output returns to its former growth paths, fiscal positions would not be restored to precrisis levels. Rather, it appears that the evolution of the exchange rate (in particular, the extent to which the exchange rate recovers to its precrisis level) and the way that various expenditure policies are reconfigured in the aftermath of the crisis will also have a significant impact on the postcrisis fiscal balance. Of course, so will the legacy of the financing costs of addressing the immediate effects and causes of the crisis.

Thus, the analysis indicates that exchange rate depreciations have tended to produce more important effects on fiscal balances than declines in output growth, that discretionary policy measures have tended to induce smaller impacts on budget positions than deteriorations in the macroeconomic environment, and that the net effect in all cases has been to move the fiscal balance significantly toward deficit.

These are the results as they appear from the perspective of July 1998, one full year into the Asian crisis. The economic situation viewed from this standpoint, however, is much worse than had been expected by national authorities, the IMF, or private forecasters when the crisis started in the summer of 1997: no one anticipated the extent to which the economic situtation would deteriorate. Indeed, the average of private forecasts for these economies moved markedly and progressively downward throughout the intervening year (see figure). Fiscal policies adjusted quite substantially to this unanticipated deterioration; and the policies in place as of July 1998 were, appropriately, quite different from those that had been planned earlier. This is well illustrated by the case of Thailand, the country that first entered the crisis.

When Thailand’s stabilization and reform program was initially agreed with the IMF in mid-August 1997, the growth of the Thai economy was projected to slow considerably but nonetheless to remain significantly positive. Thus, the initial program called for measures to strengthen the fiscal position, to enhance the credibility of the government’s policy program, and thereby help to stabilize foreign exchange and financial market conditions; to contribute to an orderly reduction in Thailand’s large current account deficit; and to make fiscal provision for the large expected public sector costs of cleaning up the financial system. By the autumn of 1997, expectations for the growth of the Thai economy had declined further for both 1997 and 1998, but they were still positive. Conditions in the foreign exchange and financial markets, however, had not stabilized. Additional fiscal measures were adopted by the Thai authorities, with IMF support, to strengthen the credibility of the adjustment program.

In the ensuing months, however, it became apparent that the economic crisis in Thailand (and elsewhere in Asia) was much deeper than expected earlier and that projected growth for 1998 had turned modestly negative. Also, the Thai trade and current accounts had shifted massively toward surplus, owing primarily to import compression, and by February 1998 a meaningful recovery of the exchange rate of the baht had been achieved. The fiscal measures of the preceding autumn were essentially reversed as fiscal policy moved to support the economy. As the outlook for the economy continued to deteriorate, the projected fiscal deficit widened, owing to the operation of the (relatively weak) automatic stabilizers and because of policy efforts to cushion the output decline and provide support to the needy (see Box 2.4)—within the limits of what can be financed without destabilizing foreign exchange financial markets.

Evolution of the Consensus Forecast for 1998 Real GDP Growth


Source: Consensus Economics, Inc., Asia Pacific Consensus Forecasts.

For Indonesia and Korea, current account deficits going into the crisis were much smaller than for Thailand, and the initially agreed fiscal tightenings to enhance credibility and provide for financial sector cleanups were roughly half the size (relative to GDP) of that in Thailand. As it became clear that the economic situation in these countries would also turn out much worse than initially expected, the planned fiscal stance has shifted substantially toward deficit. For Indonesia, which has fallen into deep crisis, large external support has been arranged to help to finance the deficit in a manner that does not further destabilize financial markets. Even with such support, tight discipline on less essential components of government spending is needed to keep the fiscal deficit (estimated to reach 10 percent of GDP) within the bounds of available finance.

1This approach is different from the standard approach used for measuring the impact on the budgetary balances of changes in macroeconomic conditions in industrial countries. The latter approach limits the endogenous effects to those of fluctuations in output around a trend growth path. This type of analysis has been modified for the exercise described here to recognize the effects of changes in exchange rates and other macroeconomic variables in the Asian context.2As noted below, this significantly affects the calculated net impact of policy changes for Indonesia.3In 1998/99, classification of the increase in subsidies in Indonesia as a policy decision, rather than an effect of the change in the economic environment, would result in the net impact of policy changes being expansionary.4Bank restructuring costs in this analysis are limited to the carrying costs of any net increase in debt associated with support of financial institutions. Because countries have adopted different support strategies, and different budgetary treatments for recording costs, comparisons of current budgetary costs across countries will not necessarily provide a good indicator of the relative severity of each country’s financial sector difficulties.

The announcement of the package, and its approval by the IMF on July 20, initially had positive effects on Russian financial markets and eased pressures on the ruble. Equity prices rebounded by over 30 percent, average treasury bill rates fell from more than 100 percent to below 50 percent, and the central bank, on July 24, lowered its refinancing rate from 80 to 60 percent. In conjunction with the approval of the augmented program and to reduce short-term debt-servicing costs, the Ministry of Finance swapped treasury bills with a market value of $4.4 billion into long-term Eurobonds, placed additional Eurobonds worth $500 million, and shifted its borrowing in rubles to longer maturities, introducing new bonds with maturities longer than one year.

The improvement in financial market conditions was short-lived, however. Confidence was weakened particularly by the lack of support for the program in the Duma, which forced the President to veto several measures approved by the Duma and to implement a number of other measures by decree, and led the IMF to reduce the amount of a disbursement from the $5.6 billion originally planned to $4.8 billion. Also, in the key energy sector, major producers voiced strong opposition to the program, and the collection of overdue tax payments from a number of oil companies proved difficult. Finally, the government-owned Sberbank took a decision not to roll over its sizable treasury bill holdings falling due in the last two weeks in July, forcing the government to borrow at relatively high rates to cover its debt service, and shaking the confidence of other investors in the government’s ability to roll over its debt in the future. As a result, from the last week of July, treasury bill rates rose and equity prices fell again, and the Ministry of Finance was forced to cancel the auctions for its new longer-term bonds three weeks in a row because of prohibitively high borrowing rates. The ruble also came under downward pressure, forcing the central bank to intervene on a large scale. Pressures intensified in the second week of August and spread to the banking sector. By August 14, with average treasury bill rates at around 300 percent, international reserves down to around $15 billion (from $18½ billion following the July IMF disbursement), and banks unable to meet payment obligations, Russia was facing a full-scale banking and currency crisis.

The authorities responded to the crisis by announcing a number of measures on August 17, including a change in exchange rate policy entailing a de facto devaluation; a unilateral conversion of short-term ruble government debt and suspension of trading in the domestic treasury bill market; a 90-day moratorium on the payment of many private sector foreign currency obligations; a strengthening of controls on capital flows; and steps to stabilize the banking sector, including interbank settlements. The new exchange rate policy moved the band to 6.0–9.5 rubles per U.S. dollar from 5.3–7.1 rubles, thus signaling a willingness to let the ruble depreciate significantly, and introduced more flexibility within the band. The other measures were aimed at supporting the new exchange rate policy, easing the budget’s cash-flow situation, and protecting the banking sector.

Market reaction to the new measures was extremely unfavorable, with negative sentiment exacerbated by uncertainty regarding the details of the debt conversion scheme and the moratorium. In the days that followed the announcement, the exchange rate vis-à-vis the U.S. dollar fell by more than 10 percent, equity prices dropped sharply, and the spread on Russian U.S. dollar Eurobonds rose by more than 1,000 basis points to over 3,000 points. In response to the strong downward pressure on the currency, the Russian central bank raised its overnight rate to 250 percent and again intervened heavily in the foreign exchange market, with international reserves falling to around $13½ billion by August 21. The financial crisis intensified following the dissolution of the government, the subsequent political uncertainty, and the (delayed) announcement of the terms of the domestic debt conversion. The conversion conditions, which broadly offered the choice to exchange treasury bill holdings for either ruble bonds with maturities of three, four, and five years with coupons of 20–30 percent, or (up to a limit of 20 percent of the total holding) eight-year U.S. dollar-denominated bonds with a 5 percent return, were seen as implying major losses for investors.

On August 26, following heavy intervention, the central bank announced it would stop selling U.S. dollars and suspended trading in the ruble on the country’s main exchanges. In the interbank market, the ruble subsequently began to trade in the 10–12 rubles per dollar range, well beyond both the announced official rate and the upper bound of the new band. Also in late August, continued severe pressures on the banking sector led to the announcement of two major bank mergers and the formation of “payment pools” between banks, while the central bank withdrew the license of a large bank and imposed a temporary administration on the country’s second-largest deposit bank (SBS-Agro Bank). On September 2, finally, the central bank abandoned the exchange rate band and let the ruble float.

Financial markets remained unsettled and highly volatile in the first two weeks of September, reflecting uncertainty about the approval of a new prime minister and the future course of economic policies. The ruble exchange rate made sharp swings, while equity prices weakened further. Following parliamentary approval of a new prime minister in mid-September, markets stabilized somewhat, with the ruble trading at around 15 rubles per dollar, close to 60 percent lower against the dollar than in early August. As pressures on the banking sector continued, the central bank initiated an arrangement whereby banks would transfer accounts of assenting depositors to Sberbank—whose deposits are state guaranteed—and began to provide additional liquidity to ailing banks. The authorities also announced their intention to revise some of the terms of the August 17 domestic debt conversion, involving a larger initial cash offer, shorter terms for the ruble securities on offer, and better conditions for the Eurobond option.

Background to the Crisis: Fiscal Problems

The recurrent financial market pressures in Russia since late 1997 and the August 1998 crisis are attributable to a combination of serious remaining weaknesses in economic fundamentals, especially in the fiscal area; unfavorable developments in the external environment; and the country’s vulnerability to changes in market sentiment arising from the financing of the balance of payments and the budget through short-term treasury bills and bonds placed in international markets. A number of shortcomings in economic reform have set the context. In addition to incomplete reforms in the structural area, chronic fiscal imbalances and poorly functioning tax and expenditure management systems have continued to present major problems. As discussed in the May 1998 World Economic Outlook (Chapter V, including Box 9), Russia has made insufficient progress in improving tax systems and budget procedures, establishing fully competent agencies to collect taxes and control expenditures, clarifying intergovernmental fiscal relations, and introducing accountability and transparency at all levels of government operations. Following a substantial revenue shortfall in 1997, the revenue performance at the federal level was again disappointing in the first half of 1998, with expected conversions of arrears and noncash revenue into cash payments not materializing and the decline in international oil prices reducing tax contributions from the oil sector. Reflecting the poor revenue performance, the federal government deficit amounted to about 5 percent of GDP (Table 2.8). Local governments and the pension fund also ran large deficits, resulting in a deficit of the enlarged government of close to 10 percent of GDP for the first half of 1998, with wage and payments arrears rising at all levels of government.

Table 2.8.Russia and Ukraine: Selected Macroeconomic Indicators
19961997First Half
Real GDP growth (percent)-5.00.9-0.51
Consumer price inflation (percent)221.811.06.4
Public finance (federal government)3
Primary balance-3.4-2.5-0.7
Money and credit growth (percent)2
Base money26.225.7-2.3
Broad money29.629.55.7
Interest rates (percent)
Refinance rate499.831.958.5
Bank loan rate4103.042.336.7
External sector
Current account balance30.6-0.1-1.0
Gross international reserves515.317.813.7
Exchange rate: rub/$
Real GDP growth-10.0-3.20.21
Consumer price inflation (percent)239.710.16.7
Public finance (general government)3
Primary balance-1.6-3.8-2.2
Money and credit growth (percent)2
Base money37.944.618.8
Broad money35.133.921.2
Interest rates
Refinance rate460.324.843.8
Bank loan rate479.649.147.3
External sector
Current account balance3-2.7-2.6-4.1
Gross international reserves52.02.41.8
Exchange rate: rub/$

Change from the same period in the previous year.

Twelve-month change at end of period.

In percent of GDP.

Period average.

Billions of U.S. dollars, end-period.

Change from the same period in the previous year.

Twelve-month change at end of period.

In percent of GDP.

Period average.

Billions of U.S. dollars, end-period.

