The staff’s revised projections are summarized in Table 3, where they are also compared with the projections in the October 1997 World Economic Outlook. For the ASEAN-4 countries, Hong Kong SAR, Japan, and Korea, Table 4 compares the latest projections for output growth and current account balances with the projections published in both the May and October 1997 editions of the World Economic Outlook. Table 5 shows the latest projections of fiscal balances for the major industrial countries.

The staff’s revised projections are summarized in Table 3, where they are also compared with the projections in the October 1997 World Economic Outlook. For the ASEAN-4 countries, Hong Kong SAR, Japan, and Korea, Table 4 compares the latest projections for output growth and current account balances with the projections published in both the May and October 1997 editions of the World Economic Outlook. Table 5 shows the latest projections of fiscal balances for the major industrial countries.

Table 3.

Overview of the World Economic Outlook Projections

(Annual percent change unless otherwise noted)

article image
Note: Real effective exchange rates are assumed to remain constant at the levels prevailing during October 9–November 5, 1997, 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 $20.42 in 1996; the assumed price is $19.39 in 1997 and $19.03 in 1998.

Average, based on world commodity export weights.

London interbank offered rate.

Table 4.

Revisions to World Economic Outlook Projections for Selected Asian Countries

(Annual percent change unless otherwise noted)

article image
Table 5.

Major Industrial Countries: Actual and Structural Budget Balances for General Government

(In percent of GDP and potential GDP, respectively)

article image

Estimated balance assuming output equal to potential. Structural balances reflect the effects of all noncyclical factors, including temporary measures, one-off operations, and accounting changes. They are not necessarily appropriate measures of the underlying fiscal position, particularly at the current juncture for likely early participants in EMU. Estimates are also contingent on estimated potential output gaps.

Figures in brackets exclude social security. The projections were finalized prior to the announcement on December 17 of an income tax cut and new public investment spending amounting to about 1 percent of GDP.

The Revised Baseline

Underlying the revised projections are assumptions about exchange rates, commodity prices, and economic policies that have been formulated in the usual way on the basis of recent information. Another important assumption underlying the projections is that the decline in net private capital flows to developing and newly industrialized economies estimated for 1997—a decline of close to $80 billion, accounted for mainly by Asia (Table 6)—is not fully reversed in 1998, and that capital flows recover only partially by 1999. This decline in private capital flows in 1997–98 relative to 1996, together with changes in exchange rates and other financial market variables that have occurred in recent months in the countries affected by the Asian crisis, dominates the revisions to the projections shown here, although the revisions also reflect other changes that have occurred and new information that has become available since the October 1997 World Economic Outlook was finalized.

Table 6.

Developing Countries, Countries in Transition, and Newly Industrialized Economies: Net Capital Flows1

(In billions of U.S. dollars)

article image

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, Taiwan Province of China, and Israel.

The decline in private external financing for emerging market economies is assumed to lead developing countries to reduce their current account deficits (relative to the October 1997 World Economic Outlook) by about $22 billion in total in 1997, this impact being cushioned by drawdowns of reserves and increased official borrowing (Table 7). In 1998, with countries adjusting more fully, the developing world’s combined current account deficit is projected to be about $51 billion smaller than was envisaged in the October World Economic Outlook. Whereas the correction in 1997 is fully accounted for by Asia (mainly China, Indonesia, and Thailand), in 1998 significant adjustment is also projected for the emerging market countries of the Western Hemisphere. The current account deficit projected for Korea, within the group of advanced economies, has also been revised down substantially for 1998. In relation to output, the developing countries’ combined current account deficit in 1998 is now projected to be ¾ of 1 percent of GDP smaller than envisaged in the October World Economic Outlook, but roughly unchanged, at 1½ percent of GDP, from its levels in 1996 and 1997.

Table 7.

Overview of Current Account Projections

(In billions of U.S. dollars)

article image

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.

The external adjustment that the emerging market economies undertake as a result of the decline in private financing may be expected to occur partly through compression of domestic demand and partly through improved international competitiveness as a result of the currency depreciations that have occurred. The first mechanism—operating through reduced funding by private investors, a higher cost of capital, loss of confidence, and a tightening of economic policies—will tend to slow the growth of activity both in the adjusting economies and, indirectly, among their trading partners in the rest of the world. The second mechanism will also tend to slow the growth of trading partners’ economies by lowering their net exports, while in the adjusting economies, the effect on aggregate demand and activity may be positive or negative, because net exports will be boosted while domestic demand will be reduced by the impact of currency depreciations on consumers’ real wealth and income.

