Back Matter

Back Matter

Ales Bulir, Marianne Schulze-Gattas, Atish Ghosh, Alex Mourmouras, A. Hamann, and Timothy Lane
Published Date:
February 2002
    • ShareShare
    Show Summary Details
    Appendix I Country Sample

    The sample of programs included as capital account crisis programs consists of:

    ArrangementOriginal Program Length
    (In months)
    Argentina (1995)1EFF12
    Brazil (1998)Stand-by36
    Indonesia (1997)Stand-by36
    Korea (1997)Stand-by36
    Mexico (1995)Stand-by18
    Philippines (1997)2EFF6
    Thailand (1997)Stand-by34
    Turkey (1994)Stand-by14

    Extension of the EFF originally approved in 1992.

    Extension of the EFF originally approved in 1994.

    In the main report, reference is also made to a sample of “other” program countries. This sample consists of the following programs:

    ArrangementOriginal Program Length
    (In months)
    Algeria (1994)Stand-by12
    Barbados (1992)Stand-by16
    Belarus (1996)Stand-by12
    Brazil (1992)Stand-by19
    Bulgaria (1994)Stand-by12
    Cameroon (1994)Stand-by18
    Central African Republic (1994)Stand-by12
    Chad (1994)Stand-by12
    Dominican Republic (1993)Stand-by9
    Ecuador (1994)Stand-by25
    Gabon (1994)Stand-by12
    Hungary (1993)Stand-by15
    India (1992)Stand-by20
    Jamaica (1991)Stand-by12
    Jordan (1992)Stand-by18
    Lithuania (1995)EFF36
    Malawi (1995)Stand-by7
    Moldova (1995)Stand-by12
    Niger (1994)Stand-by12
    Panama (1992)Stand-by22
    Pakistan (1994)EFF36
    Papua New Guinea (1995)Stand-by18
    Peru (1996)EFF33
    Poland (1991)EFF36
    Romania (1997)Stand-by13
    Russia (1995)Stand-by12
    Ukraine (1995)Stand-by12
    Uruguay (1992)Stand-by12
    Vietnam (1994)Stand-by12
    Republic of Yemen (1996)Stand-by15
    Appendix II Financial Fragilities and Official Financing

    It has often been asserted that the capital account crises of the 1990s constituted largely unexpected events. With the benefit of hindsight, however, there is now some evidence suggesting that, at least to some extent, the buildup of the financial vulnerabilities that made these crises possible reflected the consequences of a critical decision made by each of the governments of the countries affected; to finance, rather than adjust to, the effects of an exogenous shock through external borrowing or a rundown of international reserves. If this was indeed the case, an important question arises: Were countries that waited too long before acknowledging the need for adjustment and thus exposed their countries to a potential capital account crisis of a larger magnitude, rewarded with larger official aid packages?98 In order to address this issue, this appendix provides an analysis of the behavior of two simple indicators of vulnerability, to determine whether they exhibit any clear trends in the run-up to the crises. The results of this analysis are then related to the size of the official packages granted to each of these countries.

    The first of the two indicators examined below is a currency crisis index, which intends to measure exchange market pressures. This index is a weighted average of cumulative reserve losses and exchange rate depreciation, measured from a fixed starting point. The weights are given by the inverse of the variances of these two variables over a relatively long period.99 The second indicator, the ratio of international reserves to short-term debt, is a measure of financial fragility, which has been found to be a key determinant of liquidity and currency crises in recent theoretical work (i.e., Chang and Velasco, 1998) and has been found to be highly correlated with the probability of capital account crisis in empirical studies. It is important to stress here that the focus of the analysts is not on the systematic ability of these indices to forecast a crisis. Instead, this appendix provides an ex post analysis (particularly since some of the data used to construct these indices were not available at the time of the crises) of the indices for each country, with the purpose of detecting any trends prior to the crisis, which may be interpreted as an indication of specific policy actions pursued by the governments involved.

    Figures A2.1 and A2.2 show the behavior of the two vulnerability indices over a five-year span: four years prior to the countries’ adoption of IMF programs and one year after. 98 By construction, the specific values taken on by the exchange market pressure index shown in Figure A2.1 are not very meaningful, but increases in the index reflect downward pressures on the exchange rate (i.e., losses of reserves or a depreciation of the national currency).101 Specific values aside, it was possible to detect a steady increase in the index for at least six months prior to the adoption of IMF programs in three countries: Brazil, Mexico (where the index increases significantly for about a year prior to the program, with the exception of a single month: October 1994), and Thailand (where the index increases for nine consecutive months prior to the program).102 Despite not representing a strict case of a six-month deterioration in the pressure index, one could add Turkey to the list (the index deteriorates steadily only between November 1993 and April 1994, and improves for three months prior to the adoption of the program). In that case, the deterioration in the index coincides with a drastic and well-documented change in policy regime that the literature has identified as a key determinant of the currency crisis of April 1994: a switch in the source of financing of Turkey’s large fiscal imbalance, from bonds to money creation, which led to a reduction in domestic interest rates and a depletion of reserves, as predicted by the “first generation” of currency crises models.103 In the other four countries, the deterioration in the pressure index is either less protracted (Argentina and Korea) or both less protracted and less pronounced (Indonesia and the Philippines).

    Figure A2.1.Crisis Index1

    Sources: IMF, International Financial Statistics; national sources; and IMF staff estimates

    1 Weighted average of (i) the cumulative percentage change in the level of gross reserves (with a negative sign) and (ii) the exchange rate, beginning four years before the adoption of the IMF program under study. The weights used were the precision of each of the series (computed over the period January 1991–December 1999) divided by the sum of the precisions. This methodology follows Sachs, Tornell, and Velasco (1996).

    2 Reserves are net of deposits at foreign branches of local banks.

    3 Reserves are net of forward operations of the Central Bank of Thailand.

    Figure A2.2.Indicators of International Liquidity

    Sources: IMF, International Financial Statistics and World Economic Outlook database; national sources; and IMF staff estimates.

    1 Reserves exclude deposits at overseas branches and subsidiaries of domestic banks.

    2 Reserves are net of forward operations of the Central Bank of Thailand.

    Notably, of the four countries for which a steady deterioration of the pressure index was detected (including the case of Turkey), all except Brazil have been usually regarded in the literature as cases where the crises were, in some way, self-inflicted rather than the result of contagion.104 The crises in Turkey, Mexico, and Thailand certainly did not come in the aftermath of other currency crises in emerging markets and, moreover, the Mexican and Thai crises have been identified as the sources of regional contagion, as currencies in neighboring countries came under heavy pressure almost immediately after the peso and the baht were allowed to float. The Brazilian currency may have been ripe for a currency crisis during 1998 as a result of the cumulative effect of inconsistent policies over a relatively long period of time, as was discussed in Chapter II (more on this below). However, it would be difficult to ignore the possibility that the stability of the real may have been threatened also by external pressures stemming from Asia and, subsequently, from the Russian crisis of August 1998. On the other hand, the other four countries for which no steady deterioration in the index was found have been typically considered victims of contagion: Argentina, as a result of the Mexican crisis, and Indonesia, Korea, and the Philippines, following the Thai crisis (see, for example, Edwards, 2000).105

    Figure A2.2 shows the ratio of gross international reserves to short-term debt, an indicator of a country’s international liquidity.(Figure A2.2) also shows the evolution of reserves in billions of dollars.106) As mentioned earlier, this index is intended to capture a different dimension of the dynamics leading to currency crises (the vulnerability to financial panic) and, therefore, it provides information not contained in the exchange rate market pressure index. Not surprisingly, the evolution of this index tells a very different story from that obtained from Figure A2.2.

    In the case of the reserves-to-short-term-debt ratio, there would seem to be a natural numerical benchmark (one) against which one could compare the value of the index at different points in time, although, in practice, this is far from a settled issue.107 Despite its shortcomings, however, the evolution of this index reveals some distinctive patterns. There is a first group of countries (Korea and Turkey) for which the index stayed consistently and significantly below one during the four years prior to the crisis. There is a second group of countries (Indonesia, Mexico, and Thailand) for which the index stayed in the neighborhood of one for a relatively long period of time (at least two years) and then dropped significantly below one exactly three quarters prior to the adoption of an IMF program. Finally, there is a third group of countries where the index either (i) never dropped below one prior to the adoption of an IMF program (Brazil and the Philippines) or (ii) dropped slightly below one, but only briefly at the time of the crisis (Argentina).

    To the extent that one can associate financial fragility and thus the potential for financial panic with situations in which the index fell below one for more than a single quarter, the analysis of the previous paragraph would seem to suggest that, irrespective of the quality of their other policies. Turkey and Korea had been in a danger zone for several years prior to their crises. Indonesia, Mexico, and Thailand seem to have been on the edge of the danger zone for quite a while, before clearly entering it and remaining there for a full three quarters before they adopted an IMF program. Although in Mexico and Thailand the drop in the index seems to be driven by a fall in reserves, suggesting that their central banks were actively engaged in defending their currencies through intervention in the foreign exchange market, in Indonesia the drop in the index reflects entirely a buildup of short-term debt, as foreign exchange reserves were actually rising.108 In the absence of a more complete analysis of these countries’ other macroeconomic and structural policies, it is hard to determine the extent to which having the index staying close to one (as opposed to significantly above one) may in itself have reflected a potentially dangerous situation. This would imply that, like Turkey and Korea, this group of countries also spent several years in a financially dangerous zone, before experiencing their crises.