The retreat of investors from emerging markets compounded the fiscal problem. Until the early autumn of 1997, substantial short-term capital inflows had played a key role in driving interest rates on Russian treasury bills to below 20 percent. Easy access to foreign financing led to the development of financing strategies that relied heavily on a combination of short-term treasury bill sales, targeted at domestic and foreign financial institutions, and borrowing in the international markets; this resulted in considerable vulnerability to changes in market sentiment. In the wake of the Asian crisis, foreign investors began to reduce their exposure to the Russian markets, and they returned in early 1998 only when offered yields well above precrisis levels. The combination of high yields and the short maturity of the treasury bills, however, raised concerns that the government would not be able to meet the equivalent of around $1.5 billion in debt service falling due each week in the remainder of 1998. In the face of deteriorating market sentiment, beginning in June, domestic borrowing to finance the federal budget came to a virtual halt, with the government making large net repayments of treasury bills (Figure 2.14).

Figure 2.14.Russia: Monetary and Financial Market Developments

As pressure in Russia’s financial markets intensified in the late spring of 1998, international reserves fell, interest rates rose sharply, and net financing from treasury bill issues turned negative.

Sources: Central Bank of Russia; and IMF staff estimates.

1The overnight interbank interest rate was 190 percent on September 25, 1998.

2The Russian treasury bill market was suspended as of August 17, 1998.

The decline of confidence in the authorities’ ability to bring the fiscal situation under control and to roll over the treasury bills that had not been swapped into Eurobonds was the main immediate cause of the August 1998 crisis. The augmented program of July 1998 was built on a number of assumptions regarding revenues, expenditures, and financing flows for the second half of 1998 and 1999 which, if realized, would have allowed an improvement in the country’s fiscal position and an easing of financial market tensions. However, the program required the passing into law, ahead of IMF approval on July 20, of a series of measures needed for the achievement of revenue and expenditure targets; and it was also based partly on the assumption that confidence would improve such that official treasury bill holders would be willing to roll over all, and private domestic and foreign treasury bill holders would be willing to roll over at least part of, their treasury bill holdings falling due in the remainder of 1998. As financial market participants began to see that the program’s revenue and expenditure measures were not being implemented fully, and following early indications that many treasury bill holders, including the government-owned Sberbank, were demanding payment as obligations fell due, there was a growing perception of a likely financing shortfall. Expectations that it was only a matter of time before the government would run out of resources to meet its gross debt-service payments and demands for foreign exchange led to intensifying financial market pressures from the end of July, undermining the program within weeks after its approval.

A key trigger of the loss of confidence was the failure of the Russian government to obtain parliamentary approval of the fiscal measures planned under the program. Substitute action in the form of presidential decrees could not pass into law some of the missing measures needed to achieve the program’s budget targets, in the areas of personal income taxation and pension fund financing in particular. While the government announced a special Duma session in August to seek passage of remaining measures, uncertainty remained about their adoption. Further, in the absence of signs of a substantial intensification of revenue collection efforts,4 skepticism remained regarding the prompt and full implementation of even those measures that had been approved. In addition to the lack of confidence in the implementation of the anticrisis program, investor sentiment was hit by the further reassessment of risks in emerging markets. Another critical factor was the continued large-scale expansion of credit in July-August by the central bank, as it sought to protect Russian commercial banks. More technical factors also played a role. With a total of $8.6 billion of Russian Eurobonds issued in June–July, a supply overhang in these instruments emerged, as reflected in a widening Russian spread relative to other emerging market economies.

Implications for the Economy

The consequences of the August crisis for the Russian economy will depend crucially on the policy responses. The authorities’ immediate task is to stabilize financial markets, restore confidence, and prevent a further decline in the value of the currency. Even if they manage to do so quite quickly, however, the crisis and the measures announced in August and September will have important consequences, including for growth and inflation.

Growth in the Russian economy, which was positive in 1997 after eight years of deep output decline, again turned slightly negative in January–July 1998 (by comparison with the same period of 1997), reflecting the unfavorable financial environment. High interest rates impeded bank lending to the private sector and discouraged fixed investment, which was more than 5 percent lower in the first seven months of 1998 than a year earlier, following a cumulative drop by more than 75 percent in 1991–97. Growth is expected to be further adversely affected in the remainder of 1998 and in 1999 by declines in consumption and investment associated with the collapse of confidence, the sharp decline in real wealth, severe financing constraints, and disruptions in the financial and payments systems. Inflation, which on a 12-month basis had fallen to below 6 percent in July, is expected to rise quickly, owing to the rapid pass-through of the currency depreciation, reflecting not only the high share of imported goods in the consumer goods basket, but also the high degree of dollarization in the economy. According to initial estimates consumer prices rose by 15 percent in August at a monthly rate, compared with a 0.2 percent rise in July, and by as much as 43 percent in the first two weeks of September. In the retail sector, some stores appear to have reintroduced the practice of quoting U.S. dollar-based ruble prices.

The impact of the crisis and the August 17 measures on the banking sector is likely to be severe. One private international bank has referred to estimates that Russian banks’ on-balance-sheet net external foreign-currency-denominated liabilities amount to $5 billion, with an additional $5 billion off-balance-sheet forward exposure; and there are other estimates considerably higher than these. The banking sector is therefore heavily affected by the depreciation of the ruble, with the 90-day external payments moratorium offering only temporary relief.5 The sharp fall in interbank market activity, payment delays, and the freezing of the banks’ treasury bill portfolio have created severe liquidity management problems for most banks. The consolidation observed in late August among the sector’s still more than 1,500 banks is expected to accelerate, and only the systemically important and strongest banks seem likely to receive liquidity support from the Central Bank of Russia and to be in a position to renegotiate foreign currency payment obligations.

The depreciation of the ruble and the other measures announced on August 17 also have important consequences for the structural reform agenda. Government intervention in the banking sector and financial markets has increased, and the sale of state holdings in a number of major oil and energy companies announced for the remainder of 1998 is likely to be postponed, as foreign investor interest has diminished and domestic investors now lack the financial resources. More generally, the August unilateral domestic debt conversion extends the practice of nonpayment to holders of financial instruments and thereby reinforces the culture of nonpayment. The conversion is expected to stimulate further the buildup of arrears, barter arrangements (which now account for close to 50 percent of industrial sales), and dollarization.

How Have Financial Market Conditions in Other Emerging Market Economies Been Affected?

Other Transition Economies

Among other countries in transition, the effects of the Asian and Russian crises have been felt most strongly in Ukraine, which has close financial and trade ties to Russia and has been suffering from similar underlying structural and fiscal problems and associated vulnerability to changes in investor sentiment (see Table 2.8). In the face of poor revenue collection and inadequate expenditure control, the Ukrainian government in 1997 increased sharply its issuances of treasury bills, around half of which were taken up by foreign investors, and borrowed short-term from the international markets. In the wake of the Asian crisis, foreign investors began to redeem maturing bills, in spite of a marked increase in yields from around 20 percent in October 1997 to more than 50 percent in early 1998. Confronted with a substantial rise in its gross financing needs, the government issued Eurobonds twice and relied increasingly on treasury bill purchases by the National Bank of Ukraine, which by April 1998 had become the largest participant in the primary treasury bill market.

The renewed pressures in Russia from mid-May quickly spilled over to Ukraine, and official interest rate hikes in Russia were followed in Ukraine in late May and early July, while the currency and equity prices weakened. In August, in the wake of the Russian crisis, the financial situation deteriorated further. An additional increase in the Lombard rate and significant foreign exchange market intervention, which reduced usable reserves to very low levels, could not prevent the hryvnia from moving outside its 2¼ per U.S. dollar band in the interbank market. On September 4, in the context of an agreement with the IMF on an EFF arrangement, the authorities announced a policy package including adoption of a new 2½–3½ hryvnia per U.S. dollar exchange rate band, the implementation of tighter fiscal policies, and a rescheduling of domestic debt involving an arrangement to swap part of the short-term debt for longer maturities. The rescheduling is intended to be voluntary and provides, for nonresident investors, the option to exchange treasury bills for U.S. dollar Eurobonds yielding 20 percent. As of mid-September, conditions in the Ukrainian treasury bill and equity markets remained unsettled, and the currency was under further pressure as investors awaited the results of the debt rescheduling and sought reassurances that the September 4 package was being fully implemented.

Financial markets in the Czech Republic, Hungary, and Poland, which had been temporarily affected by the Asian crisis, have faced renewed, and significant, pressure since the onset of the Russian crisis.6 The currencies of all three countries had come under pressure in late 1997, with stock markets also recording significant declines. In early 1998, the currencies stabilized and, in the Czech and Polish cases, appreciated significantly, especially in real terms, while equity markets recovered, partially in the Czech Republic and more than fully in Hungary and Poland. In August and early September, however, the koruna depreciated by more than 5 percent vis-à-vis the deutsche mark, the forint fell to the bottom of its trading band, and the zloty moved from close to the upper limit to the middle of its trading band. The weakening of the Czech and Polish currencies eased concerns about excessive appreciation earlier in the year. As foreign investors reduced exposure to countries in transition, equity prices also fell sharply from their mid-July peaks, by almost 50 percent in Hungary and by 25–30 percent in the Czech Republic and Poland, while government bond yields in Hungary and Poland rose significantly. The authorities in all three countries signaled their intention not to let these financial market pressures affect the longer-run stance of interest and exchange rate policies. In the Czech Republic official interest rates were lowered in mid-August; in Hungary the monthly rate of depreciation of the crawling peg was reduced from 0.8 to 0.7 percent in late August; and in Poland a reduction in official money market rates accompanied a reduction in the monthly devaluation rate of the zloty from 0.65 percent to 0.5 percent in early September. In Hungary, however, as pressure on the currency continued in mid-September, the central bank raised its operating rate by a full percentage point.

In other countries in transition, the effects of the Asian and Russian crises have differed depending on the degree of development and international integration of local financial markets, the scale of imbalances, and trade and financial links with Russia.

In the three Baltic countries, the effects of the crises have been felt mostly in equity markets, which reached historic lows in mid-September, in Estonia and Latvia at about one-fourth their values in September 1997. In Latvia, where according to some estimates more than 8 percent of banks’ assets are exposed to Russian risk, the banking sector has also been affected, with a major bank holding significant amounts of Russian treasury bills suffering a run on deposits in late August. In Romania and the Slovak Republic, where restrictions on short-term capital flows have been maintained, currency and financial market pressures have mainly reflected country-specific factors, including rising current account and fiscal imbalances and political uncertainties. Finally, the countries of the former Soviet Union other than Ukraine, which had been relatively untouched by the Asian crisis, have felt significant pressures on exchange rates, Eurobond spreads, and domestic interest rates as a result of the Russian crisis.

The Asian crisis occurred while transition countries were rapidly gaining access to international financial markets. By the end of 1997, 14 transition countries had received credit ratings from at least one of the major agencies, compared with only 4 countries two years earlier. Gross medium- and long-term funds raised in the international bond and loan markets by the transition countries increased from less than $8 billion in 1995 to almost $30 billion in 1997 (in spite of a slowdown in the last quarter), while foreign portfolio investment rose from less than $4 billion to more than $19 billion in the same period. There was, in fact, a surge in portfolio investment in 1997, largely into short-term treasury bills in Russia and Ukraine. Overall, while total net private capital flows to all developing and transition economies fell by more than a third to $123 billion in 1997 from $215 billion in 1996, net flows to transition countries rose to $23 billion from $16 billion. In a number of countries in transition, the creation and opening up of new financial markets as part of the transition process may have motivated flows in excess of what was warranted in relation to the overall level of private capital flows to emerging markets before the Asian crisis and to underlying economic and financial conditions in the transition countries.7 The vulnerability to external financial shocks of countries such as Russia and Ukraine, which were notable destinations for investors in search of high yields, was thereby increased.

The effects of the Asian and Russian crises on financial markets in the economies in transition have been significant, but the broader economic repercussions are expected to remain limited, at least in the short run. Because of the relative insensitivity of banks’ lending rates to monetary policy (reflecting the low degree of competition in the banking systems), the still limited role of banking sector credit, and the low leverage of the corporate sector in transition countries,8 the effects of higher official interest rates on aggregate demand and economic activity tend to be relatively small. Similarly, given the still low levels of stock market capitalization, even sharp drops in equity prices tend not to have sizable direct effects on the real economy. In the short run, therefore, growth prospects in the transition countries are unlikely to be affected substantially by recent interest rate and equity price movements. However, their longer-term growth prospects may suffer significantly if the financial turbulence slows down the development of financial markets and banking sectors that should be playing an increasing role in intermediating the financial resources needed for further restructuring and new investment.