The projected outcome is characterized by a marked slowing in the growth of imports and activity among the developing countries, the newly industrialized Asian economies, and the countries in transition in 1998, relative to the October 1997 projections, with a more moderate deceleration in the same variables among the industrial countries. An upward revision to the growth of developing countries’ exports indicates how activity in those countries is expected to be supported by foreign demand as a result of exchange rate changes, while the downward revisions to the growth of exports in the other two groups indicate the contractionary impact of the slowing of expansion in the developing world.

The growth of output in the developing countries is now projected to weaken to just below 5 percent in 1998, which would be the slowest expansion for these countries since 1991 and 1¼ percentage points below the growth rate projected in the October 1997 World Economic Outlook. Growth projections have been revised down for all regions, but the largest changes are for Asia and, to a lesser extent, the Western Hemisphere—the two regions where private capital inflows have been largest in recent years and where they are now seen as declining the most. In fact, Asia is the only region where growth in 1998 is now projected to be significantly slower than in recent years: in the other regions, growth is expected to be maintained at rates broadly matching those observed in the first half of the 1990s, except that in Africa growth is projected to remain higher than in any recent year other than 1996. The relatively small effects of the crisis on growth in Africa and the Middle East are attributable to the comparatively low levels of private capital flows to most countries in these regions. There will, however, be some negative impact from the slowdown in global growth, its effects on commodity prices, and the competitiveness gains made by some trading partners.

Among the advanced economies, 1998 growth projections for the newly industrialized economies of Asia have been marked down substantially—from 6 percent to 3½ percent—with the most significant adjustment occurring in Korea. The 1998 projections have also been revised down further for Japan, which is now expected to experience markedly slower growth in both 1997 and 1998 than in 1996—another interruption, in effect, of the hesitant recovery begun in 1995. For the industrial economies of Europe and North America, the downward revisions of growth are much more moderate because of the smaller shares of trade with Asia and in view of the stronger-than-expected momentum of growth observed recently in these countries. In the European Union, growth is expected to be reasonably well-sustained at rates close to or a little above potential, while in the United States economic expansion is projected to continue in 1998 at rates close to potential, despite a slightly more marked slowing than was envisaged in the October 1997 World Economic Outlook. In both the United States and the United Kingdom, the moderation of growth should help to reduce the risk of increasing inflationary pressures. In the absence of the Asian crisis, the projections for North America and Europe would have been revised up slightly; the crisis is expected to subtract between ¼ and ½ of 1 percentage point from the growth rates that would otherwise have been projected at this stage (see Box 3).

For the world economy as a whole, output growth is now projected to slow from the 4 percent rate seen in 1996 and 1997 to about 3½ percent. While this represents a downward revision by ¾ of 1 percentage point from the October 1997 World Economic Outlook, it is notably higher than the growth rates of 1¾–2¾ percent recorded in the global slowdown of 1990–93. Indeed, the global growth rate projected for 1998 even now is slightly above the average experience of the past two decades. Nor does the slowing of world trade growth now projected for 1998 imply a rate of expansion particularly low by recent historical standards.

A notable feature of the revised projections, which indicates a tension in them, is that the adjustment of external current account balances expected to be undertaken by the developing countries and newly industrialized Asian economies—amounting to $15 billion in total, comparing 1998 with 1997, mostly accounted for by the Asian countries—is significantly smaller than the deterioration of $63 billion projected for the industrial countries. As a result, there is a widening of about $50 billion in the global current account discrepancy, as shown in Table 7. This implies that in reality either the emerging market economies will need to undertake more adjustment in response to reduced capital inflows than allowed for in this baseline scenario or that the deterioration in the industrial countries’ external positions will be smaller, which would permit larger net capital flows to the emerging market countries. The former seems the more plausible response to the general reassessment of emerging market risk that is under way in financial markets. This suggests that the risks to the baseline scenario are predominantly on the downside, particularly for imports and growth in the emerging market countries.