    Finally, neither Brazil nor the Philippines appears to have been near the danger zone, while in Argentina the temporary drop in the index reflects the effects of the crisis itself, and the authorities’ willingness to allow the automatic adjustment mechanism built in their currency board arrangement to operate. Nonetheless, it must be stressed that even though Brazil did not enter the danger zone until after the program with the IMF was signed, it experienced a large loss in gross reserves (over $30 billion, or nearly one-half of its holdings, between February and November of 1999) and a strong, albeit less dramatic, fall in the reserves-to-short-term-debt ratio, as capital outflows during the period caused some reduction in the stock of short-term debt.

    Based on the analysis presented so far, it would appear safe to say that, prior to the adoption of an IMF program, three of the countries in our sample (Mexico, Thailand, and Turkey) were subject to strong exchange rate market pressures, while being in a financially dangerous zone and exhibiting an increasing degree of financial fragility (as a result of deliberate intervention in foreign exchange markets) for at least half a year. Although Korea clearly stayed in the financially dangerous zone for a long time, Korea’s case is somewhat different in that it appears to have experienced exchange market pressures (including a sizable loss of reserves) over a shorter period of time in the run-up to its crisis. Brazil experienced exchange rate pressures and loss of a significant amount of reserves over a period of at least six months, but it did so while still preserving a relatively safe degree of reserve coverage of its short-term debt. The indices provide much weaker evidence that the other countries (Argentina, Indonesia, and the Philippines) engaged in a systematic policy of deferring adjustment over a protracted period in the face of an adverse external shock.

    Given these results, and the analysis of the official financing provided to the crisis countries in Chapter III, there does not appear to be a clear correlation between the size of the official packages and the extent to which the countries may have been attempting to defer an unavoidable adjustment in domestic policies. Although the Mexican package was among the largest on the basis of various benchmarks, the Thai and Turkish packages were small relative to the others, especially in relation to short-term debt. There is no clear pattern either in the case of the countries that did not try to postpone adjustment for a relatively long time: Indonesia’s package was among the largest according to most measures except short-term debt, while the Argentine and Philippine programs were small. The Brazilian and Korean packages were on the large side but were not among the largest two. In other words, this appendix does not find empirical evidence in favor of the hypothesis that countries that may have engaged in a systematic postponement of adjustment were rewarded by larger official packages in the context of IMF programs.

    Appendix III Calculation of Fiscal Sustainability and Fiscal Impulse Ratios

    Medium-Term Sustainability

    In the text, “medium-term sustainable” fiscal balances are reported. These are the primary balances which, given historical growth and interest rates, are sufficient to stabilize the initial public debt ratio. Even if the primary balance equals this “medium-term sustainable balance,” the actual debt-to-GDP ratio is likely to rise for at least two reasons. First, during a currency crisis, real GDP growth is likely to be lower, and real interest rates higher, than their historical norms. Second, there may be financial sector restructuring costs which explicitly or implicitly add to the stock of public debt. In the text, the estimated carry costs of financial sector restructuring are reported separately.109 (Conversely, however, there may be privatization receipts that lower the stock of public debt.) Finally, it bears emphasizing that, since the calculations embody a number of underlying assumptions, they should be considered indicative rather than precise estimates.

    In general, the widest concept of public sector for which there are available time series data on primary balances and domestic and external debt was used in the calculations. These correspond to:

    • Argentina: Consolidated public sector

    • Brazil: Public sector (federal, state, and public sector enterprises)

    • Mexico: Non financial public sector

    • Turkey: Overall public sector (consolidated budget, local authorities, state economic enterprises)

    • Indonesia: Central government

    • Korea: Consolidated central government

    • Philippines: National government

    • Thailand: Consolidated non financial public sector.

    In undertaking the calculations, three-year backward averages are used for GDP growth and interest rates.110 The estimated medium-term primary surplus at the year-end closest to the program approval date is reported in the text.

    Annual and Quarterly Fiscal Impulse

    The text reports programmed and actual annual fiscal impulses.

    Programmed Fiscal Impulse

    The programmed fiscal impulses are based on the original program projection of revenues and expenditures for the following concepts:

    • Argentina: Consolidated public sector overall balance

    • Brazil: Public sector (federal, state, and public sector enterprises) borrowing requirement

    • Mexico: Nonfinancial public sector overall balance

    • Turkey: Central government overall balance, cash basis

    • Indonesia: Central government overall balance

    • Korea: Consolidated central government overall balance

    • Philippines: National government overall balance

    • Thailand: Central government overall balance

    An HP-filter is fitted for the log of real GDP using annual data from 1980 through year t-1, to calculate potential output which, particularly for the Asian countries, differs significantly from the full sample (“actual”). Program projections are also used for real GDP and expenditure and revenue. A base year for the revenue and expenditure ratios, r0 and e0 is chosen such that the absolute value of the output gap is small. As a percent of GDP, the program fiscal balance is given by bp=rpep; while the cyclically neutral balance is given by bcp = r0e0(ypot/yprog) where ypot is potential output based on an HP-filter to log real GDP for the sample 1980 to t-1. The program fiscal stance is then fsp=bcp – bp and the program fiscal impulse is fip=Δfsp.

    Actual Fiscal Impulse

    The actual fiscal impulse is calculated in the same way, except that the HP-filter is applied to the entire sample period. 1980–99: actual outcomes are used for the revenue and expenditure ratio:ba=raea; and the actual GDP outcome is used for the cyclically neutral balance: bca=r0e0(ypot/ya), where ypot is potential output based on an HP-filter to log real GDP for the entire sample. Then the actual fiscal stance is fsa=bca-ba, and the actual fiscal impulse is fia=Δfsa. These are reported in Text Table 5.3, while the corresponding primary balances (i.e., excluding interest reports) are given in Table A3.1 above.

    Table A3.1.Primary Fiscal Balances and Fiscal Impulse Ratios: Programs versus Outcomes
    Real GDPPrimary balanceCyclically neutral balanceFiscal stanceFiscal impulseReal GDPPrimary balanceCyclically neutral balanceFiscal stanceFiscal impulse
    (Percent change)(Percent change)
    (As percent of GDP)(As percent of GDP)
    Argentina (1995)−1.6−−0.2
    Brazil (1999)1−1.01.8−0.2−2.0−−2.2−2.2
    Indonesia (1998)−2.3−0.7−13.11.1−1.3−2.4−2.0
    Korea (1998)12.51.8−1.3−3.1−2.8−6.7−3.1−
    Mexico (1995)1.53.4−2.5−5.9−1.1−6.24.4−4.5−8.9−4.5
    Philippines (1998)−0.7−
    Thailand (1998)13.51.2−0.1−1.3−3.1−10.8−−1.3
    Turkey (1994)−1.66.4−2.7−9.1−9.3−4.73.7−2.8−6.5−7.3
    Sources: IMF, World Economic Outlook; IMF, MONA; and IMF staff estimates.

    Dates refer to year t (or closest fiscal year), rather than to program approval year.

    Quarterly Fiscal Impulse

    Text Figure 5.1 also reports quarterly fiscal impulses. These are calculated in an analogous fashion, except that the real GDP series is seasonally adjusted using multiplicative X–11, and the HP-filter is applied to the logarithm of the seasonally adjusted series to obtain potential output. Rather than choose a single base year, quarter-specific revenue and expenditure base ratios, rio andeioi=1 . . ., 4 are calculated as the average of the quarter-specific revenue and expenditure ratios over the full sample (since, by definition, average potential and actual real GDP should be equal).111 Quarterly impulses are then defined as the quarter-to-quarter change in the fiscal stance, and graphed against the quarter-to-quarter change in (seasonally adjusted) real GDP. In some cases, a somewhat narrower fiscal concept was used for the quarterly impulses due to data availability limitations.

    Appendix IV Vector Autoregression Estimates of Real Money and Real GDP Relationship

    The text makes reference to impulse response functions of real GDP growth to real money growth or real credit growth; these are reported in Table A4.1 below.

    Table A4.1.Impulse Response Functions 1
    Impulse Response of Real GDP Growth to
    10 Percent Shock to Real Money Growth2
    Impulse Response of Real GDP Growth to
    10 Percent Shock to Real Money Growth 3
    Source: IMF staff estimates.

    Vector autoregressions with four lags over 1985:1–1999:4, as available.

    Assumes 10 percentage point increase in real money Growth in the first quarter.

    Assumes 10 percentage point increase in real credit Growth in the first quarter.

    Appendix V Chronologies of Events in Countries’ Capital Account Crises

    Argentina (1995)

    The capital account crisis in Argentina was short lived, mainly because the banking system was strong enough to withstand a period of high interest rates and because the authorities acted expeditiously.

    In 1991, Argentina launched a comprehensive set of reforms aimed at ending a protracted period of high inflation and restructuring its economy in a fundamental way. A key element of the program was the adoption of a currency board through the introduction of the “Convertibility Plan” in March 1991 and the enactment of the necessary supporting legislation. In addition, the government adopted a trade reform; abolished all price controls; deregulated wholesale and retail trade; privatized ports and public utilities; and reformed the financial sector. These efforts were supported initially by a Stand-By Arrangement (SBA) approved in July 1991 and subsequently by a three-year Extended Fund Facility (EFF) approved in March 1992.