Latin America

After weathering the fallout from the Asian crisis that hit their financial markets in October 1997, with the help of tighter monetary and fiscal policies (as discussed in the May 1998 World Economic Outlook), the Latin American economies faced renewed financial market pressures in the wake of the Russian crisis. The pressures were exacerbated by falling commodity prices and downgraded credit ratings for Brazil and Venezuela. Sovereign bond spreads widened dramatically in late August and early September, reaching levels not seen since the “tequila crisis,” while equity prices fell by 40 to 60 percent between mid-July and early September before recovering somewhat. Authorities in the region generally again responded by tightening monetary and fiscal policies, and, in some cases, intervening in the foreign exchange markets. In early September, Colombia and Ecuador adjusted the trading bands of their currencies, effectively devaluing them by 9 and 15 percent, respectively. While contagion from Asia and Russia and the repercussions on fiscal and external positions may in part explain the financial market pressures that arose, country-specific factors, including macroeconomic imbalances and structural weaknesses unrelated to the crisis, as well as political uncertainties, also played a role.

In Brazil, short-term interest rates had been lowered to pre-October 1997 levels by the end of July as international reserves increased. Capital inflows had been sustained in part by private sector bond issues, the privatization program, and high real interest rates. But the economy remained vulnerable on account of a large and widening fiscal deficit, a large current account deficit, and a sizable stock of short-term public debt, increasingly and largely indexed to overnight interest rates or the U.S. dollar. In the wake of the Russian crisis and the subsequent reassessment of risk by investors, private capital outflows accelerated, equity prices fell sharply, and sovereign bond yield spreads increased by around 1,400 basis points. The authorities responded by intervening heavily in the foreign exchange market to defend the real, raising the benchmark interest rate in two stages from 19 to almost 50 percent in early September, relaxing restrictions on short-term capital inflows, promising to increase primary fiscal surpluses after 1998, and announcing fiscal cuts for 1998 partly to offset the increase in debt-servicing costs related to higher interest rates.

In Venezuela, as discussed earlier, economic activity has contracted during 1998, owing in large measure to the decline in world oil prices. The fiscal position has shifted from a surplus in 1997 to a large deficit in 1998, and the current account has moved into deficit. Exchange market pressures intensified sharply with the Russian crisis. In response, the authorities intervened to support the exchange rate, while allowing the bolivar to drop to the lower end of its band, tightened liquidity conditions, and announced additional measures aimed at reducing the fiscal deficit.

In Argentina, by the end of July interest rates had declined to pre-October 1997 levels with an easing of financial market pressures attributable in part to a moderation of growth and tight fiscal policy, although the widening current account deficit and slow progress with labor market reforms remained sources for concern. During August and early September, however, pressures reemerged, with equity prices falling sharply and spreads on stripped Brady bonds increasing by over 600 basis points. Domestic interest rates increased by smaller amounts as gross liquid reserves at the central bank fell only slightly, reflecting in part the credibility of the currency board as well as banking sector reforms, improved public debt management, and recent measures to reduce fiscal spending in 1998 and freeze public spending in 1999. Nevertheless, the economy, which has substantial trade exposure to Brazil, remained vulnerable.

In Mexico, contagion from other emerging markets together with concerns about oil price declines, the widening current account deficit, and the fragility of the banking system have also been reflected in financial market pressures in recent months. These concerns have been partially allayed, however, by a tightening of monetary policy and the prospect of tax reform aimed particularly at boosting non-oil revenue. Equity and sovereign bond prices have fallen sharply, but by less than in other parts of Latin America, reflecting a more flexible exchange rate—the peso has depreciated by over 10 percent since July and by about 25 percent since the beginning of the year—a smaller current account deficit, and stronger links to the still buoyant U.S. economy.

In Chile, the decline in copper prices and the country’s substantial exposure to Asia through trade, along with the turmoil in other emerging markets, have significantly worsened short-term prospects for growth and the external current account and reduced capital inflows, putting downward pressure on the exchange rate. A tight monetary stance and a package of measures introduced in June—which included a reduction in public expenditure, a narrowing of the exchange rate band from 25 percent to 5½ percent combined with a faster rate of crawl of the band, the issuance of medium-term dollar-indexed bonds, and a lowering of the unremunerated deposit on certain capital inflows—helped to reduce the pressure on the currency. In mid-September, following the subsequent reemergence of pressures, official interest rates were increased, the deposit requirement on capital inflows was reduced to zero, and the trading band for the peso was widened to 7 percent, with allowance for a further gradual widening to 10 percent by end-1998, in a further effort to reduce volatility in interest rates and stem capital outflows.

Other Developing Countries

Most countries in the Middle East and Africa had experienced relatively little financial market contagion from the Asian crisis. Recent developments in Russia have had a larger impact, although not uniform across all countries, as equity prices have plummeted and sovereign bond spreads have risen sharply. In Egypt, the stock market has weakened significantly during 1998, reflecting more the effects of the decline in oil prices and the decline in tourism earnings than direct contagion from Asia or Russia. The exchange rate has not come under more than moderate pressure in recent months, while equity prices have fallen less since July than in other emerging markets. In contrast, financial markets in Turkey, which has strong trade links with Russia, came under strong pressure in the wake of the Russian crisis, with the large-scale capital inflows that had been attracted earlier in the year by high interest rates and improved prospects of disinflation being partially reversed. In Africa, financial market turbulence has been most in evidence in South Africa. The rand, which has been on a depreciating trend since early 1996, came under particularly intense downward pressure during May–July and again in August. The authorities responded by tightening monetary conditions in both instances. After vigorous intervention in the foreign exchange market earlier in the year that left net international reserves at a precarious level, the authorities intervened little during the latest turmoil. Although external developments and depressed commodity prices contributed to the pressure on the rand, the persistent sluggishness of domestic growth, high unemployment, and inflation above partner-country levels have been important sources of vulnerability.

Prospects for Global Flows of Funds and Current Account Balances

Associated with the Asian crisis have been substantial shifts in international financial flows, with important implications for adjustments of external positions by many countries—both those in crisis and their partners. With the crisis in Russia and contagion to other countries, further adjustments to financial and trade flows will be needed. In 1997, net private capital flows to emerging markets (defined here as all developing and transition countries, together with the newly industrialized Asian economies) are estimated to have fallen by $91 billion from the record level of $215 billion reached in 1996, with Asia more than fully accounting for the decline (Table 2.9). In all other developing country regions, and the transition countries as a group, net inflows continued to rise last year, although there was a notable slowing of the growth of inflows to the developing countries of the Western Hemisphere.

Table 2.9.Developing Countries, Countries in Transition, and Newly Industrialized Asian Economies: Net Capital Flows1(Billions of U.S. dollars)
Net private capital flows313.5144.2155.7195.3214.9123.556.7129.2
Net direct investment13.064.885.399.6120.4147.2127.5118.6
Net portfolio investment4.464.0104.440.780.269.935.341.9
Other net investment-3.815.4-
Net official flows26.217.4-
Change in reserves4-14.4-79.6-75.4-121.0-106.2-37.7-31.7-67.3
Developing countries
Net private capital flows317.8129.4133.8148.2190.4139.065.8116.1
Net direct investment12.257.976.586.5108.5126.5108.297.8
Net portfolio investment4.951.185.722.252.755.532.038.4
Other net investment0.620.4-28.439.529.3-43.0-74.4-20.1
Net official flows27.216.810.332.13.2-3.327.63.3
Change in reserves45.1-54.8-42.3-67.1-95.2-57.83.6-37.0
Net private capital flows32.
Net direct investment1.
Net portfolio investment-0.8-
Other net investment2.
Net official flows6.
Change in reserves40.1-2.3-5.0-1.8-7.3-12.8-0.9-1.6
Net private capital flows313.155.864.791.899.028.8-44.311.0
Net direct investment4.532.944.451.
Net portfolio investment1.56.711.310.010.211.6-12.22.6
Other net investment7.
Net official flows7.
Change in reserves4-2.1-28.9-39.7-29.1-48.1-19.24.8-37.9
Middle East and Europe
Net private capital flows32.323.
Net direct investment1.
Net portfolio investment5.112.313.
Other net investment-3.97.9-3.2-6.4-
Net official flows4.8-0.4-1.0-1.1-0.6-0.6-1.0-1.7
Change in reserves46.6-5.5-2.7-10.9-11.6-10.8-8.4-5.7
Western Hemisphere
Net private capital flows3-0.246.446.538.281.887.575.266.6
Net direct investment5.319.
Net portfolio investment-0.932.360.91.740.038.027.820.7
Other net investment-4.6-5.1-39.310.32.6-4.6-1.61.6
Net official flows8.21.2-3.820.5-13.5-8.0-0.1-3.9
Change in reserves40.5-18.15.0-25.3-28.2-
Countries in transition
Net private capital flows3-1.012.518.843.216.222.720.031.1
Net direct investment-
Net portfolio investment10.420.518.824.320.714.311.8
Other net investment-0.8-4.2-7.111.0-21.6-16.2-13.01.5
Net official flows0.21.1-12.1-8.49.915.01.7
Change in reserves4-3.6-7.6-6.9-36.2-0.1-6.10.8-8.7
Newly industrialized Asian economies5
Net private capital flows3-
Net direct investment0.90.53.4-0.2-
Net portfolio investment-0.52.4-1.8-0.33.3-6.3-11.0-8.3
Other net investment-3.7-
Net official flows-1.1-0.5-0.3-0.615.810.8-5.7
Change in reserves4-15.9-17.2-26.2-17.7-10.926.1-36.1-21.6

Net capital flows comprise net direct investment, net portfolio investment, and other long- and short-term net investment flows, including official and private borrowing.

Annual averages.

Because of data limitations, “other net investment” may include some official flows.

A minus sign indicates an increase.

Hong Kong SAR, Korea, Singapore, and Taiwan Province of China.

Net capital flows comprise net direct investment, net portfolio investment, and other long- and short-term net investment flows, including official and private borrowing.

Annual averages.

Because of data limitations, “other net investment” may include some official flows.

A minus sign indicates an increase.

Hong Kong SAR, Korea, Singapore, and Taiwan Province of China.

Data on gross private financial flows to emerging markets indicate that gross (or new) financing peaked in the second and third quarters of 1997, and that in the first half of 1998 it was running at about half of precrisis levels (Table 2.10). Asia accounts for most of the decline in gross flows since mid-1997, but flows to other regions have also been adversely affected. In August, gross financing virtually dried up, reflecting the turbulence in Russia and other emerging markets.

Table 2.10.Gross Private Financing to Emerging Market Economies(Billions of U.S. dollars)
Middle East and Africa15.531.02.85.810.611.73.311.
Western Hemisphere63.190.116.729.430.114.121.724.
Bond issues101.9127.927.743.045.012.425.330.49.713.60.4
Western Hemisphere47.254.211.918.7203.814.815.65.35.4
Other regions11.628.23.18.410.
Other fixed income9.410.
Western Hemisphere
Other regions0.20.1
Loan commitments90.7123.623.332.729.937.511.
Western Hemisphere12.330.
Other regions22.
Equity issues16.424.
Western Hemisphere3.
Other regions3.
Source: Capital Data Loanware and Bondware, Ltd.
Source: Capital Data Loanware and Bondware, Ltd.

Net private capital flows in 1998 as a whole are projected to be a further $67 billion lower than in 1997, at around $57 billion—some $65 billion lower than projected in the May 1998 World Economic Outlook, and about one-fourth the net inflow recorded in 1996. Some gradual recovery in private flows to emerging market economies is assumed in late 1998 as confidence picks up, and this is reflected in a recovery in private flows projected for 1999, although the timing and degree of recovery remain highly uncertain.