Effects of the East Asian Financial Crises on Trade with Industrial Countries

The turmoil affecting Asian emerging market economies has substantially reduced their immediate growth prospects, since the abrupt tightening of financial conditions is expected to sharply compress domestic demand. The effect of the contraction in domestic spending on overall output will be partially offset, however, by improvements in their external balances, as lower spending directly reduces imports and as currency depreciations raise relative import prices. At the same time, depreciated currencies will enhance the competitiveness of domestic producers and boost exports. Both responses will be contractionary from the point of view of trading partners, with the magnitude of the effect in any country depending on the share of trade with the Asian region in its total trade and on the importance of trade in relation to its overall output. (Another factor influencing the effect in any country will be the commodity composition of its trade, but the following analysis abstracts from this.)

The table shows these shares for the major industrial countries as well as Australia and New Zealand, on the basis of bilateral trade data for 1996. Of the major industrial countries, Japan has easily the largest share of trade with the Asian region, with roughly one-third of its total trade taking place with the ASEAN-4 countries and the newly industrialized Asian economies combined. The share of the United States is about one-half that of Japan, and shares of the other G-7 countries are smaller still. Of the other industrial countries, Australia and New Zealand have significant shares of trade with Asian emerging market economies, with Australia’s share approaching 35 percent of total trade.

The effect of changes in trade with Asia on the GDP of industrial countries also depends on the importance of trade in their total output. In this respect, there are wide differences across countries. For the United States and Japan, the ratio of trade to GDP is relatively low at somewhat over 10 percent for the average of exports and imports; for the other major industrial countries, this ratio ranges from 23 percent for Italy to 43 percent for Canada.1 Taking both the shares of trade with Asian partners and the ratios of trade to GDP into account allows the calculation of simple “rules of thumb” for assessing the effect of a slowdown in trade with Asia on demand for industrial country output.

Industrial Countries: Shares of Trade with Emerging Market Economies and Ratios of Trade to GDP

(In percent)

article image

Average share of merchandise exports and imports to or from each group of countries in total trade, based on data for 1996 from the IMF’s Direction of Trade Statistics.

Comprises Argentina, Brazil, Chile, China, Czech Republic, Egypt, Estonia, Hong Kong SAR, Hungary, India, Indonesia, Israel, Korea, Latvia, Malaysia, Mexico, Philippines, Poland, Romania, Singapore, South Africa, Taiwan Province of China, Thailand, Turkey, and Venezuela.

Average ratio of real exports and imports to real GDP in 1996.

Suppose, for instance, that imports of the countries most affected by financial turmoil—the ASEAN-4 plus Korea—decline by 10 percent from the levels that would have been observed in the absence of the financial crises, whereas imports of the other Asian newly industrialized economies are reduced by 5 percent.2 This would imply a reduction in aggregate exports of industrial countries of about 1 percent, or about 0.2 percent of their combined GDP. The direct effect would be relatively large for Japan, at about 0.4 percent of GDP; at the other end of the scale, Canadian GDP would be reduced by slightly more than 0.1 percent.3 The effects for Australia and New Zealand would be similar to those for Japan. If, in addition, export volumes of the countries affected by the financial crises expanded by one-half as much as the decline in import volumes in response to currency depreciations,4 this would raise the direct effect on the industrial countries as a group to 0.3 percent of GDP, with proportionate increases for the individual countries.

Beyond these direct effects, the crises will have “second-round” effects in the industrial countries, as incomes and profits are reduced by the deterioration in external positions and as domestic demand is lowered in turn. Simulations of these second-round effects using the IMF’s multicountry macroeconometric model, MULTI-MOD, indicate a trade “multiplier” for the industrial countries of about 1.5 on average—in other words, the overall effect on output would be roughly 50 percent larger than the direct impact on the real trade balance. Using the illustrative magnitudes mentioned above, this would imply an effect on all industrial countries of around 0.4 percent of GDP, with that for Japan rising to about 0.8 percent of GDP.