    Between 1991 and 1994, real GDP growth averaged close to 9 percent, while consumer price inflation declined from hyperinflationary levels to 3.9 percent at end-1994. Growth was driven by a large expansion in demand, particularly in investment, leading to a five-fold increase in imports and a deterioration in the current account, which moved from near balance in 1991 to a deficit of about 4 percent of GDP in 1994. The Fiscal accounts recorded a significant improvement through 1993, but worsened in 1994, reflecting the shift of employees’ social security contributions to the private pension system and a worsening in the provincial finances. Nonetheless, by end-1994 the magnitude of the fiscal imbalance was not alarming: the overall public sector deficit in 1994 was 2.3 percent of GDP and the public sector debt was 32 percent of GDP.

    The reforms also led to a very fast process of re-monetization: fueled by rapid Growth in bank deposits financed with capital inflows, the M3 monetary aggregate-to-GDP ratio grew from 6 percent in 1990 to 19 percent in 1994. At the same time, bank credit to the private sector rose at an average rate of 2½ percent a year in real terms. Financial deepening was accompanied by improvements in productivity in the banking industry. As part of the reforms carried out since 1991, the government eliminated arrears and restored access to domestic and international capital markets, including through a Paris Club agreement and a Brady Bond restructuring in 1992. Since 1993, the government was able to place long-term notes in domestic and external markets for more than $8 billion (3 percent of GDP).

    Through September 1994, Argentina had made all of its scheduled purchases under the EFF on time, but the program went off-track in the second half of 1994, and the Mexican crisis of December 1994 spilled over swiftly into Argentina. Interest rates rose sharply and the Argentine stock and bond markets fell. The pressure continued in early 1995 and, by the end of the first quarter of 1995, the central bank lost one-third of its gross reserves and the domestic deposit base shrank dramatically. The associated liquidity crunch led to another surge in interest rates (the interbank rate peaked at 70 percent), a sharp fall in asset prices (stock and bond prices fell by about one-half from end-1994 to March 1995), and declining profitability of financial institutions.

    Contagion from the Mexican financial crises highlighted the fragilities of the 1991 currency board arrangement, particularly the absence of a formal deposit insurance scheme and of a lender of last resort. Despite the limited powers contemplated for it by that law, the central bank announced a series of measures. The central bank accommodated the outflow by gradually lowering reserve requirements on U.S. dollar and peso deposits. In addition, bank deposits in the central bank were dollarized in an attempt to give confidence to the markets and a fund was set up to purchase nonperforming loans of distressed financial institutions. In February, the central bank charter was modified to give the central bank additional powers to assist troubled financial institutions. Despite all these measures, deposit withdrawals and reserve losses continued, and, as a result, the central bank’s ratio of reserves to monetary liabilities approached the statutory limit of 80 percent, which effectively eliminated the margin for further central bank support to the financial system.

    On March 23, 1995, the authorities provided the details of an IMF-supported adjustment program. The program had two main objectives; first, to reinforce the public finances and demonstrate unequivocally the government’s ability to service its maturing obligations and, second, to deal with a fragile financial system in the absence or a lender of last resort, by setting two trust funds to provide resources for restructuring the financial system. The IMF supported the program with an extension of the 1992 three-year EFF to a fourth year, including an augmentation of SDR 1, 537.1 million (100 percent of quota). Although the program was approved only on April 6, 1995, there was a positive market reaction to its March announcement, when interbank and prime rates fell significantly. Moreover, in April 1995. Argentina was able to place two $1 billion bond issues, with domestic and international investors, respectively.

    Despite a reflow of deposits and a continued decline in interest rates during the remainder of 1995, credit to the private sector remained tight as banks struggled to rebuild their liquidity positions and increased lending to the government. This contributed to a significant decline in economic activity, particularly in the second and third quarters of the year. The compression of domestic demand combined with strong external demand led to a quick, albeit small, improvement in the external current account: the deficit narrowed from 3.7 percent of GDP in 1994 to 1.3 percent in 1995. Economic activity bottomed out in the last quarter of 1995, and GDP fell by some 3 percent in 1995.

    The thrust of the economic program was maintained throughout 1995. Despite the difficult economic and social situation, the authorities observed the discipline imposed by the currency board arrangement. As a result, by year-end more than 90 percent of deposits withdrawn during the crisis had been recovered, peso and U.S. dollar denominated prime interest rates were close to their pre-crisis levels, and spreads on Argentina’s sovereign debt narrowed considerably. Although there was some fiscal underperformance vis-à-vis the IMF targets, a tight fiscal stance was maintained in cyclically adjusted terms. A strong recovery of economic activity followed in 1996–97.

    A few characteristics differentiate Argentina from other country cases of capital account crises. First, its banks were healthy prior to the crisis: risk-based capital asset ratios and liquidity ratios were high at 18 percent and 20 percent, respectively. Second, debt management had become a policy priority early in the reform process and the authorities avoided any bunching of amortization payments. Third, crisis management was expeditious and had an immediate and positive impact on market confidence.

    Brazil (1997–99)

    By the mid-1990s, stabilization programs, including the 1994 Real Plan—an exchange rate-based stabilization program—had succeeded in ending Brazil’s chronic high inflation. Private capital inflows had picked up, including substantial foreign direct investment and portfolio flows, and the current account deficit had widened. The Brazilian economy did slow down appreciably as contagion from Asia and Russia spread worldwide during 1998. Still, positive Growth was registered in 1998, and the 1999 V-shaped recession was mild and short-lived in comparison with the contraction in Mexico during the “tequila crisis”112 and the experiences of Asia and Russia in the late 1990s.

    A variety of factors can explain why Brazil’s experience was less painful than those of Mexico, Asia, and Russia. On the face of it, Brazil was vulnerable on many fronts—certainly more so than Mexico in 1994–95. Of greatest concern were Brazil’s fiscal and current account deficits, growing public and external indebtedness financed at short maturities, and a rigid and less than fully credible crawling peg exchange rate regime. The public sector continued to be plagued by an expensive and actuarially unbalanced public pension system, excessive government payrolls, and an inefficient taxation system. After an initial primary surplus of 5.3 percent of GDP in 1994, the balance worsened to a deficit of 1 percent of GDP in 1997, financed mostly at short maturities. The fiscal deterioration was accompanied by an appreciation of the real exchange rate by about 20 percent between April 1994 and end-1997. The external current account deficit widened from approximate balance in 1994 to a deficit of 4 percent of GDP in 1997.

    Brazil’s vulnerabilities became evident following the outbreak of the Asian crisis when reserves fell by nearly $8 billion in October 1997 alone. To stem the outflow of reserves, the central bank doubled interest rates to about 45 percent in October, and the government adopted a fiscal package with an annual yield estimated at about 2½ percent of GDP. The outflow of reserves stopped and capital inflows resumed in the last two months of 1997, reducing the loss of reserves for the year as a whole to $8 billion.

    Doubts about the authorities’ resolve to implement the previously announced budget cuts in an election year and the presentation of relatively loose budget proposals for 1999, together with the persistence of a substantial current account deficit, signaled that Brazil remained vulnerable to external shocks in the fall of 1998, when the Russian default shook financial markets. In response, the authorities again tightened fiscal and monetary policies—interest rates were raised to 40 percent—to stem the outflow of reserves.

    Brazil approached the IMF and other international financial institutions for support at a relatively late stage in November 1998. The Board approved the three-year SBA on December 2, 1998. The program was premised on the notion that confidence would return—and the fixed exchange rate regime could be maintained—through ample official financing and a strong policy package. In view of the imbalances in the public finances, the IMF-supported program had at its core a fiscal package sufficient to stabilize the debt-to-GDP ratio and forestall unsustainable debt dynamics. To this end, the official financing package envisaged under the SBA amounted to $41.8 billion, including $18.1 billion from the IMF.

    The early December 1998 agreement had a calming effect on markets and led to easing of the pressures on reserves. Subsequent policy implementation faltered, however, as the central bank reacted to the easing of market pressures by beginning a rapid, premature reduction in interest rates, from 40 percent in November to 29 percent in December. Fiscal policy implementation suffered setbacks as well, aggravating earlier concerns: Congress defeated an important component of the fiscal package in early December and delayed other measures, while the state of Minas Gerais announced that it would not honor its debt to the federal government. As a result, outflows continued as foreign bank creditors refused to roll over their maturing credit lines, and the currency was subject to intense pressures in late 1998 and early 1999. After some interest rate increases and a one-day experiment with a wider band for the currency, the central bank was forced to float the real on January 15, 1999, and to seek to renegotiate and reinforce the IMF-supported program.

    The exchange rate regime change required modifications to the monetary framework underpinning Brazil’s IMF-supported program. Although these negotiations were under way in early 1999, the central bank continued to lose reserves ($8.5 billion during January-February) as the high interest rate policy proved insufficient in the absence of stronger measures to stem capital outflows. The reserve loss was stemmed in March 1999 following an agreement with the IMF on a revised monetary framework and the bailing-in of foreign creditors. Specifically, Brazil adopted inflation targeting to replace the nominal anchor lost when the exchange rate was allowed to float. A semi-voluntary form of private sector participation was secured when commercial bank creditors agreed in mid-March to maintain their interbank credit lines with Brazilian banks at their end-February level.