The financial crisis in Russia, the devaluation of the ruble, and the unilateral debt restructuring announced by the Russian authorities have substantially worsened prospects and increased risks for net private capital inflows into emerging market economies in Latin America and central and eastern Europe. While these countries’ access to private flows was undiminished in the wake of the Asian crisis, some are likely to be adversely affected by these more recent developments. One early indication of this is demonstrated by Eurobond spreads that have widened to levels last seen in January 1995, close to the peak of the Mexican crisis (Figure 2.15). The baseline projections assume that these spreads will decline and access to private financing will be maintained, although at lower levels than in 1997 and early 1998. Thus net private inflows into the emerging market economies of the Western Hemisphere are projected to decline in 1998 and 1999, to a level next year that would be about $20 billion below the 1997 peak. Net private inflows into countries in transition are also projected to decline in 1998, mainly reflecting Russia. If interest rate spreads do not decline, and if private capital flows do not recover as assumed, domestic demand and economic activity in the emerging market economies will, of course, weaken, by more than projected in the baseline as imports are compressed and current account balances turn toward surplus. Higher borrowing costs could also put strains on fiscal balances.

Figure 2.15.Emerging Markets: Bond Spreads1

(Percent a year)

In all regions, emerging market bond spreads (relative to U.S. treasury securities) increased sharply following the Russian crisis in August 1998.

Sources: Bloomberg Financial Markets, LP; Reuters; and IMF staff estimates.

1Secondary market yield spreads on U.S. dollar-denominated bonds over comparable U.S. treasury securities.

2The Russian Eurobond spread reached 63.8 percent in the first week of September 1998.

The sharp declines in private capital flows to emerging market economies that began with the eruption of the Asian crisis have already required substantial adjustments of external positions, especially in Asia, even though the adjustments are being cushioned in several cases by drawdowns of reserves and by official borrowing. Among the crisis countries, Indonesia, Korea, and Thailand have all been supported by substantial official flows catalyzed by IMF arrangements. Looking ahead, similar adjustment patterns can be expected in Russia and to a lesser extent in several countries of the Western Hemisphere, particularly if net private inflows fall below those assumed in the baseline projections.

For the five countries most severely affected by the Asian crisis—Indonesia, Korea, Malaysia, the Philippines, and Thailand—current account balances are projected to amount to a combined surplus of $57 billion in 1998, compared with combined deficits of $24 billion in 1997 and $54 billion in 1996 (Table 2.11). The swings between 1997 and 1998 alone are projected to range from about 4 percentage points of GDP in Indonesia and the Philippines to 13–15 percentage points of GDP in Korea and Thailand (Table 2.12). These large shifts in external balances—and they could turn out to be even larger than projected in the baseline—are being generated partly by the large currency depreciations that have occurred and partly by the compressions of domestic demand that have resulted from the tightening of financial conditions, including through the implementation of adjustment measures. As discussed earlier, the adjustments in the Asian crisis countries thus far have occurred mainly through a sharp decline in imports, but the large improvements in competitiveness may be expected increasingly to boost exports, especially as financial conditions ease. With regard to currency movements, it is notable that in the 14 months to September 1998 the real effective exchange rates of Korea, Malaysia, the Philippines, and Thailand declined by 18–28 percent, and that of Indonesia by 60 percent, with the general weakness of partner currencies in Asia and higher domestic inflation offsetting a relatively small fraction of their nominal depreciations vis-à-vis the U.S. dollar (Table 2.13).

Table 2.11.Overview of Current Account Projections(Billions of U.S. dollars)


from May 1998

Advanced economies51.234.369.439.6-18.4-11.0-55.5
Major industrial countries-0.9-21.46.0-67.4-120.9-15.5-61.1
United States-115.3-134.9-155.2-236.3-290.4-9.3-39.4
United Kingdom-5.8-2.97.3-18.7-21.1-1.8-6.8
Other advanced economies52.255.763.4107.0102.44.65.6
Taiwan Province of China5.511.
Hong Kong SAR-5.5-1.6-
Industrial countries50.338.764.0-19.5-66.4-16.7-65.0
European Union53.990.8123.396.692.8-12.1-29.7
Euro area54.787.3111.6110.9108.3-9.6-21.5
Newly industrialized Asian economies5.81.08.961.750.75.89.5
Developing countries-95.3-71.4-61.8-78.3-63.212.629.3
Middle East and Europe-
Western Hemisphere-35.9-37.8-64.9-80.5-71.02.815.8
Countries in transition-1.6-18.0-25.0-30.8-25.16.917.2
Central and eastern Europe-5.9-16.6-20.3-21.7-
Excluding Belarus and Ukraine-4.1-14.9-18.2-19.7-
Transcaucasus and central Asia-1.5-4.0-4.1-5.0-
In percent of total world current account transactions-0.4-0.4-0.1-0.5-0.70.1-0.1
In percent of world GDP-0.2-0.2-0.1-0.2-0.4

Reflects errors, omissions, and asymmetries in balance of payments statistics on current account, as well as the exclusion of data for international organizations and a limited number of countries.

Reflects errors, omissions, and asymmetries in balance of payments statistics on current account, as well as the exclusion of data for international organizations and a limited number of countries.

Table 2.12.Selected Economies: Current Account Positions(Percent of GDP)
Advanced economies
United States-1.6-1.8-1.9-2.8-3.3
United Kingdom-0.5-0.20.6-1.4-1.5
Hong Kong SAR1-3.9-1.1-3.2-0.01.2
New Zealand-3.1-3.9-7.7-6.5-6.0
Taiwan Province of China2.
European Union0.
Developing countries
Côte d’Ivoire-6.0-4.8-4.5-4.1-3.2
Saudi Arabia-
South Africa-2.0-1.3-1.5-1.1-1.1
Countries in transition
Czech Republic-2.7-7.6-6.1-3.0-3.0
Slovak Republic2.0-11.0-11.0-10.0-6.0

Data include only goods and nonfactor services.

Based on data for the current balance, including a surplus on unrecorded trade transactions, as estimated by IMF staff.

Data include only goods and nonfactor services.

Based on data for the current balance, including a surplus on unrecorded trade transactions, as estimated by IMF staff.

Table 2.13.Recent Exchange Rate Movements on a Bilateral and Multilateral Basis1(Percent)

Exchange Rate
Nominal Effective

Exchange Rate
Real Effective

Exchange Rate

U.S. dollar

Japanese yen




United States17.610.916.510.512.7
United Kingdom1.919.
New Zealand-26.3-13.3-20.5-13.5-20.2-14.7
Hong Kong SAR17.58.214.812.616.1
Taiwan Province of China-19.4-5.2-12.9-8.2-14.9-12.4
South Africa-27.0-14.1-24.4-21.0-21.1-19.0
Sources: IMF, Direction of Trade Statistics (DOTS) and Information Notice System (INS) databases; WEFA, Inc; and IMF staff calculations.

Change from June 1997 to September 1998; positive number means appreciation. September 1998 was calculated based on the average exchange rates up to and including September 21.

Partner country weights capture both bilateral and third-country effects based on data for 1988–90.

Partner country weights capture only bilateral trade, based on data for 1994–96.

Sources: IMF, Direction of Trade Statistics (DOTS) and Information Notice System (INS) databases; WEFA, Inc; and IMF staff calculations.

Change from June 1997 to September 1998; positive number means appreciation. September 1998 was calculated based on the average exchange rates up to and including September 21.

Partner country weights capture both bilateral and third-country effects based on data for 1988–90.

Partner country weights capture only bilateral trade, based on data for 1994–96.

Other emerging market countries experiencing weaker capital inflows are also projected to improve their current account positions in 1999, in part through adjustment measures. And adjustments in current account balances could be larger than projected if interest spreads remain high and net capital inflows are lower than in the baseline.

The improvements in current account positions occurring in countries where financing has become more constrained are contributing to corresponding deteriorations in the current account balances of other countries—both emerging market and advanced economies—which are experiencing weaker export demand, deteriorations in competitiveness, and declines in the prices of commodity exports. Among the developing country regions, the largest projected shift into deficit in 1998 is for the Middle East and Europe—a shift of about $24 billion, $10 billion larger than the shift projected in the May 1998 World Economic Outlook—reflecting the deterioration in the current account balances of the oil-producing countries. This deterioration is not expected to be significantly reversed in 1999. In Saudi Arabia, for example, after virtual balance in 1996 and 1997 a deficit equivalent to 8¼ percent of GDP is projected for 1998. The weakness of commodity prices is also a factor tending to increase Africa’s deficit.

In the industrial countries as a group, current account balances are projected to deteriorate by more than $80 billion in 1998, and by a further $47 billion in 1999. The main component of the shift into deficit in 1998 is an $81 billion widening projected for the current account deficit of the United States; other projected changes are mutually offsetting. The current account surplus of the EU is projected to narrow by $27 billion, while the current account surplus of Japan is projected to increase by $37 billion. Australia’s current account deficit is projected to widen to almost $20 billion or 5½ percent of GDP this year, reflecting regional developments.

The projected increase in the U.S. current account deficit, to about 2¾ percent of GDP in 1998 and 3¼ percent in 1999 from its levels of 1½–2 percent over the past several years, reflects the downturn in Asia and expected adjustments in Latin America together with the continued buoyancy of domestic demand and the strength of the dollar. The decline in the EU’s surplus is accounted for mainly by a widening of the United Kingdom’s deficit to 1½ percent of GDP in 1998, reflecting the appreciation of sterling and the strength of domestic demand. Elsewhere in Europe, lower import prices and improvements in competitiveness in recent years are likely to offset the decline in exports to Asia, leaving the current account balances of the major continental countries broadly unchanged. Norway’s current account surplus, however, is expected to decline sharply to below 1½ percent of GDP this year, from more than 5 percent in 1997, owing to lower oil export revenues.

In Japan, the substantial negative impact of adjustments in neighboring economies is expected to be outweighed by the weakness of domestic demand, lower commodity prices, and the yen’s depreciation against other industrial country currencies, so that the current account surplus this year is projected to widen to 3½ percent of GDP, the largest surplus since 1987 and ½ of 1 percent larger than projected in May.

The projected changes in current account balances imply a widening of the global current account discrepancy to $70 billion in 1998 and $107 billion in 1999.9 Even though the discrepancy in 1997, at $17 billion, was exceptionally small by the standards of preceding years, the projected increases in the discrepancy seem unlikely to be realized once actual data become available. This inconsistency or tension in the current projections is greater than is typically the case, perhaps reflecting the size of the adjustments in trade flows now under way and the paucity of consistent trade and financial flow data available for 1998. It is likely to stem from a combination of two sources of possible forecast error: first, underestimation of the projected adjustments in the external balances of emerging market countries experiencing reduced capital inflows—that is, the current account surpluses of these countries may turn out to be greater than now projected; and second, overestimation of the shifts into deficit of other economies.10

The consequences for the global outlook are difficult to determine but point in the direction of downside risk. In particular, data available for the first half of 1998 show a potentially larger trade surplus in Asia than is portrayed in the projections. On this basis, one plausible scenario would incorporate more trade adjustment in the crisis economies than in the baseline projections. In this case, import compression and the slowdown in economic activity in these countries may still be underestimated in the current projections. Similarly, the rise in the global discrepancy could point to unresolved tensions in the more recent evolution of the crisis. For example, the projections for the United States may allow for a larger decline in exports than is consistent with the import projections for its trading partners in Latin America. Here, the tension would be resolved by greater-than-assumed adjustment in Latin America.

The large shifts in external positions stemming from the crisis in emerging markets have been both inevitable and, in the circumstances, to some extent also desirable. They have not only helped to cushion the contraction in output in the crisis countries but also to reduce the risk of overheating in countries that are at a mature stage of the business cycle. However, these adjustments will eventually need to be reversed through a process that is likely to involve shifts in relative cyclical positions, changes in the pattern of the global flows of funds, and exchange rate changes. A benign scenario for this process would involve a gradual return of confidence in emerging markets, a decline in interest rate spreads to more normal levels, a pickup in domestic demand in the crisis countries, and a progressive reduction of the large surpluses emerging in these countries in 1998–99. Recovery in Japan, led by resurgent domestic demand and a strengthening of the yen, would help to reverse the recent rise in Japan’s surplus. The U.S. and U.K. economies would experience a cyclical moderation of growth, while the euro area would continue to absorb economic slack, with the dollar and sterling adjusting vis-à-vis the euro in accordance with the likely shifts in relative cyclical positions and in the relative attractiveness of financial assets. Finally, the resumption of stronger growth in Asia would help the recovery of commodity prices and strengthen external positions of commodity exporters. In this scenario, the adjustment of external positions would contribute to the forces of recovery from the Asian crisis.