For the time being, the financial crises in Asia have primarily affected the Asian economies, with secondary effects on other emerging markets of varying intensity and duration. Although containment of contagion from the Asian crises and a return to relative financial stability appear to be reasonable prospects if policy measures are taken to restore confidence quickly—both in Asia and other developing country regions—a less benign outcome is easily imaginable. The main risk is that financial contagion from the Asian crises will have a pronounced effect on the flow of external financing to all emerging markets, requiring a large and simultaneous correction in their external balances. To judge the possible magnitude of the impact on industrial countries under this scenario, the table shows the shares of industrial countries’ trade with all major emerging market economies. Japan’s trade share is only slightly larger than that with the Asian region as a whole, whereas the shares of the United States and some of the European countries in trade with the emerging market economies are up to twice as large. It is evident that a more widespread shock to trade that affected all emerging markets would have much larger effects on the industrial countries than those discussed above, and that these effects would also be more evenly distributed.

The above calculations are, of course, only illustrative. The magnitudes of the trade shocks assumed for Asian and other emerging market economies are rather arbitrary. It remains to be seen how deep and widespread the effects of the Asian financial crises will be, and also what the timing will be of their effects on capital flows and trade.

In addition, the baseline outlook for both the industrial and emerging market economies is affected by many other factors, including policy responses in industrial countries and developments in exchange markets. Thus, while it is relatively clear (as reflected in the revised baseline projections discussed in the text) that the current account positions of the United States and the European Union will deteriorate in the near term to absorb the improvement in the external positions of the emerging market countries of Asia, the change in the outlook for Japan’s external balance is more difficult to ascertain. On the one hand, Japan’s trade balance will suffer more, directly, from the slowdown in the rest of Asia than those of other industrial countries; but on the other hand, weaker domestic demand in Japan and the weaker yen than assumed in the October 1997 World Economic Outlook will attenuate any deterioration in the external position. This illustrates how the above calculations should be considered as rough rules of thumb for assessing the approximate size of the direct impact of the crisis through trade, as opposed to estimates of how the external positions of the industrial countries will evolve over the coming year, taking all factors into account.

1 Adjusting for these differences in the trade intensity of output, the ratio of trade with Asian emerging market countries to total GDP is quite similar across the G-7 countries (excluding Japan) at about 2 percent of GDP, whereas for Japan it approaches 4 percent of GDP.2 This would correspond to an improvement in the combined current account surplus of the affected countries of roughly $60 billion (excluding the re-exports of Hong Kong SAR to China). By comparison, in the wake of Mexico’s 1994–95 crisis, its import volume dropped by 27 percent in 1995 relative to its trend growth in the previous three years, against the background of a 50 percent depreciation of the peso against the U.S. dollar. In Argentina, import volumes fell by 12 percent in 1995, as domestic spending slumped while the exchange rate remained at its pegged value to the dollar.3 In addition to the effects on trade volumes, Canada could be affected by adverse movements in world commodity prices, although the magnitude of the impact is difficult to judge at this stage.4 Mexico’s export volumes jumped by 33 percent in 1995 compared with 10 percent growth in the previous three years, while Argentina’s export growth rose to 23 percent from 7 percent in previous years.

The recent evolution of private sector forecasts for the major industrial countries and the other major economies affected by the Asian crisis is illustrated in Table 8. The latest available consensus private sector forecasts refer to projections formulated between mid-October and November in some cases, and therefore do not take full account of recent developments in Asia.

Table 8.

Selected Countries: GDP Growth Forecasts

(In percent a year)

article image
Sources: IMF; and Consensus Economics.

Consensus Economics, Consensus Forecast surveys.

May 1997 World Economic Outlook.

October 1997 World Economic Outlook.

Forecasts for the selected developing countries in Asia and the Western Hemisphere are from the November and October surveys respectively, which are the latest available.

Comprises Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela.

Alternative Scenarios

Considerable uncertainty remains about the depth and duration of the crisis. The baseline scenario described above presents a cautiously optimistic view of its implications: in the directly affected economies, economic activity would rebound by 1999, and the implications for other countries would be relatively moderate. Some precedents—including developments in Mexico, Argentina, and elsewhere in the wake of the 1994–95 “tequila crisis”—suggest that such a progressive return of confidence is feasible if adequate policy measures are implemented quickly.