    The strengthened program was successful in restoring confidence quickly, and the performance of the Brazilian economy in 1999 was significantly better than expected. The central bank was able to gradually reduce interest rates from a peak of 45 percent to 21 percent in mid-July. After an initial period of overshooting, the currency stabilized at around R$l.75 to the U.S. dollar and voluntary capital flows resumed in April. Real GDP grew by slightly less than 1 percent in 1999, in contrast with the 4 percent decline projected originally. The unemployment rate declined slightly, but wage pressures were moderated considerably by slack labor markets, and consumer inflation was moderate in 1999–2000. The fiscal position also improved in 1999, as the program’s fiscal targets were met or exceeded. Progress was also registered in structural fiscal reforms, including significant steps in the reform of the social security system.

    The authorities followed a policy of relatively clean float, limiting intervention in the foreign exchange market to counter disorderly market conditions. They also met understandings reached with the staff to not reduce interest rates while intervening in the foreign exchange market and to promptly consult with the staff about appropriate policy responses in the event of substantial pressure on net international reserves.

    Despite its numerous vulnerabilities, Brazil enjoyed several advantages that are responsible for the soft landing of 1998–99. These advantages included steadily growing inflows of foreign direct investment, large foreign exchange reserves, and a relatively healthy banking system. Judicious policy implementation in the fall of 1997 also seems to have played a role in restoring confidence and averting a full-blown crisis during the initial phase of contagion.

    Indonesia (1997–98)

    Indonesia demonstrates consequences of an unresolved capital account crisis. Although the country’s starting macroeconomic position was probably more favorable than that of other Asian countries, the long-term costs of the crisis have been drastic owing to domestic political turmoil and general procrastination.

    The currency, the rupiah, came under pressure in July 1997, soon after the float of that the baht. Although the pre-crisis current account deficit was modest at around 3 percent of GDP, export Growth remained high and the fiscal balance stayed in surplus. Indonesia’s short-term private sector external debt grew rapidly, however, and evidence of weaknesses in the financial sector raised doubts about the government’s ability to defend the currency. Following the float of the rupiah in mid-August 1997, the exchange rate was initially relatively stable, albeit volatile, but began to fall sharply in October—the cumulative depreciation soon became the largest in the region (about 30 percent from July 1997).

    In early November 1997, the IMF approved a three-year Stand-By Arrangement equivalent to $10 billion (490 percent of quota) and additional financing commitments and pledges totaled $26 billion. The key objectives of the adjustment program were to restore market confidence, bring about an orderly adjustment in the current account, limit the decline in output Growth, and contain the inflationary impact of exchange rate depreciation. The initial response to the program was positive, market confidence improved, and the rupiah strengthened.

    Contagion from other emerging markets soon reemerged and the exchange rate fell precipitously during December 1997-January 1998, notwithstanding massive interventions that lowered gross official reserves to less than four months of imports. The downfall of the rupiah against the U.S. dollar from Rp2,400 in the pre-crisis period to Rpl4,000–15,000 led to a collapse in corporate balance sheets and resulting sharp economic contraction. The increase in the cost of living and poverty led to widespread social and political unrest. Key factors contributing to the macroeconomic deterioration included stop-and-go monetary policy, swinging between support for the exchange rate and strong liquidity expansion in the face of financial sector difficulties and runs on deposits. Implementation of important structural measures was uneven and market sentiment worsened ahead of the presidential election in mid-1998.

    A strengthened program was announced in early January 1998, but markets remained skeptical. In addition to a commitment to tight monetary policy and an extensive package of structural reforms, the program stressed a comprehensive bank restructuring plan, which did not move quickly enough to address the problems of corporate debt, however. Implementation of structural reforms continued to lag, and the macroeconomic program quickly ran off track, with base money growing rapidly, fueled by the liquidity support for failing financial institutions. Program implementation was further sidetracked by futile discussions about the introduction of a currency board and preparations for the March presidential election. The economic downturn deepened and the economy hovered on the verge of a vicious circle of currency depreciation and hyperinflation.

    Following the re-election of President Suharto and formation of a new government, the first review was completed on May 4, 1998 on the basis of another substantial modification of the program. After a promising start, including successful talks with private creditors regarding the restructuring of corporate sector obligations and the rollover of short-term bank debt, the program was cast off track by severe civil unrest, which led to the resignation of President Suharto on May 21, 1998. The rupiah nose-dived and hit an all-time low of Rpl6,650 against the U.S. dollar in mid-June 1998 (a cumulative depreciation of 85 percent since June 1997). Although some progress was made on restructuring of interbank debt, restoration of a trade facility, and creation of a framework for the voluntary restructuring of corporate debt involving a government exchange guarantee scheme (INDRA scheme), the economy experienced a severe slump (real GDP fell by a staggering 14 percent in 1998).

    In view of the deep-seated nature of Indonesia’s structural and balance of payments problems, the IMF’s Executive Board in late August 1998 approved the authorities’ request to replace the Stand-By Arrangement with an Extended Arrangement with the same access ($6.3 billion, or 312 percent of quota, for the remaining 26 months) and phasing as envisaged under the Stand-By Arrangement. The program succeeded in improving market sentiment, the domestic currency appreciated to about Rp7,500 against the U.S. dollar, and, as a result, inflation declined sharply. Unlike in other cases of capital account crises, however, a V-shaped recovery in economic activity did not take place and the economy declined by a further 1 percent in 1999. Although macroeconomic policies remained broadly on track in 1999, continued political and social unrest weighed on economic policies. Implementation of much needed structural policies remained poor and the slow progress in bank and corporate restructuring was particularly damaging to the economy.

    The EFF was cancelled in early 2000 and, at the same time, a new EFF for the period through December 2002 was approved, following the election of President Wahid and formation of his government. The arrangement provided access equivalent to some $5 billion and focused on macroeconomic stabilization, bank and corporate restructuring, rebuilding of public institutions, and improvement in natural resource management. Progress under this arrangement has been mixed, especially in the structural area. Although the economy bottomed out in 1999 and grew modestly—by 2.5 percent—in 2000, the recovery remained fragile. The continued social unrest hampered economic policies and, to date, Indonesia has been unable to benefit from a sustained rebound in economic activity.

    Korea (1997–98)

    Korea’s external position became vulnerable during the 1990s as domestic banks borrowed offshore to finance domestic firms. Although contagion from Thailand and Indonesia, and a lack of promptness in responding to market pressures, brought Korea to the brink of default, the authorities soon started to act decisively by implementing their program and bringing private lenders to the negotiating table. In the end, the ensuing crisis was relatively mild.

    Korea initially appeared little affected by the crisis in the region, with the exchange rate remaining essentially stable through October 1997. With a high level of short-term debt in relation to international reserves, however, the economy was vulnerable to a shift in market sentiment. Although macroeconomic fundamentals were favorable, concerns about the soundness of financial institutions and chaebol (industrial conglomerates) had intensified in early 1997. As Korean banks began to face difficulties rolling over their short-term foreign liabilities, in mid-1997 the Bank of Korea found itself lending its dwindling foreign exchange reserves to the banks’ offshore branches and the government announced a guarantee of foreign borrowing by Korean banks. External financing conditions deteriorated further in late October 1997 and the won fell sharply. Monetary policy was tightened briefly, but was soon relaxed in light of concerns about the impact of higher interest rates on the highly leveraged corporate sector. By early December 1997, the won had depreciated by over 20 percent against the U.S. dollar and usable foreign exchange reserves had declined to $6 billion (from $22.5 billion at the end of October 1997).

    On December 4, 1997, the IMF’s Executive Board approved a three-year Stand-By Arrangement with Korea, equivalent to $21 billion (1,939 percent of quota), with additional financing totaling $37 billion. To establish conditions for an early return of market confidence, the underlying program aimed to bring about an orderly reduction in the current account deficit, build up foreign exchange reserves, and contain inflation through a tightening of monetary policy and some fiscal measures. In addition, the program included a range of structural reforms in the financial and corporate sectors to address the root causes of the crisis. Upon approval of the program, Korea was able to draw $5.5 billion from the IMF.

    The positive impact of the announcement of the program was short-lived and the won dropped sharply. Confidence was undermined by doubts about the commitment to the program as the leading candidates for the mid-December presidential election hesitated to publicly endorse it. Moreover, with new information becoming available about the state of financial institutions, the level of usable reserves, and short-term obligations falling due, markets became concerned about a widening financing gap.

    With the won in free-fall and a few days of reserves left, a temporary agreement was reached with private bank creditors on December 24, 1997 to maintain their exposure to Korea. In addition, Korea requested a rephasing of purchases under the Stand By Arrangement to permit an advancement of drawings. At the same time, the program accelerated financial sector restructuring to facilitate capital inflows into the domestic stock and bond market. Interest rates had been raised significantly, and conditions for the provision of foreign currency liquidity support to banks had been tightened.

    In January 1998, early signs of stabilization emerged. Rollover rates increased significantly after the agreement with the banks on a voluntary rescheduling of short-term debt (equivalent to some $22 billion), usable international reserves stabilized, and the won appreciated moderately against the U.S. dollar. The current account had moved into surplus, but owing to the large depreciation of the exchange rate, inflation started to rise. In February, against the background of contracting domestic demand, program targets were adjusted to accommodate fiscal stimulus: the fiscal target for 1998 was lowered from a surplus of 0.2 percent of GDP in the original program to a deficit of 0.8 percent of GDP Monetary policy was expected to remain tight as long as the exchange market situation continued to be fragile. The program’s comprehensive structural reform agenda was expanded further to include commitments in financial sector restructuring, capital account and trade liberalization, the social safety net, labor market flexibility, and corporate restructuring and governance.