However, the adjustment process might proceed less smoothly if the return of confidence in emerging markets were significantly delayed and if financial markets eventually were to question the sustainability of the large current account deficit of the United States, which appears to be the main counterpart of the sharp shifts into surplus in the crisis countries. In this case, the current account imbalances would initially widen further, with a risk that subsequent market reactions would be more abrupt and with potentially adverse effects on the exchange rates of the major currencies, inflation, interest rates, and stock markets. Such a scenario might well be associated with significantly weaker global economic growth. The adjustment process could also be severely impeded if countries where current account deficits widen engage in defensive reactions, in the form of protective trade measures or competitive currency depreciations, with their damaging repercussions for global trade and growth.

The wider crisis scenario could also be transmitted to partner countries through the banking system. The Russian crisis, coming on top of the Asian crisis, has adversely affected the balance sheets of banks in the industrial countries. German banks have outstanding loans of about $30 billion to Russia, 60 percent of which is government guaranteed, with half of the remainder being provisioned for. Other European banking systems combined have smaller exposures than German banks, some of them are state-guaranteed, and provisions have been increasing. U.S. banks have about $7 billion in exposure to Russia, which has already resulted in some significant provisions and losses. U.K. and Japanese banks have relatively small exposures, at less than $1 billion each, although one U.K. bank has reported substantial losses. Moody’s has put the ratings of several European banks on review and has placed a negative outlook on others, although it does not view events in Russia as a major systemic risk to banks in the industrial countries. Nonbank financial institutions have been adversely affected as well, with a number of hedge funds and securities firms reporting substantial losses. A handful of hedge funds have been unable to meet margin calls or have sought bankruptcy protection.

Foreign Exchange and Financial Market Developments in the Industrial Countries

The deepening crisis in emerging financial markets and the deteriorating economic situation in Japan have had sizable spillover effects on financial markets in the industrial countries. Markets in the United States and Europe initially benefited, as bond yields declined, equity markets reached new highs, and the U.S. dollar and core European currencies strengthened on a multilateral basis. Downward pressures on financial markets were confined mainly to Japan and major commodity-exporting countries such as Canada and Australia, which were perceived as most vulnerable to the negative effects of a prolonged slowdown in Asia. However, the nature of the spillover effects changed in August and early September as the crisis took on a more global dimension. Government bond yields fell further—most sharply in the United States, the United Kingdom, and the core ERM countries—amid a global flight to quality, which was reflected in widening international yield differentials on government bonds and widening yield spreads on corporate bonds. In addition, stock markets in the major industrial countries suffered sizable downward corrections, and the U.S. dollar weakened against the yen and the ERM currencies. Volatility increased sharply across all markets.

In foreign exchange markets, the multilateral value of the U.S. dollar reached its highest level since December 1986 in mid-July, buoyed by strong domestic demand growth in the United States, interest rate differentials favoring dollar-denominated assets, and safe-haven demand in the face of deteriorating sentiment toward emerging markets (Figure 2.16). However, the dollar fell sharply in late August and early September as the financial market crisis spread to Russia and Latin America, prompting a downward correction in U.S. equity prices and revised expectations regarding the near-term direction of U.S. monetary policy.

Figure 2.16.Major Industrial Countries: Effective Exchange Rates

(Logarithmic scale; 1990 = 100)

The yen continued to weaken up to August, while concerns about the effects of falling commodity prices led to a decline in the Canadian dollar.

1Defined in terms of relative normalized unit labor costs in manufacturing, as estimated by the IMF’s Competitiveness Indicators System, using 1989–91 trade weights.

2Constructed using 1989–91 trade weights.

The depreciation of the Japanese yen accelerated in May and early June as the extent of the economic contraction in the first quarter became apparent, and as concerns increased about prospects for the early resolution of problems in the banking system. An 8-year low of more than ¥145 to the dollar was reached in mid-June before a rebound sparked initially by coordinated intervention by the U.S. and Japanese authorities—the first U.S. intervention since 1995—and supported in early July by the announcement of new initiatives by the Japanese authorities in the banking sector. The yen subsequently fell back to a new low before rebounding in late August and early September. Associated with the weakness of the yen and the Japanese economy were renewed currency weakness in neighboring emerging market economies. Moreover, the deep recessions in the Asian emerging market economies and Japan contributed to the weakness of commodity prices, which was a key factor in the decline of the Australian and New Zealand dollars against the U.S. dollar and the major European currencies. Lower commodity prices also contributed to downward pressure on the Canadian dollar, which fell to record lows against the U.S. dollar in August, prompting the Bank of Canada to raise official interest rates by a full percentage point.

In Europe, movements in the pound sterling, which have been in a narrow range vis-à-vis the U.S. dollar, have been affected mainly by changing expectations regarding U.K. monetary policy. The pound rebounded temporarily in June following the release of data showing inflation above its target and wages accelerating, and after the Bank of England unexpectedly raised official rates by a further 25 basis points. However, sterling subsequently weakened again as signs emerged that the economy was slowing and as wage and price pressures began to moderate, suggesting that short-term interest rates had peaked. The prospective euro-area currencies have generally strengthened on an effective basis, particularly since late August as the dollar has weakened. Most currencies have remained near their central ERM parities, the exceptions being the Irish pound, which has remained moderately appreciated following its revaluation by 3 percent in March, and the Greek drachma (not an initial participant in the third stage of EMU), which entered the ERM in March at a central rate that entailed a 12.3 percent devaluation against the European currency unit (ECU). The drachma initially strengthened to about 9 percent above its ERM central parity before giving up part of those gains in August as the Russian crisis deepened.

Government bond yields rose slightly in April and early May in most industrial countries except Japan amid concerns about the inflation risks associated with continued above-potential growth in the United States. However, yields declined steadily from mid-May as the deteriorating economic situation in Asia and associated weakness in commodity prices eased inflation concerns and as turmoil in emerging markets, including Russia, triggered a flight to quality. This trend accelerated in late August, particularly in the United States, the United Kingdom, and core ERM countries, as the emerging market crisis deepened. By early September, nominal bond yields had fallen by 150 basis points or more from a year earlier, in some cases reaching postwar lows, though the declines from earlier lows were somewhat less pronounced in real terms. In Japan, bond yields reached unprecedented levels of well below 1 percent as economic and financial sector conditions continued to deteriorate and inflation dropped to near zero (Figure 2.17). Elsewhere, government bond yield differentials widened somewhat in the future euro area, and bond yields also rose for a period in Canada, Australia, and New Zealand, reflecting downward pressures on the currencies of these countries. Corporate bond yield spreads also widened sharply in North America and Europe.

Figure 2.17.Major Industrial Countries: Nominal Interest Rates

(Percent a year)

Long-term interest rates in the industrial countries have declined further during 1998. Short-term interest rate convergence in the euro area has continued.

Sources: WEFA, Inc.; and Bloomberg Financial Markets, LP.

1Yields on goverment bonds with residual maturities of ten years or nearest.

2Three-month maturities.

Since mid-April, monetary authorities in most industrial countries have held short-term interest rates steady as the deflationary effects of the Asian crisis and its repercussions have helped to keep inflationary pressures well-contained, even in countries such as the United States where resource utilization has been high. Among the major commodity-exporting countries, strong downward pressures on exchange rates prompted official interest rate hikes amounting to a cumulative 375 basis points in Norway between late June and August, and a 100 basis point hike in Canada in August. Exchange rate pressures also prompted rises in short-term market rates for a period in New Zealand and, to a lesser extent, in Australia, while official rates were raised on two occasions in Denmark, most recently by 100 basis points in mid-September. Earlier, in June, inflation concerns had prompted a 25 basis point hike in the United Kingdom. By contrast, in Sweden, official interest rates were lowered by 25 basis points in early June, while in Greece the central bank cut its Lombard rate by 300 basis points at the end of July. In Japan, the central bank reduced its target for the overnight call rate from ½ to ¼ of 1 percent in early September amid continuing concerns about deflationary pressures and the weaknesses in the banking sector.

Within the future euro area, the scope for adjusting monetary policy in response to differences in cyclical conditions has been constrained by the need for short-term interest rates to converge by the January 1, 1999 start date of EMU. This process will entail significant rate reductions in some countries, including Ireland and—to a lesser extent—Portugal, where domestic demand growth is already strong, raising additional concerns about overheating (see Chapter V). Moves toward convergence of short-term rates since mid-April have been fairly limited, with official rates being cut by 50 basis points in Italy and by smaller amounts in Portugal and Spain.

After further strong gains in early 1998, stock market prices in many industrial countries generally consolidated or rose to new highs in mid-July before sizable downward corrections prompted by the prospect of a marked slowdown in world economic growth (Figure 2.18). Although the global interest rate environment remained highly favorable for equities, concerns that current market valuations may be difficult to sustain in the face of slower growth in corporate earnings, combined with a general increase in risk aversion, prompted a reversal of the price gains recorded earlier in the year in the United States and the United Kingdom, and a partial reversal of the larger earlier gains in the major continental European markets. In Japan, equity prices remained under downward pressure, falling to 12-year lows in late August and early September on concerns about the economy and the banking sector. Markets also weakened significantly in other countries with large direct and indirect exposures to the economic downturn in Asia, including Canada, Australia, and New Zealand.

Figure 2.18.Advanced Economies: Equity Prices

(U.S. dollar terms; logarithmic scale; January 1990 = 100)

Equity markets in the United States and the major European countries performed strongly up to mid-July but then dropped sharply.

Source: WEFA, Inc.

Inflation Concerns in Advanced Economies

What Is the Role of Transitory Factors in Influencing Inflation in the United States?

Inflation has remained low in the United States despite robust output growth and labor market conditions that previous experience suggests would result in inflationary pressures (Table 2.14). The U.S. unemployment rate fell to a 28-year low of 4.3 percent in April 1998, and since late 1994 it has been consistently below the level of around 6 percent at which labor shortages were previously found to result in wage and inflationary pressures.

Table 2.14.United States: Economic Indicators(Annual percent change unless otherwise noted)
United States
Real GDP2.73.93.5
Consumer price index2.92.31.6
GDP deflator2.51.91.2
Unemployment rate26.55.04.5
Producer price index1.7-0.1-5.6
Productivity, manufacturing3.64.03.4
Unit labor costs, manufacturing-0.51.2
Nominal effective exchange rate-
Import prices1.6-4.3-5.8
Oil prices2.0-5.4-31.1
Nonfuel commodity prices6.2-3.3-13.9

Projections, except for producer price inflation and the nominal effective exchange rate, where changes shown refer to data through August 1998.

Percent of labor force.

Projections, except for producer price inflation and the nominal effective exchange rate, where changes shown refer to data through August 1998.

Percent of labor force.

The success of the United States in maintaining low inflation close to full resource utilization can be attributed in part to factors that are likely to allow continued strong growth at a lower inflation rate than in the past. The Federal Reserve’s successful efforts to lower inflation appear to have reduced inflation expectations in the United States, and to have allowed lower interest rates and higher levels of investment and output than was the case without the added credibility. Moreover, there is evidence that the unemployment rate consistent with stable inflation—the nonaccelerating inflation rate of unemployment (NAIRU)—has fallen in recent years, from around 6 percent to between 5 and 5½ percent, though there is a large degree of uncertainty surrounding such estimates.11 The apparent fall in the NAIRU can be explained in part by demographic shifts, as the aging of the baby boom generation has resulted in a larger proportion of older workers who tend to change jobs less frequently, thus reducing the “frictional” unemployment present in the economy.12

For 1995 and 1996, the combination of low inflation and unemployment below 6 percent appears to be consistent with a decline in the NAIRU from 6 percent before 1995 to 5.5 percent (Figure 2.19). However, even assuming a NAIRU as low as 5 percent does not explain the fall in inflation in 1997; indeed, all else equal, inflation would have been expected to rise as unemployment stayed below 5 percent from the second quarter of 1997 on, which is below essentially all estimates of the NAIRU. This suggests that factors other than labor market developments account for the decline in inflation since the beginning of 1997.