At this stage, however, there are no clear signs that the crisis is subsiding, and doubts persist in financial markets about the policy response in several of the affected countries. Against this background, scenario 1—generated by the IMF’s multicountry macroeconometric model (MULTIMOD)—illustrates the potential implications of a substantially more pronounced and prolonged cutback in capital flows to emerging markets (Table 9). The additional curtailment of capital flows is assumed to reach a magnitude of roughly $100 billion (almost 2 percent of emerging market countries’ GDP) and to be associated with a sharp widening in risk premiums for emerging market borrowers, implying a sizable depreciation of their currencies and rise in domestic interest rates.11 These, along with declining confidence, contribute to contractions in domestic demand and an associated strengthening in these countries’ trade balances. (for further details on the simulations, see Box 4.) While the effect in the first year is cushioned to some extent by a drawdown of foreign exchange reserves, by the second year the trade balance strengthens essentially by the full magnitude of the reduction in net capital inflows. The resulting compression of domestic demand is accompanied by a decline in output of about 4½ percent relative to baseline by the second year. Mirroring the strengthening trade balance of emerging market economies, exports weaken in the industrial countries. Assuming that monetary policy remains essentially unchanged from the baseline scenario, output in industrial countries declines, relative to baseline, by ¾ of 1 percent by the second year.

Table 9.

Alternative Scenarios: Some Simulation Results

(Deviations from baseline; in percent unless otherwise noted)

article image

Baseline is based on current World Economic Outlook database, with shocks starting in 1997. Scenario 1 models a jump in the risk premium for emerging markets in the first year that gradually dissipates over the first five years; a similar jump in the risk premium in scenario 2 dissipates over three years. Also, the monetary reaction function in the emerging market economies differs somewhat across the two scenarios, and in industrial countries monetary policy is somewhat more accommodative in scenario 2 owing to the weakening in output. See Box 4 for details.

Comprises Argentina, Brazil, Chile, China, Czech Republic, Egypt, Estonia, Hong Kong SAR, Hungary, India, Indonesia, Israel, Korea, Latvia, Malaysia, Mexico, the Philippines, Poland, Romania, Singapore, South Africa, Taiwan Province of China, Thailand, Turkey, and Venezuela.

Comprises Canada, France, Germany, Italy, Japan, the United Kingdom, the United States, and the following smaller industrial countries: Australia, Austria, Belgium, Denmark, Finland, Greece, Ireland, the Netherlands, New Zealand, Norway, Portugal, Spain, and Switzerland.

Rest of the world, excluding transition economies.

The magnitude of the potential downside risks depends on the policy steps undertaken in the directly affected economies as well as in other countries. This is illustrated in scenario 2, where appropriate policy actions are assumed to facilitate a quicker restoration of confidence; with larger international financial support, the overall drop-off in financing flows to emerging market economies is limited to about $50 billion (some 1 percent of emerging market countries’ GDP, or about half the drop-off in scenario 1). In these circumstances, the decline in GDP in these countries is also likely to be about half of the decline observed in scenario 1, and the recovery in economic activity could be expected to occur more rapidly. With stronger external financing flows, the exchange rate depreciation would be more limited. In addition, it is assumed here that monetary policy in industrial countries, recognizing the weakening in activity emanating from the develop ments in emerging markets, responds with a somewhat more accommodative stance than is assumed in the baseline scenario. In these circumstances, the output decline in industrial countries would be limited to ¼ of 1 percentage point, and the spillover effects on developing countries would also be quite small.

Alternative Scenarios: Assumptions and Interpretation

To illustrate some of the risks that attach to the baseline scenario, Table 9 in the text summarizes two scenarios based on the IMF’s multicountry macroeconometric model (MULTIMOD):1 scenario 1, which entails a further drop-off in external financing flows to emerging markets peaking at about $100 billion (some 2 percent of these countries’ GDP); and scenario 2—a crisis-containment scenario—which examines a smaller and less prolonged drop-off in financing flows. Additional details on the simulations are provided here.2


In the simulation exercises, the drop-off in external financing flows to emerging markets is modeled as mainly resulting from a sharp rise in the risk premium these economies face in international capital markets. In addition, and as already evident in some cases, these economies are assumed to experience declines in domes-tic demand. Between the two scenarios, there are also some differences in monetary policy. More specifically, the assumptions (relative to the baseline scenario) are:

  • Risk premium. In scenario 1, the risk premium for emerging markets is assumed to rise by 9 percentage points during the first two years and by 5 percentage points over the next three years before reverting to its baseline level.3 In scenario 2, the increases would be 9 percentage points for the first year and 5 percentage points for the next two years.