    The program remained on track and market confidence in the new government’s commitment strengthened. By July 1998, Korea had made substantial progress in overcoming its external crisis. A global sovereign bond issue was launched successfully, the central bank registered significant capital inflows into the domestic stock and bond markets, and usable reserves exceeded $30 billion. The sharp decline in economic activity, however, was weighing heavily on corporations, necessitating an acceleration of structural reforms. Interest rates had been lowered somewhat, but monetary policy continued to focus on maintaining exchange market stability. In view of the weaker outlook for Growth, the program incorporated an additional fiscal stimulus.

    Macroeconomic policies were further eased in the second half of 1998 to mitigate the severity of the recession: interest rates declined to pre-crisis levels, and a supplementary budget was prepared to support economic activity and strengthen the social safety net. Although output declined by 6 percent in 1998, exports recovered, and the current account surplus exceeded 12 percent of GDP.

    In 1999, the macroeconomic situation was fully stabilized and Korea experienced a V-shaped recovery (although there remained serious structural issues in the financial corporate sectors that remained to be tackled). With GDP growing by more than 6 percent amid price stability, the won appreciated by 13 percent, while the current account stayed in a healthy surplus of 6 percent of GDP. Although Korea stopped drawing from the IMF at mid-1999, it continued with scheduled quarterly reviews. Of the total committed amount at the onset of the crisis, only about one-half was actually disbursed, almost all of it from international financial institutions.

    Mexico (1994–95)

    The Mexican financial crisis of 1994–95—the first “crisis of the 21st century”—plays a central role in any analysis of crises driven by capital account reversals as it bears many similarities to subsequent events in Asia and Brazil.

    The Mexican crisis erupted after several years of impressive macroeconomic performance and an economic boom made possible by large capital inflows. These inflows, which amounted to about $95 billion during 1990–94, were facilitated by a return of confidence following the “lost decade” of the 1980s. Similar to the Asian episode, Mexican capital inflows were short-term and unhedged in the context of a fixed exchange rate regime which had led to an appreciating real exchange rate and a growing external current account deficit. Mexico’s vulnerabilities became a full-blown crisis when the confluence of changes in external conditions and a succession of domestic political shocks led to a loss of market confidence in the authorities’ ability to defend the exchange rate. On the external front, U.S. monetary policy was gradually tightened over the course of 1994. Internally, Mexico suffered from a series of adverse political shocks that included a revolt in the southern state of Ciapas and political assassination.

    Capital outflows and exchange rate pressures first surfaced in March–April 1994. The Mexican authorities responded by doubling interest rates from 9 to 18 percent and using $11 billion worth of reserves to defend the peso. However, the extent of intervention was kept secret. Moreover, the government decided to change the composition of its debt by issuing dollar-linked securities (tesobonos) to replace maturing peso-denominated securities (cetes) in an effort to reduce its direct funding costs. Although interest rates were some 6 percentage points lower than cetes rates, tesobonos offered an exchange rate guarantee. These actions were successful in calming market conditions in the period to August.

    Market turbulence resumed in October–November after yet another assassination, prompting a decline in the stock market and a resumption of out-flows. Monetary policy was tightened, but this did not prove sufficient to stem the loss of reserves (about $4 billion in the last two weeks of November). The peso came under renewed pressure in mid-December and it was finally allowed to float on December 22 after reserves had dipped to a low of $6 billion as compared to $25 billion in early November. The immediate problem following the float of the peso was how to refinance $29 billion of tesobonos and $18 billion in short-term bank credit lines, with the amortization schedule particularly heavy during the first quarter.

    Prior to the approval of the program, an $18.5 billion official line of credit was committed on January 2, 1995. Markets viewed the size of the package as insufficient, however, while the Mexican government was perceived as procrastinating. With the crisis deepening, a larger package of $40 billion in loan guarantees was proposed on January 12. Finally, on January 31, with Mexico on the brink of default, the U.S. administration proposed a package of more than $50 billion from the U.S. Treasury, the IMF, the Bank for International Settlements, and private institutions.

    The IMF-supported program provided for financing of SDR5.3 billion (300 percent of quota), to be disbursed over an 18-month period. In light of the unrelenting crisis, the financing was increased to SDR12.1 billion (688 percent of quota) on February 1, 1995. The program aimed to bring about a gradual real depreciation of the peso (by 14 percent over the life of the program) which was expected to cut the current account deficit in half (to 4 percent of GDP) by triggering a modest (7 percent) decline in imports and a much larger (25 percent) increase in exports of manufactures.

    With the program failing to stem capital outflows, the crisis continued unabated. Between January and March, the exchange rate depreciated sharply, interest rates rose substantially, and the banking system came under severe pressure. Financial markets doubted the realism of the program and the adequacy of its financing. The program had to be revised in March 1995 through a tightening of monetary conditions, a further consolidation of public finances (by the equivalent of 1.7 percent of GDP), and by a comprehensive set of policies to deal with the banking crisis. The revised program was ultimately successful in stemming capital outflows, stabilizing the peso, and amortizing the entire stock of tesobonos.

    The financial crisis led to a sharper-than-expected contraction in real GDP of about 7 percent in 1995, while end-of-period consumer price inflation amounted to 52 percent. The tightening of financial policies and a sizable real depreciation of the peso led to a large turnaround in the balance of payments: the current account shifted to approximate balance, fueled by a 32 percent increase in exports (in dollar terms) and a 21 percent decline in imports. The 1995 fiscal targets were met with significant margins, with the overall budget achieving a surplus of about 1 percent of GDP. Implementation of policies agreed in the context of the program led to a consolidation of financial market stability in 1996, quick recovery, and restoration of low inflation.

    The 1995 downturn combined with high domestic interest rates led to a banking crisis. The government strategy of dealing with the crisis emphasized burden-sharing between bank shareholders and the government, and the spreading of the fiscal costs of bank restructuring over time. Immediate steps taken in January-June 1995 included (1) a government commitment to protect all depositors; (2) provision of peso liquidity by the Bank of Mexico to the banking system; (3) a temporary bank recapitalization through the PROCARTE program; (4) a program of dollar loans to help banks meet their external short-term obligations through FOBAPROA, the deposit protection agency administered by the Bank of Mexico; and (5) a strengthening of banking supervision and relaxation of limits on foreign participation in Mexican banks. While these measures contained the crisis and prevented the collapse of several banks, the banking system remained fragile in 1995—96.

    The extent of the Mexican crisis caught the financial markets by surprise and prompted a lot of soul searching as to the reasons behind the failure to detect the early warning signs. Three reasons were identified. First, weak financial systems were unable to intermediate efficiently large capital inflows into productive investments, creating firm-level vulnerabilities. Second, the Mexican government procrastinated for several months in 1994, in effect undermining confidence in its policies. Third, the authorities’ lack of transparency, especially with regard to international reserves in the early days of the crisis, fueled negative market sentiment.

    The Philippines (1997)

    When the 1997 crisis struck, the Philippines had a long-running IMF program already in place.113 The authorities acted decisively, the program was quickly amended, and, in the end, the country weathered the crisis better than expected.

    The Philippines’ better-than-expected performance can be explained as a combination of a comparatively smaller asset bubble and a faster and more decisive reaction by the authorities. First, in their macroeconomic performance in the 1970s and 1980s, the Philippines lagged behind their neighbors and only after reforms in the early 1990s did the economy begin to attract substantial foreign capital. Consequently, signs of overheating and asset bubbles began to show much later than in other countries and the resulting financial imbalances were correspondingly smaller. Second, at the time of the crisis, an IMF arrangement was already in place and, when the crisis struck, the authorities acted without delay. Moreover, the consensus for sound economic policies survived the May-July 1998 transition from one presidential administration to another.

    At the onset of the crisis, the Philippines presented a somewhat different pattern of strengths and vulnerabilities than the “tigers” most heavily affected by the crisis. Current account deficits averaged 4–5 percent during pre-crisis years and private sector credit grew 50 percent in 1996. By regional standards, however, the pre-crisis Growth rates of real GNP were more modest than in other countries and external exposure by domestic private corporations, including short-term debt, was relatively small. Moreover, levels of corporate leverage were significantly lower, major banks were well capitalized, and the reforms of the past 10 years or so had created a reasonably open, market-oriented economy. In summary, the structural vulnerabilities were much smaller and their roots—unlike in other countries—were known and some of them had already been addressed in previous IMF programs.

    The initial impact of the Asian crisis—in terms of the initial drop in share prices, currency depreciation, and loss in international reserves—was comparable to that of the Philippines’ neighbors. On the other hand, the ensuing recession was relatively mild and most of the macroeconomic and financial indicators rebounded faster than in the Philippines’ neighbors. This may have reflected, in large part, the lesser degree of vulnerability of the Philippine economy. In particular, the rebound was associated with more favorable export performance than in neighboring countries: unlike in other Asian crisis countries, where monthly U.S. dollar export Growth rates were negative for most of 1998, the Philippines recorded an uninterrupted period of double-digit Growth rates. The high level of exports throughout the crisis likely reflected the stability of firms’ balance sheets vis-à-vis exchange rate fluctuations. Of course, as a result of the massive depreciation of the peso, imports fell between 1997 and 1998 by some 8 percent of GNP, forcing a trade balance adjustment of 13 percent of GNP.