Figure 2.19.United States: Actual and Predicted Inflation1


Low inflation in 1995–96 appears consistent with a decline in the NAIRU from about 6 percent to the 5–5.5 percent range, but this does not explain the even lower inflation in 1997 and early 1998.

1The forecasts of inflation are based on a model in which inflation is related to the current and past values of the unemployment gap (the difference between the actual rate of unemployment and the non-accelerating inflation rate of unemployment (NAIRU)); past rates of inflation; and past rates of increase of unit labor costs, import prices, oil prices, and nonfuel commodity prices. The model is estimated using quarterly data from 1975 to 1994, assuming a NAIRU of 6.0 percent, and then used to provide forecasts of inflation for values of the NAIRU of 6.0, 5.5, and 5.0 percent.

In fact, it would appear that the performance of inflation in 1997–98 results in part from a confluence of favorable developments, the influence of which could wane in the period ahead. For example, slow growth in Asia has curbed demand for U.S. exports and thus has helped to avert the need for the Federal Reserve to raise interest rates to stem overheating. (The crisis in Asia has thus effectively changed the composition of demand and output in the United States, with a slowdown in net exports in lieu of a retrenchment in domestic consumption and investment that would otherwise have taken place as a result of monetary tightening.)

Further, the exchange value of the U.S. dollar has risen sharply in the past three years, with its nominal effective value increasing by over 25 percent since April 1995. This has had both a direct effect of reducing prices of imported goods, and an indirect effect of putting competitive pressure on producers of tradables and thus limiting the scope for price increases. Weak demand in Asia has also contributed to declines in the prices of commodities, thereby reducing cost pressures, as reflected especially clearly in the producer price index, which has declined since the beginning of 1997. With firms facing competitive pressure both domestically and externally, these cost reductions have contributed to lower price inflation. Meanwhile, in the labor market, although the growth of labor earnings has picked up somewhat, it has remained subdued in relation to the experience of recent decades despite low rates of unemployment. The low-inflation environment has also been enhanced by slower growth of health care costs, reflecting the rapid expansion of health maintenance organizations that put an emphasis on cost containment. Falling computer prices have been another contributory factor.

The United States has also enjoyed relatively strong productivity gains in recent years that have largely offset wage increases and have even led to declines in unit labor costs in manufacturing in some years. This increased productivity growth may stem in part from the environment of stable, low inflation and also from the effects of corporate restructuring. There may also have been benefits from the large-scale investments in information technology. Finally, revisions to the calculation of the U.S. consumer price index will have the effect of reducing the upward bias in the measurement of inflation by about 0.4 percentage point over the next five years, so that measured inflation will tend to be somewhat lower on this account.

Some of the developments that have dampened inflation recently are of a temporary nature, implying that unless productivity growth remains strong there are notable risks to the inflation outlook as these transitory factors are reversed. The current value of the dollar appears to be significantly above its medium-term equilibrium level, and as growth picks up in Europe and Asia the currency may be expected to depreciate, putting upward pressure on import prices. Similarly, renewed growth in Asia could provide the impetus for a reversal of the decline in commodity prices. And despite the absence of serious cost pressures in the United States so far, it remains an open question whether labor costs and the prices of other factors of production will remain subdued if resource utilization remains elevated.

Does the Unemployment Rate Provide a Good Indicator of the State of the Labor Market, and How Much Slack Is There Currently in the Labor Markets of Advanced Economies?

After falling to 4.3 percent in April and May 1998, the unemployment rate in the United States stood at 4.5 percent in June 1998, close to the lowest level reached in three decades. In contrast, unemployment has trended sharply upward over the past decade in many of the continental EU countries, reaching postwar highs in 1997 in countries such as France, Germany, and Italy. While unemployment in Germany and France has declined somewhat in the first five months of this year, it is still more than 2½ times the current unemployment rate in the United States (Figure 2.20). There have been a few notable exceptions to the trend of rising unemployment in the EU—for example, the unemployment rate dipped sharply in the United Kingdom, to about 4.8 percent, in May 1998. Unemployment in the Netherlands has also been falling in recent years.

Figure 2.20.Selected Advanced Economies: Unemployment1

(Percent of labor force)

Unemployment has declined in the United States but increased sharply in a number of European countries over the past two decades.

1Shaded areas indicate IMF staff projections.

While differences in measured rates of unemployment do provide useful information about the relative strengths of labor markets in different countries, they do not always provide the best indicator of the capacity of an economy to create jobs and employ people gainfully. In part this is because of the differences in how unemployment is measured among countries—labor force surveys being used in some cases, and numbers of people claiming benefits in others. Table 2.15 shows how assessments of labor market performance can change significantly depending on the measure of unemployment used. For instance, the unemployment rate in the United Kingdom in 1997, based on the standardized measure of the Organization for Economic Cooperation and Development (OECD), was almost 1½ percentage points higher than the national definition based on benefit claimants. In the case of the Netherlands, by contrast, the OECD’s standardized measure paints a more impressive picture of labor market performance than the national definition, although allowance needs to be made for substantial withdrawal from the labor force.

Table 2.15.Selected Advanced Economies: Unemployment Rates, 1997(Percent of labor force)


United States4.94.95.0
United Kingdom5.57.06.7

Nonaccelerating inflation rate of unemployment (IMF staff estimates).

Nonaccelerating inflation rate of unemployment (IMF staff estimates).

Changes in the measured rate of unemployment may at times be mainly a reflection of changes taking place in participation rates, as people move in and out of the labor force, rather than an indication of changes in employment.13 Although structural factors such as increased participation of women in the labor force govern long-term trends in participation rates, there is also a tendency for participation rates to be procyclical, reflecting cyclical variation movements in the probability of successful job search (Figure 2.21). Consequently, movements in the measured rate of unemployment usually understate both the extent of the improvement in labor market performance during recoveries, as well as the deterioration during recessions.

Figure 2.21.Selected Advanced Economies: Labor Force Participation

(Percent of total labor force)

Changes in participation rates reflect both long-term trends and procyclical movements.

Source: OECD, Analytical Data Bank.

Focusing on the employment, rather than unemployment, performance of different countries is one way of avoiding some of the pitfalls involved in using unemployment data to make judgments about labor market performance. But here there also are problems. Thus, potential employment growth depends on the growth of the working-age population, and one of the explanations for the relatively large growth of employment in the United States is its relatively rapid population growth by the standards of other industrial countries. To assess the performance of labor markets, one should therefore examine employment growth relative to the growth of the working population. In addition, employment data may conceal factors such as whether the jobs that have been created are temporary, part-time, or relatively poorly paying. Movements in employment may nevertheless offer a valuable macroeconomic indicator of labor market performance beyond what can be seen from unemployment data. Figures 2.21 and 2.22 show that the United States’ impressive record in bringing unemployment down in the 1990s is matched by its performance in employment creation, which has permitted a substantial increase in participation rates. Aggregate employment in the United States has increased by more than 30 percent since 1980. This contrasts with the experience in continental Europe. There has, for instance, been virtually no increase in employment in Germany since 1980 as unemployment has increased steeply, while in Italy employment has actually fallen during this period. The United Kingdom’s recent success in reducing unemployment is put into perspective not only by the standardized measures referred to earlier, but also by its relatively sluggish employment creation. In this context, the job creation records of Ireland and the Netherlands appear particularly impressive.

Figure 2.22.Selected Advanced Economies: Employment1

(1980 = 100)

The performance of France, Germany, and Italy in employment creation since 1980 has been disappointing.

1Shaded areas indicate IMF staff projections.

What accounts for the differences in the employment performances of the advanced economies? The successful employment growth record of the United States is related to the high degree of flexibility of its labor market—both real and relative wages respond relatively quickly to market forces in the United States, and there is also a much smaller wedge between the gross wage paid by employers and the net wage received by employees than is the case in the other industrial countries. Factors contributing to the high degree of flexibility in U.S. labor markets include low minimum wages, low rates of unionization, and relatively low levels of unemployment and social welfare benefits. In contrast, the poor employment creation record of countries such as France, Germany, and Italy has been attributed to a variety of labor market regulations and norms that have had the effect in these countries of discouraging employers from hiring and workers from actively seeking employment, although more recently some modest reforms have been introduced in these countries. Although the institutional structures of labor markets in both the United Kingdom and the Netherlands were closer to those of the other EU countries up until the early 1980s, both these countries implemented a series of reforms in the mid-1980s that succeeded in removing many of the rigidities that had earlier characterized their labor markets. In the United Kingdom, the reforms involved the liberalization of hiring and firing laws, abolition of minimum wages for all categories of workers except agricultural workers, and the reduction of unemployment benefits relative to earnings. (A national minimum wage, due to be introduced in 1999, is likely to affect employment negatively.) In the Netherlands, there were modest reductions in unemployment benefits, although with more stringent eligibility requirements for receiving them. Minimum wages were reduced substantially, and nonwage labor costs, particularly for unskilled workers, were brought down significantly.14

Although the unemployment rate provides some indication of the degree of excess demand or supply in the labor market, it is a very imprecise measure. Thus, a relatively high unemployment rate need not imply an effective excess supply of workers at the going wage. Workers may be unwilling to accept jobs at the going wage even at high rates of unemployment if there are generous unemployment benefits or other social welfare programs. Also, since high unemployment rates are generally associated with high proportions of long-term unemployed, leading in turn to the deskilling and demotivation of workers, employers may be faced with shortages of suitable workers even in conditions characterized by high unemployment.

In this regard, unemployment relative to the natural rate of unemployment or the NAIRU is a more appropriate indicator of potential inflationary pressures. The NAIRU is determined in part by institutional features of the labor market, such as wage-setting patterns; the generosity of the unemployment benefits system; and the ease with which training programs are available to workers. But the NAIRU may change over time, reflecting not only institutional changes in the labor market but also unemployment itself. Thus, a period of sustained low unemployment can lower the NAIRU, as previously demotivated workers reenter the labor market. Table 2.15 provides the IMF staff’s estimates of the NAIRU for a group of selected advanced economies. Even though there are wide margins of uncertainty about these estimates, it is apparent that the NAIRU is higher in countries where labor market rigidities are more pronounced.

When actual unemployment is below the NAIRU, it is a sign that inflationary pressures may be building up in the economy. Unemployment in both the United States and the United Kingdom is currently below estimates of their respective NAIRUs, indicating a risk of incipient inflationary pressures that may be building in these economies. In the major continental European countries, in contrast, unemployment is above the estimated NAIRU, suggesting that there is no immediate inflationary danger from the rebound of activity in these countries.

Implications of Asset Price Inflation for Monetary Policy in the Advanced Economies

In advanced economy equity markets, the sharp price declines experienced in July and August have largely reversed the gains accrued in 1998. Even with these recent declines, however, asset prices, particularly the prices of equities, have risen substantially in many advanced economies over the past several years. This has raised concerns on the part of policymakers—although inflation, as measured by consumer price indices or other broad measures of the general level of output prices, has remained low in most cases. Concerns arise because increases in asset prices have often in the past seemed to be a leading indicator of increasing demand and inflation. Part of the explanation is that rising stock market wealth leads to increased consumption, while associated declines in finance costs lead to higher investment. Especially in economies such as the United States and United Kingdom with high degrees of resource utilization, the rise of stock market values over the past years has added to the perceived risks of overheating. These risks have been offset to some extent by the summer corrections, since this sharp reminder that markets fluctuate both up and down may well give pause to the translation of increased stock market wealth into consumption.

In the United States, before the recent price declines, the Standard & Poor 500 stock price index through July had risen by more than 20 percent since the beginning of the year, following sharp increases in 1995–97 (Table 2.16). Other equity indices in the United States have shown similar experiences. In the United Kingdom, the Financial Times Stock Exchange 100 index rose by over 17 percent in 1998 before the July reversal, following a 26 percent rise in 1997. Equity prices in continental Europe have shown even larger increases since late 1996, after relative sluggishness earlier and continue to show sizable gains in 1998. Prices of real estate have not risen by nearly as much as equity prices, although there have been regional increases in the United States and the United Kingdom, and they have declined since early 1997 in Germany as well as Japan.