  • Domestic demand shock in emerging markets. To reflect the effect of weakening confidence and financial sector stress, private consumption and imports are assumed to fall short of the levels that would be predicted on the basis of the explanatory variables embedded in MULTIMOD, which do not include confidence factors. In scenario 1, the negative shock to consumption is about 1¼ percent of GDP, and about ¾ of 1 percent of GDP in the case of imports. In scenario 2, the shocks are about two-thirds this size.

  • Monetary policy in emerging markets. Scenario 1 assumes that monetary policy in the emerging market economies—in the context of trading off higher interest rates against lower exchange rates—accommodates a relatively larger depreciation of domestic currencies, other things being equal. In scenario 2, by contrast, monetary policy is assumed to place a relatively larger weight on exchange rate stabilization, other things being equal, through stronger interest rate adjustments.

  • Monetary policy in industrial countries. In scenario I, it is assumed that nominal interest rates remain unchanged relative to the baseline scenario. In contrast, scenario 2 allows for a monetary reaction function similar to a Taylor rule, which implies that interest rates would decline to some extent and tend to dampen the negative effects on output and employment.4

Results and Interpretation

In scenario 1, the simulation results indicate a sharp decline in output in emerging markets, bottoming out at some 4½ percent relative to baseline by the second year (see Table 9 in Section V). The associated decline in domestic demand and depreciation of the nominal effective exchange rate by about 7¼ percent contribute to the improvement in the current account. Although a drawdown in foreign exchange reserves is assumed to cushion these effects somewhat in the first year, by the second year the current account improvement is assumed to match the almost $100 billion decline in net external fin anc ing. The effect on inflation is relatively muted because the inflationary impact of the exchange rate depreciation is counterbalanced by weaker economic activity.

The strengthening in the emerging market economies’ trade balance would be largely mirrored by a weakening balance for the industrial countries. With monetary policy keeping short-term interest rates close to the baseline trajectory, industrial country GDP falls by about ¾ of 1 percent relative to baseline over the first two years. With weaker demand in the emerging markets and in the industrial countries, growth would also suffer in those developing countries not included among the emerging market economies for the purposes of this exercise, falling almost ½ of 1 percent below baseline by the second year before gradually recovering thereafter.

In scenario 2, stronger policy adjustments in the emerging markets, coupled with additional financial support from official and private sources, are assumed to limit the reduction in external financing flows to about $50 billion (almost 1 percent of GDP in the recipient countries, or about half the reduction in scenario 1). Under these circumstances, the fall in domestic demand and output (relative to baseline) would also be about half as large. Moreover, the external support would help to limit the exchange rate depreciation in these economies and thereby also the negative spillovers to the rest of the world.

The negative effects on the global economy would also be mitigated in scenario 2 by a somewhat more accommodative monetary policy stance in the industrial countries. In this case, in view of the adverse repercussions from developments in emerging markets, monetary policy in the industrial countries would allow short-term interest rates to decline somewhat. With supporting policies being implemented in emerging market economies as well as in industrial countries, the spillover effects on other developing countries would be quite small, with GDP declining by less than ¼ of 1 percent relative to the baseline scenario.

1 See P.R. Masson, S. Symansky, and G. Meredith, MULTI-MOD Mark II: A Revised and Extended Model, Occasional Paper 71 (Washington: IMF, 1990). For the analysis here, the standard version of MULTIMOD was extended by introducing a separate bloc for emerging market economies.2 Although MULTIMOD has a nonlinear structure, proportionally smaller or larger drop-offs in external financing flows would lead, as a close approximation, to proportionally smaller or larger changes in the other variables shown in Table 9.3 The risk premium refers here to the amount that the short-term interest rate on a country’s own-currency liabilities exceeds the sum of the interest rate on short-term liabilities of the U.S. Treasury and the expected rate of depreciation of the domestic currency against the U.S. dollar. The size of the risk-premium shock is calibrated to generate the respective drop-offs in foreign financing flows discussed above. It can also be thought of as capturing the possibility that, in some cases, countries may be temporarily shut out altogether from borrowing in private capital markets.4 See J. Taylor, “Discretion Versus Policy Rules in Practice,” Carnegie-Rochester Conference on Public Policy (December 1993), pp. 195–214.

These scenarios illustrate some of the downside risks facing the emerging market economies as well as the global economy. However, they also underscore that adequate and timely policy responses can play an important role in limiting these risks and containing the crisis.