    As in other countries, capital outflows and sharp falls in the stock market led to mounting pressures on the peso. The authorities initially tightened monetary policy by raising interbank interest rates from about 15 percent to reach peaks of 40–60 percent (for a few days in late August 1997) and intervened in the foreign exchange market to maintain the de facto peg of the peso. However, the system lost its viability with the flotation of the Thai baht and the peso was accordingly floated in July 1997. The new exchange rate arrangement was accompanied by strengthened fiscal, monetary, and structural policies in the context of the existing IMF program. The thrust of the program was on improved fiscal performance and the related long-standing revenue measure: the Comprehensive Tax Reform Package.

    After initial monetary and fiscal tightening, as the peso stabilized, 114 the stance became neutral and eventually shifted toward supporting the emerging recovery. Interest rates were brought down during the second half of 1998 and monetary policy was eased significantly in early 1999, after firm turnarounds in the balance of payments and inflation were established. Fiscal policy followed a similar path: from a pre-crisis surplus target of 1 percent of GNP for 1998, the program was revised to an eventual deficit target of 3 percent of GNP.

    Financial markets remained volatile until the last quarter of 1998, as a result of both external developments (short-term outflows amounted to $3 billion, a reversal of some 7 percent of GNP, and the net position on medium-term loans worsened by another $2.5 billion) and domestic uncertainties (including political uncertainties associated with a new administration that took office in early 1998). Since September 1998, however, financial markets strengthened continually, with share prices up by more than 100 percent above their trough, the peso appreciating by some 20 percent in real effective terms, and official reserves rising well above their pre-crisis level. By mid-1999, the economic slowdown appeared to be over, with industrial production showing significant Growth. After recording zero Growth in 1998, GNP grew by more than 3½ percent in 1999.

    Thailand (1997–98)

    Contagion started to spread from Thailand in 1997, where a period of fast Growth led to large current account deficits. In addition, chaotic deregulation in the early 1990s created an unsupervised and vulnerable financial sector and contributed to the emergence of asset bubbles, especially in the real estate sector.

    Pressures on the baht started in late 1996 and built up in early 1997 against the background of an unsustainable current account deficit, significant real appreciation of the domestic currency, rising short-term foreign debt, and growing problems in the financial sector. Reserve money Growth accelerated sharply as the Bank of Thailand provided liquidity support for ailing financial institutions. The policy response to the pressures in the foreign exchange market focused on intervention, the introduction of capital account controls, and minor fiscal tightening.

    The baht was floated on July 2, 1997, following mounting speculative attacks and amid concerns about the reserve position. The accompanying policy response was inadequate and failed to bolster market confidence. The baht depreciated by 20 percent against the U.S. dollar during July, partly because short-term interest rates were allowed to decline sharply after a temporary increase.

    On August 20, 1997, the IMF’s Executive Board approved a three-year Stand-By Arrangement with Thailand, equivalent to $4 billion (505 percent of quota), with additional financing totaling $13 billion. The underlying adjustment program was aimed at restoring confidence, bringing about an orderly reduction in the current account deficit, while maintaining positive Growth rates, reconstituting foreign exchange reserves, and limiting the rise in inflation to the one-off effects of the depreciation. Key elements of the program included measures to restructure the financial sector (including closure of insolvent financial institutions); fiscal adjustment equivalent to some 3 percent of GDP; and control of domestic credit, with indicative ranges for interest rates. Upon approval of the program, Thailand drew more than $5 billion from the IMF and other sources.

    In subsequent months, the baht continued to depreciate as the country was unable to roll over its short-term debt. While macroeconomic policies were in line with the program and nominal interest rates were raised, market confidence was adversely affected by delays in the implementation of financial sector reforms, political uncertainty, and initial difficulties in communicating key aspects of the program to the public. It also became clear that the slowdown in economic activity was more pronounced than anticipated. Against this background, a new government took office in November 1997 and soon thereafter the program was strengthened and additional fiscal measures were introduced to achieve the original fiscal target. Reserve money and net domestic assets of the Bank of Thailand were to be kept below the original program limits, the indicative range for interest rates was raised, and a specific timetable for financial sector restructuring was announced.

    The crisis in Thailand started to abate in early 1998, in tandem with other countries in the region. After falling to an all-time tow against the U.S. dollar in early January 1998, the baht began to strengthen in early February as market confidence revived. Contracting domestic demand helped to keep inflation in check and contributed to a larger-than-expected adjustment in the current account and Growth deceleration. In view of stabilizing exchange market conditions and the changed economic outlook, the program was revised significantly in March 1998. Monetary policy focused on the exchange rate, while fiscal policy shifted to a more accommodating stance. In addition, the program included measures to strengthen the social safety net, and broadened the scope of structural reforms to strengthen the core banking system and promote corporate restructuring.

    By mid-1998, the domestic currency appreciated some 35 percent vis-à-vis the U.S. dollar from its low in January and foreign exchange reserves increased. Nevertheless, the economy sank into a deep recession. To stimulate demand, the fiscal deficit target for 1997–98 was increased from 2 percent to 3 percent of GDP. In addition, the social safety net was strengthened and the program for financial sector and corporate restructuring was further specified. However, the financial sector began to show signs of strain, economic activity slowed down again, and exports failed to pick up. The large adjustment in the current account reflected mostly a compression of imports. Restructuring of financial institutions was complicated by corporate sector problems.

    At end-1998, the policy framework was refocused on supporting the nascent recovery without sacrificing stabilization gains. Foreign exchange market conditions remained relatively stable—in spite of the Russian crisis—providing room for a further lowering of interest rates, while the fiscal position remained expansionary. The program for financial and corporate sector restructuring was broadened significantly, and the structural reform agenda in other areas (privatization, foreign ownership, and social safety net) was strengthened. The implementation of structural measures remained inconsistent, however.

    In 1999, the Thai economy started to grow again (by about 4 percent), mostly fueled by buoyant exports, but the recovery was weak and less pronounced than in neighboring countries. The current account balance remained highly positive, despite a fiscal deficit of 3 percent. The rate of Growth stayed comparatively low in 2000, partly owing to sluggish domestic demand. Thai financial markets underperformed compared to their peers as foreign investors started to pull out of the country because of unresolved issues related to financial and corporate sector restructuring.

    Thailand was hard hit by the 1997–98 crisis—only Indonesia experienced a larger decline in real GDP—and the subsequent recovery was milder than in other countries. More important, investor confidence remained low, pending the resolution of structural vulnerabilities.

    Turkey (1994)

    The Turkish 1994 crisis foreshadowed several of the subsequent capital account crises with a massive swing in capital outflows of some 8 percent of GNP. The IMF-supported program was successful in the sense that it quickly restored confidence, even though the national authorities procrastinated on longer-term structural measures.

    Following the liberalization of the economy during the 1980s. Turkey enjoyed high, albeit very variable, real GNP Growth rates (Table A5.1). At the same time, macroeconomic imbalances became increasingly pronounced as a result of expansionary public sector policies, especially after 1988. These imbalances resulted in persistently high inflation—in the range of 60–70 percent per year—and a weakening external position.