Table 2.16.Selected Countries: Changes in Asset Prices(In percent)
Equity PricesHousing Prices
Country1998 to

September 24


United States8.629.514.54.33.3
United Kingdom0.626.111.48.4-2.3
Sources: Share prices from Bloomberg Financial Markets, LP. Indices are: United States, Standard & Poor 500; United Kingdom, FTSE 100; Japan, Nikkei 225; Germany, Frankfurt Commerzbank; France, Paris CAC 40; Italy, Milan Banca Commerciale; Canada, Toronto Stock Exchange Composite; Netherlands, CBS General Index; Spain, Madrid stock index. Housing prices for United States and United Kingdom are from Bloomberg Financial Markets, LP.; prices for Japan, Netherlands, and Spain from IMF staff country data; Germany and France from Bank for International Settlements (BIS).

Land price used for Japan instead of housing price.

Sources: Share prices from Bloomberg Financial Markets, LP. Indices are: United States, Standard & Poor 500; United Kingdom, FTSE 100; Japan, Nikkei 225; Germany, Frankfurt Commerzbank; France, Paris CAC 40; Italy, Milan Banca Commerciale; Canada, Toronto Stock Exchange Composite; Netherlands, CBS General Index; Spain, Madrid stock index. Housing prices for United States and United Kingdom are from Bloomberg Financial Markets, LP.; prices for Japan, Netherlands, and Spain from IMF staff country data; Germany and France from Bank for International Settlements (BIS).

Land price used for Japan instead of housing price.

The strong multiyear performance of equity markets could present a challenge for monetary policy, because of its possible implications for aggregate demand and inflation expectations. With regard to effects on consumption, using data through the mid-1980s for the United States, it has been estimated that in the long run consumption tends to increase by $0.03–$0.05 for every dollar increase in stock market wealth, compared with a $0.07–$0.09 increase for each dollar increase in nonequity wealth, with the relative volatility of stock values appearing to lead consumers to exercise greater caution in changing consumption than in response to capital gains from other sources.15 Assuming a wealth effect of $0.04 per dollar, this evidence suggests that the more than 25 percent increase in stock market capitalization in the United States since end-1995 will eventually translate into an $85 billion increase in consumption, equal to about 1 percent of 1997 GDP. At its peak in July, however, stock market wealth had increased by 50 percent since end-1995, with an implied doubling of the effect on consumption.

Although there are lags of up to three years involved in the materialization of wealth effects on spending, the gains have accumulated over more than four years and appear to have already contributed to increased consumption, as evidenced by the decline in the household saving rate in the United States from 3.4 percent in 1995 to 2.1 percent in 1997, with a further decline to 0.9 percent in the first half of 1998. There may have been an even larger effect on investment, with business sector fixed investment growing by 9–10 percent annually in 1995–97, as firms took advantage of lower capital costs to increase borrowing. These effects are similar to the experience of the 1980s, when the household saving rate fell from 9.4 percent in 1981, before the asset price boom, to 5.3 percent in 1988 just after the stock market reversal. Other countries, including the United Kingdom and Sweden, also experienced marked swings in consumption and investment in part associated with changes in asset prices in the 1980s.

To the extent that the wealth effect translates asset price gains, with lags, into added consumption, asset price inflation represents incipient demand and price pressures. Policymakers need to take such indicators into account in setting monetary conditions. Conversely, of course, monetary policy influences asset prices through interest rates and the cost of borrowing. Indeed, asset price reversals in the late 1980s in a number of advanced economies were closely associated with monetary tightenings. The link between monetary policy and asset price changes may well be enhanced during asset booms because borrowing on the basis of appreciating collateral and to finance asset accumulation increases consumer and firm financial leverage, and thus potentially heightens the sensitivity of demand to interest rate changes.

In the past, asset market reversals following sustained appreciations that were in retrospect seen as not being justified by underlying fundamentals may have had long-lasting effects on overall activity and the structure of the real economy. In Japan, Sweden, the United Kingdom, and the United States, for example, overinvestment and real estate speculation fueled by easier access to credit in the 1980s left firms in weak positions when faced with business cycle downturns, so that asset price reversals led to widespread balance sheet adjustments and financial fragility that prolonged recessions and weakened recoveries.16 In Japan, the effects of the asset price collapse are still being felt nearly a decade later, even though the asset price increases appear to have spilled over to a relatively small extent into inflation in goods prices.17 Because asset price developments may have a substantial impact on macroeconomic stability, they are an essential part of the overall array of indicators that monetary authorities must monitor and assess on an ongoing basis.

Such concerns do not imply, however, that monetary policy should primarily aim at stabilizing asset prices rather than inflation in goods and services. As recent events have made clear, asset prices are much too variable for this to be feasible, and the relationship between asset prices and monetary policy instruments is poorly understood. Just as high asset prices do not alone imply the need for a tightening of monetary policy, a decline in asset prices similarly does not by itself call for a loosening, especially if the declines represent a retreat from overvalued levels. This is particularly true in the United States and other advanced economies with continued strong growth and solid fundamentals.

Issues Relating to Growth and Inflation in Developing Countries and Countries in Transition

How Have Recent Commodity Price Movements Affected Growth and Inflation Prospects, Particularly in the Middle East and Africa?

A common feature of the economies of the Middle East and Africa is a heavy dependence of a number of countries on commodity exports for foreign exchange earnings.18 As a consequence, the adverse effects of recent primary commodity price declines seem likely to be potentially greater for the Middle East and Africa than for other regions.19

Oil-Exporting Countries in the Middle East and Africa

With direct links to the rest of the economy being relatively minor, and oil companies being fully or majority state-owned in most cases, the main direct impact of changes in oil prices is on the revenues accruing to the state through these companies. Many oil-exporting countries are likely to face difficult budgetary decisions over the near and medium term if the oil price path remains at levels much below those that prevailed in 1997. In general, those problems are likely to be more severe for the oil-exporting countries in the Middle East and Africa than elsewhere because they rely more heavily on oil revenues to finance government expenditure (Table 2.17).

Table 2.17.Major Oil-Exporting Developing Countries: Budgetary Impact of Oil Price Decline
1996-97 AveragesGovernment

oil revenue as

percent of GDP
Impact of oil

price decline on

fiscal receipts

(percent of GDP)
CountryGovernment oil

revenue as percent

of total revenue

fiscal balance

(percent of GDP)
Congo, Rep. of67-6.518.85.4
Equatorial Guinea49-
Middle East
Islamic Rep. of Iran58-2.310.81.6
Saudi Arabia77-3.419.37.9
United Arab Emirates72-9.816.45.5
Latin America
Trinidad and Tobago221.82.41.0
Sources: National authorities; and IMF staff estimates.

Nigeria recorded an overall deficit of -0.4 percent of GDP in 1997.

Combined public sector.

Sources: National authorities; and IMF staff estimates.

Nigeria recorded an overall deficit of -0.4 percent of GDP in 1997.

Combined public sector.

A key question facing governments heavily dependent on oil revenues is how to deal with current and intertemporal budget constraints in light of lower oil prices. How this is done affects both near-term non-oil output (and its future growth) and the prospects for inflation. For a number of countries that are particularly oil-rich (in per capita terms), notably in the Persian Gulf region, the need for near-term fiscal and current account adjustment may be lessened by their ability to attract net capital inflows that can generate budgetary resources. However, if oil prices do not rebound strongly, even these countries will need to undertake revenue-raising and expenditure-reducing measures to ensure medium-term external and fiscal sustainability. Other oil-exporting countries, particularly those in sub-Saharan Africa (for example, Cameroon and Nigeria), are more likely to be constrained by external credit requirements and will face a need to enact stronger adjustment measures in both the near term and the medium term.

Current projections of fiscal positions in the oil-exporting countries of the Middle East and Africa indicate a considerable weakening in 1998 and 1999 compared with the projections made early in 1997. In comparison with the earlier projections, there is a modest decline in projected GDP growth and a small increase in projected rates of inflation for this group of countries (Table 2.18).

Table 2.18.Middle East and African Countries: Growth, Inflation, and Fiscal Balances
Country GroupsNumber of



May 1997



GDP growth (percent a year)
Oil-exporting countries
Middle East123.93.71.8-1.9
Sub-Saharan excluding Nigeria64.46.03.3-2.7
Oil-importing countries
Middle East22.76.83.3-3.5
Africa—Group A1133.64.02.8-1.2
Excluding South Africa125.45.24.6-0.6
Africa—Group B1303.
Inflation (year-to-year percent change in consumer price index)
Oil-exporting countries
Middle East128.
Sub-Saharan excluding Nigeria646.421.620.4-1.2
Oil-importing countries
Middle East24.
Africa—Group A11314.29.18.8-0.3
Excluding South Africa1219.78.811.52.7
Africa—Group B1305.74.94.7-0.2
Central government fiscal balance (percent of GDP)
Oil-exporting countries
Middle East12-2.1-1.9-3.3-1.4
Sub-Saharan excluding Nigeria6-6.7-4.0-7.8-3.8

Group A African oil-importing countries rely for 10 percent or more of their export earnings on nonfuel primary commodities for which prices have fallen from 1997 to 1998 by more than 10 percent. Group B comprises all other oil-importing African countries.

Group A African oil-importing countries rely for 10 percent or more of their export earnings on nonfuel primary commodities for which prices have fallen from 1997 to 1998 by more than 10 percent. Group B comprises all other oil-importing African countries.

Oil-Importing Countries in the Middle East and Africa

For oil-importing countries in the two regions, the fall in oil prices will have beneficial effects on their current account positions, although deeper fiscal adjustment in the Persian Gulf region may produce some offset via reduced remittances and investment in some countries (for example, Jordan and Lebanon). Even in oil-importing countries, growth may suffer because of the impact of the price decline on neighboring oil exporters.

A number of oil-importing countries in Africa, furthermore, depend to a large degree on exports of certain nonfuel primary commodities whose prices have also fallen quite sharply over the past year. Thirteen of the 43 African oil-importing countries rely heavily for their foreign exchange earnings on nonfuel primary commodities that declined in price by 10 percent or more from 1997 to 1998 (Figure 2.23). For example, metal (including precious metal) prices have been subject to quite significant price declines over the past year. The price of copper by June 1998 was 27 percent lower than a year earlier, and this decline is having a serious impact on Zambia and to a lesser extent on the Democratic Republic of Congo.20 Aluminum prices have also declined quite sharply, by about 10 percent, and this will have a negative impact on Bahrain, which uses its own natural gas for refining imported bauxite. Gold prices declined quite sharply during 1997 and have fluctuated around $300 per ounce since. Gold exports are very important for such countries as South Africa and Ghana, while also contributing to the export performance of Mali and Zimbabwe.

Figure 2.23.Prices of Selected Commodities

Prices of many commodities have dropped significantly since the middle of 1997, adversely affecting developing countries in Africa and the Middle East.

Source: IMF staff estimates.

The loss of earnings from exports of the leading non-oil primary commodity exports in 1998 is expected to exceed the gain from lower prices for petroleum imports for 10 of the 13 African oil-importing countries experiencing adverse shocks from lower nonfuel commodity prices. The remaining 33 oil-importing countries in Africa will likely benefit from the lower cost of oil imports, with the gain equal to about 3 percent of their import bill. On balance, the current growth rate and inflation rate projections for 1998 for African oil-importing countries differ little from those made in 1997.

How Low and How Fast Should the Transition Economies Further Reduce Inflation?

Following notable successes in bringing down inflation to the moderate range, countries in transition have been facing difficulties in further reducing inflation to the low single-digit rates typical of the advanced economies in western Europe. In 1997, 12-month inflation rates rose in all the central and eastern European countries except Hungary, Latvia, Lithuania, and Poland, and in Hungary and Poland they remained well above 10 percent; inflation also remained in the double digits in Estonia (Figure 2.24). Inflation rates in some of the advanced reformers in central and eastern Europe and the Baltics now exceed rates observed in countries such as Azerbaijan, Kazakhstan, and the Kyrgyz Republic, where very significant progress in bringing down inflation was made in the past three years. Overall, in 1998, only four transition countries—Azerbaijan, Croatia, Latvia, and the former Yugoslav Republic of Macedonia—are expected to experience inflation at or below 5 percent. These developments raise questions about the factors underlying the recent lack of progress in lowering inflation further in central and eastern Europe and the appropriate goal, speed, and design of further disinflation policies.21

Figure 2.24.Selected Countries in Transition: Inflation

(Twelve-month percent change in the consumer price index)

Progress in reducing inflation has stalled in some of the central and eastern European economies; inflation has picked up again in Russia following the August crisis.

Sources: National authorities; and IMF staff estimates.

The recent inflation experience of the transition countries reflects a variety of factors. In Albania, the collapse of revenue in the wake of the civil unrest contributed to a sharp increase in inflation in 1997, but strong stabilization efforts over the past year have helped to reduce inflation markedly. In Bulgaria and Romania, overall macroeconomic instability resulted in sharp increases in inflation in 1997 also; although in Bulgaria inflation has since declined sharply in the context of a currency board arrangement, it has remained high in Romania. In the Czech Republic, large increases in regulated prices and wage pressures, especially in state-owned enterprises, were the principal factors behind a modest acceleration, while in Slovenia a currency depreciation and adjustments in administered prices contributed to a similar outcome. The persistence of inflation in Hungary, with year-on-year price increases still running at 13½ percent in August 1998, reflects a combination of inertia in inflation expectations, driving high nominal wage growth, and a relatively high rate of decline in the crawling peg of the forint. In Estonia, inflation remained in the double digits until July because of large adjustments in remaining administered prices and strongly rising domestic demand.

Although progress in further reducing inflation has stalled in several countries, the common goal of the reform-minded transition countries continues to be to bring down inflation rates to western European levels, and the countries that aspire to EU membership have explicitly affirmed their intention to do so (see Chapter V). This objective is appropriate and achievable over a reasonable span of time. As discussed in the October 1996 World Economic Outlook (Chapter VI), inflation may negatively affect both the level and rate of growth of output. The negative impact on growth is most clearly established for annual inflation rates in the double digits or higher. For inflation rates in the single digits but above advanced-economy levels, there is no strong evidence that inflation is detrimental to growth, but neither are there indications that it is beneficial.22 In this range, however, the negative effects on the level of output and the risk of slipping back into the double digits, where the effect on growth is unambiguously harmful, seem higher than at low single-digit rates.23 Achieving the low inflation goal is, however, complicated by the need in many transition economies to increase further many remaining administered prices. With the main contribution to inflation now coming from these price adjustments, policies can focus on underlying inflation so as to allow the needed relative price changes to take place without an undue upward push to the overall price level or to inflation expectations. In the Czech Republic, for instance, the National Bank has begun to formulate policies in terms of net inflation, which excludes changes in administered prices.

Given the objective of reducing inflation to levels observed in the advanced western European economies, authorities in the countries in transition still need to set a timetable for further disinflation. The Czech authorities have set a target of lowering “net” inflation from around 6½ percent in mid-1998 to between 3½ and 5½ percent by the end of 2000; Slovenia plans to reduce inflation from the current 9 percent to 3–5 percent by the end of 2001. Such timetables for further disinflation have to be assessed in light of the relative output costs of fast versus slow disinflation. Fast disinflation will be more costly the more monetary and fiscal policy lack credibility; the more rigid are nominal prices and wages; and the less flexible are relative prices and real wages.24

With regard to the credibility of policies, fiscal and financial sector fragilities that put the credibility of disinflation policies to the test form the main threat to successful fast disinflation in the countries in transition. As was illustrated by the reversal of the initial disinflation achievements in Bulgaria and Romania, financial sector problems that are not properly addressed may end up forcing central bank intervention and, accordingly, weaken the credibility of monetary policy. Similarly, in Russia a tight monetary policy in the presence of large fiscal deficits and lack of longer-term fiscal credibility resulted in high interest rates and, in the absence of measures to correct the underlying fiscal imbalances, an unsustainable accumulation of debt, which forced the authorities to change their exchange policy in August 1998, with immediate inflationary consequences. Second, with regard to nominal flexibility, price- and wage-setting arrangements in the transition countries are in general relatively flexible (except in Poland and Slovenia, where formal indexation has been widely applied), and facilitate fast disinflation. Finally, with regard to real rigidities, most transition countries in central and eastern Europe (the Slovak Republic is an important exception) have made significant progress in raising administered prices to cost-recovery levels, and further increases are expected to be gradual and without major impact on relative prices and real wages, making the adjustment process less of an impediment to fast disinflation. Current conditions in the transition countries, therefore, seem to be conducive to further reductions in inflation over a reasonable span of time, provided that disinflation efforts coincide with strong policies to foster fiscal sustainability and strengthen financial systems, with further progress in raising administered prices to cost-recovery levels, and with structural reform in the goods and labor markets aimed at increasing price and wage flexibility.

To achieve the low-inflation objective within the appropriate time frame, various monetary and exchange rate strategies can be effective. As discussed in the October 1997 World Economic Outlook (Chapter V), transition countries have adopted a range of monetary and exchange rate arrangements to bring down inflation. Under any arrangement, a key issue is how to respond to the capital inflows and real exchange rate appreciation that tend to accompany successful disinflations. The recent experience of the Czech Republic, Hungary, and Poland illustrates some of the trade-offs involved. In the Czech Republic and Poland, capital inflows were strong and real exchange rates appreciated significantly in the first half of 1998, as tight monetary policies aimed at reducing inflation were implemented and exchange rate arrangements are in place that allow for flexibility, following the koruna’s flotation in May 1997 and the widening of the margin within which the zloty is managed in February 1998. In Hungary, in contrast, a narrow crawling-peg framework and sterilization of capital inflows have kept the real exchange rate broadly constant since the beginning of the year and insulated the price system from the volatility of capital inflows, but have at the same time reduced the scope for monetary policy tightening and thereby contributed to an overshooting of the inflation targets. Because real exchange rates are a main determinant of competitiveness, their movements during the process of disinflation require careful monitoring, especially in countries such as Poland and the Baltics, which are recording sizable current account deficits. Although standard measures indicate that these countries have maintained their competitive positions and the deficits are mainly the counterpart of a surge in investment, constant vigilance and cautious fiscal policies are warranted.

It is unclear to what extent exports from Korea and other countries have been constrained by lack of financing. In the case of Korea, the dollar value of export letters of credit declined by 10 percent in the first quarter of 1998, reflecting strains in both the industrial and corporate sectors.

At the beginning of 1998, the private sector’s debt burden ranged from close to 130 percent of GDP in Indonesia to over 200 percent of GDP in Korea and Thailand. For details on the debt problem, see Chapter III.

Under this (Frankfurt) agreement, interbank debt will be rescheduled; trade credits will be rolled over and maintained at an agreed level; and a debt-restructuring agency (INDRA) will be established to facilitate the repayment of all corporate debt, which has been restructured on specified terms.

For instance, the government continued to hesitate to implement fully the steps that had been announced to seize the assets and limit the oil export rights of major oil companies with overdue tax liabilities.

Indications are that most Russian banks did not meet obligations under ruble-U.S. dollar forward contracts that fell due on September 14.

For a more detailed analysis of the contagion effects of the Asian crisis on these three, and other, transition countries, see Steven Fries, Martin Raiser, and Nicholas Stern, “Macroeconomic and Financial Stability: Transition and East Asian ‘Contagion,’” EBRD Working Paper 28 (London: European Bank for Reconstruction and Development, 1998).

See Hans Peter Lankes and Nicholas Stern, “Capital Flows to Eastern Europe and the Former Soviet Union,” EBRD Working Paper 27 (London: European Bank for Reconstruction and Development, 1998).

Banking sector claims on the nongovernment sector in central and eastern European countries other than Croatia, the Czech Republic, and the Slovak Republic are only in the 15–30 percent of GDP range, and are even lower in Russia and other countries of the former Soviet Union.

The global current account discrepancy reflects the fact that current account imbalances do not sum to zero across all countries. Over history, this reflects a number of factors including differences in accounting practices. In the projection period, changes in the discrepancy reveal inconsistencies among country projections, or tensions in the global projections.

The recent drop in commodity prices is a third potential source of inconsistency but is likely to be a smaller factor in the current circumstances.

See Robert J. Gordon, “The Time-Varying NAIRU and Its Implications for Economic Policy,” Journal of Economic Perspectives, Vol. 11 (Winter 1997), pp. 11–32; and Douglas Staiger, James H. Stock, and Mark W. Watson, “The NAIRU, Unemployment and Monetary Policy,” Journal of Economic Perspectives, Vol. 11 (Winter 1997), pp. 33–49.

See Joseph Stiglitz, “Reflections on the Natural Rate Hypothesis,” Journal of Economic Perspectives, Vol. 11 (Winter 1997), pp. 3–16, for a more detailed discussion of the NAIRU.

The participation rate is defined as the percentage of the working-age population that is either employed or actively seeking employment.

These issues were discussed in more detail in the October 1997 World Economic Outlook.

See Flint Brayton and Peter Tinsley, eds., A Guide to the FRB/U.S.: A Macroeconomic Model of the United States, Federal Reserve Board Working Paper 1996–42 (Washington: Board of Governors of the Federal Reserve System, 1996); and Martha Starr-McCluer, “Stock Market Wealth and Consumer Spending,” Finance and Economic Discussion Series, No. 1998-20 (Washington: Board of Governors of the Federal Reserve System, April 1998).

See Garry J. Schinasi and Monica Hargraves, “‘Boom and Bust’ in Asset Markets in the 1980s: Causes and Consequences,” in Staff Studies for the World Economic Outlook (Washington: IMF, December 1993), pp. 1–27.

See Alexander W. Hoffmaister and Garry J. Schinasi, “Asset Prices, Financial Liberalization, and Inflation in Japan,” in Ulrich Baumgartner and Guy Meredith, eds., Saving Behavior and the Asset Price ‘Bubble’ in Japan: Analytical Studies. Occasional Paper 124 (Washington: IMF, April 1995).

Nine of the 14 developing countries in the Middle East rely on oil exports for more than one-half of their export earnings. Fourteen of the 51 developing countries in Africa depend on a single primary commodity for 50 percent or more of their export earnings (4 on crude petroleum and 10 on a single nonfuel primary commodity). For 20 more African countries, the leading two or three primary commodity exports easily account for at least half of their earnings from exports of goods and services. The degree of commodity dependence is typically much less for developing and transition economies in Europe, Asia, and the Western Hemisphere. For example, for only three such countries does a single commodity account for more than one-half of their export earnings; this list includes Venezuela on account of its heavy dependence on oil exports.

Many other countries are, of course, being adversely affected also. In particular, in some transition countries energy is a leading sector helping to drive economic recovery and attract foreign investment, and the decline in energy prices, in addition to having a direct negative effect on output, could slow the growth of investment inflows and hence weaken longer-term growth prospects.

Countries in other regions, such as Chile, Kazakhstan, Mongolia, Papua New Guinea, and Peru, are also clearly affected by the copper price decline.

For a more detailed analysis, see Carlo Cottarelli and György Szapáry, eds., Moderate Inflation: The Experience of Transition Economies (Washington: IMF and National Bank of Hungary, 1998); and Carlo Cottarelli and Peter Doyle, eds., Disinflation in Transition—1993–97 (Washington: IMF, forthcoming).

See the evidence in Cottarelli and Szapáry, Moderate Inflation: The Experience of Transition Economies, and in Atish R. Ghosh and Steven Phillips, “Inflation, Disinflation, and Growth,” Working Paper 98/68 (Washington: IMF, May 1998).

For a discussion of the mechanisms whereby low inflation still may have a negative effect on the level of output, see, for example, John Driffill, Grayham Mizon, and Alistair Ulph, “Costs of Inflation,” in Benjamin Friedman and Frank Hahn, eds., Handbook of Monetary Economics, Vol. 2 (Amsterdam and New York: North-Holland, 1990); and Howell Zee, “Welfare Cost of (Low) Inflation: A General Equilibrium Perspective,” Working Paper 98/111 (Washington: IMF, August 1998).

For an analysis of these three factors, see Olivier Blanchard, “Optimal Speed of Disinflation: Hungary,” in Cottarelli and Szapáry, eds., Moderate Inflation, pp. 132–46.

    Other Resources Citing This Publication