    Table A5.1.Macroeconomic Indicators in Capital Account Crisis Programs
    1. Argentina (1995)1992199319941995199619971998
    Real GDP (Growth, percent per year)−
    Consumer Price Index (Growth, percent per year)24.910.
    Real exchange rate (Growth, percent per year)13.411.7−0.4−5.0−0.3563.1
    Private Savings (as percent of GDP)13.415.817.517.917.215.715.6
    Public Savings (as percent of GDP)−1.6−1.5−0.4−0.5
    Investment (as percent of GDP)16.719.119.917.918.119.419.9
    Current Account (as percent of GDP)−2.8−3.4−4.3−1.9−2.4−4.1−4.8
    Capital and financial account (as percent of GDP)
    Net private capital flows (as percent of GDP)4.5−
    General government balance (as percent of GDP)0.4−0.2−1.8−2.3−3.2−2.1−2.1
    External debt (as percent of GDP)27.430.533.338.140.342.647.3
    2. Brazil (1998)1995199619971998199920002001
    Real GDP (Growth, percent per year)
    Consumer Price Index (Growth, percent per year)
    Real exchange rate (Growth, percent per year)−2.3−33.69.1
    Private Savings (as percent of GDP)22.921.421.021.824.119.5
    Public Savings (as percent of GDP)−3.2−3.4−3.4−4.9−8.3−3.1
    Investment (as percent of GDP)22.320.921.521.220.420.5
    Current Account (as percent of GDP)−2.6−3.0−3.8−4.3−4.7−4.2
    Capital and financial account (as percent of GDP) 1
    Net private capital flows (as percent of GDP)
    General government balance (as percent of GDP)−7.0−5.9−6.1−7.9−10.0−4.6
    External debt (as percent of GDP)22.623.224.830.745.639.7
    3. Indonesia (1997)1994199519961997199819992000
    Real GDP (Growth, percent per year)−
    Consumer Price Index (Growth, percent per year)
    Real exchange rate (Growth, percent per year)−0.7−3.45.1−5.6−51.645.2−2.0
    Private Savings (as percent of GDP)21.622.021.919.410.811.014.7
    Public Savings (as percent of GDP)
    Investment (as percent of GDP)31.131.932.131.816.812.217.9
    Current Account (as percent of GDP)−1.7−3.3−3.2−
    Capital and financial account (as percent of GDP)−4.2−4.1−4.2
    Net private capital flows (as percent of GDP)−3.0−3.6−0.9
    General government balance (as percent of GDP)−1.1−2.3−1.5−3.1
    External debt (as percent of GDP)57.056.353.463.3148.4112.196.5
    4. Korea (1997)1994199519961997199819992000
    Real GDP (Growth, percent per year)−6.710.98.8
    Consumer Price Index (Growth, percent per year)
    Real exchange rate (Growth, percent per year)−6.0−25.613.58.1
    Private Savings (as percent of GDP)25.425.823.422.423.022.320.9
    Public Savings (as percent of GDP)10.29.710.
    Investment (as percent of GDP)36.537.237.934.221.226.728.7
    Current Account (as percent of GDP)−1.0−1.7−4.4−1.712.76.02.4
    Capital and financial account (as percent of GDP)−10.75.2−2.7
    Net private capital flows (as percent of GDP)−4.6−
    General government balance (as percent of GDP)−0.9−3.8−2.72.5
    External debt (as percent of GDP)18.017.522.228.143.5
    5. Mexico (1995)1992199319941995199619971998
    Real GDP (Growth, percent per year)−
    Consumer Price Index (Growth, percent per year)
    Real exchange rate (Growth, percent per year)8.27.6−3.6−
    Private Savings (as percent of GDP)
    Public Savings (as percent of GDP)
    Investment (as percent of GDP)23.321.021.719.823.125.924.3
    Current Account (as percent of GDP)−6.7−5.8−7.0−0.6−0.7−1.9−3.8
    Capital and financial account (as percent of GDP)
    Net private capital flows (as percent of GDP)
    General government balance (percent of GDP)1.50.7−0.2−0.20.3−1.0−1.3
    External debt (as percent of GDP)31.332.733.859.049.638.238.4
    6. Philippines (1998)1995199619971998199920002001
    Real GDP (Growth, percent per year)−
    Consumer Price Index (Growth, percent per year)
    Real exchange race (Growth, percent per year)2.79.3−0.5−18.48.6−6.8
    Private Savings (as percent of GDP)14.415.515.420.927.029.0
    Public Savings (as percent of GDP)−0.7
    Investment (as percent of GDP)22.524.024.820.218.617.6
    Current Account (as percent of GDP)−2.7−4.8−5.32.410.012.4
    Capital and financial account (percent of GDP)2.18.411.7−5.6−6.7−7.7
    Net private capital flows (as percent of GDP)−4.0−1.2−7.7
    General government balance (as percent of GDP)−1.4−0.4−0.8−2.7−4.3−4.7
    External debt (as percent of GDP)54.955.061.681.775.775.7
    7. Thailand (1997)1994199519961997199819992000
    Real GDP (Growth, percent per year)−1.4−
    Consumer Price Index (Growth, percent per year)
    Real exchange rate (Growth, percent per year)0.1−1.86.8−7.0−
    Private Savings (as percent of GDP)22.521.020.620.326.124.123.7
    Public Savings (as percent of GDP)12.112.813.
    Investment (as percent of GDP)40.241.841.633.320.319.922.4
    Current Account (as percent of GDP)−5.4−7.8−7.9−2.112.810.27.6
    Capital and financial account (as percent of GDP)−13.6−14.5−8.2
    Net private capital flows (as percent of GDP)−5.3−16.3−13.5−10.7
    General government balance (as percent of GDP)−0.9−2.6−2.9−2.3
    External debt (as percent of GDP)51.659.959.672.393.978.465.8
    8. Turkey (1994)1991199219931994199519961997
    Real GDP (Growth, percent per year)−
    Consumer Price Index (Growth, percent per year)
    Real exchange rate (Growth, percent per year)2.6−3.78.9−
    Private Savings (as percent of GDP)19.622.124.223.821.821.620.4
    Public Savings (as percent of GDP)0.7−0.8−2.7−1.1−0.1−1.80.8
    Investment (as percent of GDP)22.423.226.321.324.924.525.1
    Current Account (as percent of GDP)0.2−0.6−3.52.8−0.5−1.4−1.4
    Capital and financial account (as percent of GDP)−−3.6−
    Net private capital flows (as percent of GDP)−−
    General government balance (as percent of GDP)−6.9−6.7−9.1−6.0−5.8−10.3−9.4
    External debt (as percent of GDP)33.034.836.950.
    9. Russia (l998)1995199619971998199920002001
    Real GDP (Growth, percent per year)−4.2−3.40.9−
    Consumer Price Index (Growth, percent per year)197.447.614.727.785.720.8
    Real exchange rate (Growth, percent per year)9.722.15.6−11.4−29.312.1
    Private Savings (as percent of GDP)33.133.728.918.528.334.8
    Public Savings (as percent of GDP)−6.3−8.2−5.9−3.4−0.91.6
    Investment (as percent of GDP)25.424.623.115.715.118.0
    Current Account (as percent of GDP)1.40.9−0.1−0.612.418.4
    Capital and financial account (as percent of GDP)−8.3−13.4
    Net private capital flows (as percent of GDP)7.30.4−5.3−4.0−5.8−6.9
    Central government balance (as percent of GDP)−5.8−6.5−6.7−4.9−1.42.6
    General government balance (as percent of GDP)−6.1−8.9−7.5−7.0−0.24.5
    Primary balance (as percent of GDP)−2.5−3.0−2.8−
    External debt (as percent of GDP)37.932.530.954.879.961.5
    10. Malaysia (1997)1994199519961997199819992000
    Real GDP (Growth, percent per year)9.29.810.07.3−
    Consumer Price Index Growth, percent per year)
    Real exchange rate (Growth, percent per year)−−2.4−
    Private Savings (as percent of GDP)16.817.519.918.824.724.224.3
    Public Savings (as percent of GDP)17.216.416.118.114.914.113.7
    Investment (as percent of GDP)41.643.641.543.026.622.327.0
    Current Account (as percent of GDP)−7.6−9.7−4.4−5.913.115.99.7
    Capital and financial account (as percent of GDP)5.910.66.98.2−9.7−14.3−6.7
    Net private capital flows (as percent of GDP)−0.3−−0.3−0.32.2
    General government balance (as percent of GDP)−0.4−3.70.4
    External debt (as percent of GDP)19.918.015.616.825.1

    Capital inflows and outflows prior to 1993 had been modest (typically less than 1 percent of GNP) and total external debt at end-1992 stood at 35 percent of GNP (about three-quarters of which represented public debt).

    Real GNP continued to grow rapidly in 1993, as strong increases in both consumption and investment were fueled by a loosening of monetary policy and by an increase in the fiscal deficit. The overall public sector deficit rose to 13 percent of GNP (up from 12 percent in 1992), while the central government deficit grew from 5.4 percent to 6.3 percent of GNP. This widening deficit primarily reflected higher interest payments, with the primary deficit actually improving from 7 percent of GNP to 5.6 percent of GNP in 1993.115

    Against the backdrop of growing resource demands by the public sector, the current account deteriorated from a deficit of about 1 percent of GNP in 1992 to more than 5 percent of GNP in 1993. The associated relaxation of monetary policy reflected the authorities’ attempt to reduce the interest cost of public debt, with the ex post real interest rate on treasury bills falling from 20 percent a year in the first quarter of 1993 to about 9 percent for the rest of the year.

    At end-1993, foreign exchange reserves stood at some $8 billion or about 2½ months worth of imports, while total external debt stood at $67 billion (37 percent of GNP). Three-quarters of this was medium- and long-term debt, and most of the medium- and long-term debt was in the form of public sector debt. The exchange crisis broke in January 1994, and through the period January–March 1993 (when the central bank stopped intervening), the Turkish lira depreciated some 60 percent and the central bank lost $3 billion of its foreign exchange reserves. The capital account, which had registered a net inflow of $8.7 billion in 1993, recorded a net outflow of $4.2 billion in 1994. Much of this occurred through the banking system, which initially had been borrowing heavily from abroad to take advantage of high domestic interest rates; short-term flows to domestic banks swung from an inflow of $4 billion in 1993 to an outflow of $7 billion in 1994: a swing of more than 7 percent of GNP.

    The Treasury’s continued reluctance to sell securities at market rates paralyzed the primary and secondary bond markets and led to heavy borrowing from the central bank. In turn, this placed mounting strains on the interbank market, with interest rates reaching well into the four-digit levels. In the end, the banking system proved vulnerable to the exchange rate depreciation (even the generous net foreign exchange position limits were routinely flouted), the slowdown in economic activity, and the extremely high interbank interest rates. As a result, three small banks and a number of brokerage houses collapsed in April 1994. In May, the authorities narrowly avoided large-scale bank runs by extending the government’s guarantee to cover 100 percent of all domestic and foreign currency deposits.

    The program, approved in July 1994, centered on frontloaded fiscal adjustment—amounting to some 9 percent of GNP—while the low level of reserves and turbulence in the financial markets precluded an exchange rate anchor. The Treasury eschewed further monetary financing, and successfully reentered the bond market in June 1994. The macroeconomic adjustment was to be followed by far-reaching structural reforms, particularly in the banking, social security, and agricultural sectors, as well as privatization, so as to lower the level of public debt and the strain on public finances. The program foresaw a swing in the current account of about 4 percent of GNP, all of which could be accounted for by an improvement in the public sector savings-investment balance.

    The program was successful in that the external situation was stabilized. The current account swung into a surplus of 3 percent of GNP in 1995 (at the expense of a GNP decline of some 5 percent against a program expectation of a 1½ percent decline), and the short-term capital outflows were reversed. In the short term, public finances also improved, with the overall deficit falling from 13 percent of GNP in 1993 to 10 percent of GNP in 1994, and 6.4 percent of GNP in 1995. Many of the structural reforms, however, were not undertaken (particularly privatization and the social security reform) which would, in later years, again place unsustainable demands on public sector finances.116

    Taken as a whole, the Turkish 1994 crisis and program—which predated the Mexican crisis by several months—combined elements of traditional balance of payments crises with features of the later capital account crises. On the one hand, there were clear macroeconomic imbalances prior to the currency crisis. On the other hand, the massive swings in capital flows—amounting to some 8 percent of GNP—were a presage of crises to come.

    Table A5.2.Balance of Payment Developments in Selected Asian Countries

    (Quarterly flows, as a percent of quarterly)1

    Current Account−−3.5−−8.3−1.87.715.010.0
    Financial Account2.5−5.1−3.6−5.70.1−13.4−11.3−−7.811.03.2−6.1−10.4−7.1
    Excluding IMF and Reserve Assets−15.6−26.20.9−9.8−17.2−6.70.4−4.7−−8.2−4.6−6.3−13.4−9.5−10.8
    A. Direct Investment−−0.7−
    B. Portfolio Investment−9.9−−4.7−3.2−2.41.5−−0.2
    Equity securities−10.0−15.0−3.60.6−−0.2−−−0.5
    C. Order Investment0.5−9.1−4.22.6−8.2−−−5.4−9.9−11.5−15.1−17.2−19.1
    Trade credits0.−2.0−7.0−1.6−0.3−0.4−0.2−1.5−1.1−0.2
    Monetary authorities5.−
    of which: IMF5.−0.3−
    General government0.−1.20.5−1.7−
    of which: Short-term0.−9.7−3.7−9.1−2.3−1.3−0.7−0.20.0
    Other sectors−4.1−9.3−9.2−14.5−4.7−1.9−0.7−1.35.6−−2.1−2.2−5.2−6.3−2.3
    of which: Short-term0.−5.3−2.4−5.8−1.10.6−3.6−1.5−0.9−3.4−3.0−3.9−6.5−2.8
    Currency and deposits−0.7−−−5.4−5.6−13.2−9.6−11.0−2.8
    Other liabilities1.2−0.70.0−0.5−0.9−−0.70.8−0.60.7−0.5
    Other sectors1.0−0.70.0−−−0.3
    Net Errors and Omissions−−1.4−3.30.1−2.9−1.7−0.7−1.3−5.74.8−2.7−1.3−1.6−4.6−2.9
    Reserve Assets (-: increase)212.521.1−8.6−8.18.2−11.7−14.1−6.99.9−3.6−7.70.415.65.25.2−1.83.2
    Source: IMF, Balance of Payments Statistics Yearbook.

    In percent of equal quarters of annual GDP.

    Change in reserve assets equals –(current account +(financial account – reserve assets – IMF) + IMF + net errors and omissions + capital account (not shown)).

    Table A5.3.Balance of Payment Developments in Selected Latin American Countries

    (Quarterly flows, as a percent of quarterly GDP)1

    Current Account−4.4−4.10.3−1.6−4.6−5.8−4.1−5.3−7.5−7.1−1.90.5−3.5−
    Financial Account5.54.7−−1.54.911.7−10.0−13.8−1.3
    Excluding IMF and Reserve Assets7.8−4.3−0.13.6−8.42.7−−3.4−−7.5−7.3−0.6
    A. Direct Investment2.
    B. Portfolio Investment4.3−−3.82.0−−5.1−10.5−−0.2−0.90.3
    Equity securities1.−4.4−−
    C. Other Investment1.1−−9.3−1.8−4.02.9−1.8−−0.3−1.9−7.7−6.1−1.3
    Trade credits0.1−0.1−0.10.0−0.6−1.3−0.8−−1.3−1.9−0.91.7
    Monetary authorities−0.2−−0.16.4−0.3−0.213.5−−
    of which: IMF−0.2−−0.3−0.210.4−
    General government1.30.81.6−−−1.0−1.8−1.9−2.6−1.3
    of which: Short-term0.52.50.1−0.4−2.8−1.70.3−−0.23.3−7.3−6.2−2.8−3.8
    Other sectors0.−0.1−−−0.4
    of which: Short-term0.−−0.8−0.20.4−0.3−0.10.1−0.1−0.2
    Currency and deposits−0.3−−0.10.0−0.6−0.6−1.10.2−2.7−−
    Other liabilities−−1.5−
    Other sectors0.−1.5−
    Net errors and Omissions−1.1−0.61.2−−−2.71.0−1.4−
    Reserve assets (-: increase)2−2.09.2−3.7−0.512.70.47.9−5.7−0.19.5−1.0−−2.4−7.3−1.1
    Source:IMF, Balance of Payments Statistics Yearbook.

    Equal quarters of annual GDP.

    Change in reserve assets equals –(current account + (financial account – reserve assets – IMF) + IMF + net errors and omissions + capital account (not shown)).


      Agénor, Pierre-Richard, C. JohnMcDermott, and E. MuratUçer,1997, “Fiscal Imbalances, Capital Inflows, and the Real Exchange Rate: The Case of Turkey,European Economic Review, Vol. 41 (April), pp. 81925.

      Aghion, Philippe, PhilippeBacchetta, and AbhijitBanerjee,2000, “A Simple Model of Monetary Policy and Currency Crises,European Economic Review, Vol. 44 (May), pp. 72838.

      Alesina, Alberto, and AllanDrazen,1991, “Why Are Stabilizations Delayed?American Economic Review, Vol. 81 (December), pp. 117088.

      Basurto, Gabriela, and Atish R.Ghosh,2000, “The Interest Rate-Exchange Rate Nexus in the Asian Crisis Countries,Staff Papers, Special Issue—IMF Annual Research Conference, International Monetary Fund, Vol. 47, pp. 99120.

      Berg, Andrew,1999, “The Asia Crisis: Causes, Policy Responses, and Outcomes,IMF Working Paper 99/138 (Washington:International Monetary Fund).

      Berg, Andrew,CatherinePattillo,1999, “Are Currency Crises Predictable?: A Test,Staff Papers, International Monetary Fund, Vol. 46 (June), pp. 10738.

      Blanchard, Olivier J., and DannyQuah,1989, “The Dynamic Effects of Aggregate Demand and Supply Disturbances,American Economic Review, Vol. 79 (September), pp. 65573.

      Boorman, Jack, TimothyLane, MarianneSchulze-Ghattas, AlesBulir, Atish R.Ghosh, JavierHamann, AlexMourmouras, and StevenPhillips,2000, “Managing Financial Crisis: The Experience in East Asia,Carnegie–Rochester Conference Series on Public Policy, No. 53 (Amsterdam: North-Holland).

      Burnside, Craig, Martin S.Eichenbaum, and SergioRebelo,1999, “Prospective Deficits and the Asian Currency Crisis,Policy Research Working Paper No. 2174 (Washington: World Bank).

      Calvo, Guillermo A.,1996, “Varieties of Capital–Market Crises” (unpublished; College Park:University of Maryland). Available via the Internet:

      Calvo, Guillermo A.,1999, “Fixed versus Flexible Exchange Rates: Preliminaries of a Turn-of-Millennium Rematch” (unpublished; College Park:University of Maryland). Available via the Internet:

      Calvo, Guillermo A.,, and Enrique G.Mendoza,1996, “Mexico’s Balance–of–Payments Crisis: A Chronicle of a Death Foretold,Journal of International Economics, Vol. 41 (November), pp. 23564.

      Chalk, Nigel A., and RichardHemming,2000, “Assessing Fiscal Sustainability in Theory and Practice,IMF Working Paper 00/81 (Washington:International Monetary Fund).

      Chand, Sheetal K.,1992, “Fiscal Impulses and Their Fiscal Impact,IMF Working Paper 92/38 (Washington:International Monetary Fund).

      Chang, Roberto, and AndresVelasco,1998, “Financial Fragility and the Exchange Rate Regime,NBER Working Paper No. 6469 (Cambridge, Massachusetts:National Bureau of Economic Research).

      Chu, Ke-Young, and SanjeevGupta,1998, eds.,Social Safety Nets: Issues and Recent Experiences (Washington:International Monetary Fund).

      Claessens, Stijn, SimeonDjankov, and GiovanniFerri,1998, “Corporate Distress in East Asia: Assessing the Impact of Exchange Rate and Interest Rate Shocks” (unpublished; Washington:World Bank).

      Claessens, Stijn, SimeonDjankov, and TatianaNenova,2000, “Corporate Risk Around the World,Policy Research Working Paper No. 2271 (Washington:World Bank).

      Cline, William R.,2000, “The Role of the Private Sector in Resolving Financial Crises” (unpublished; Washington:Institute for International Finance).

      Corsetti, Giancarlo, PauloPesenti, and NourielRoubini,1999, “What Caused the Asian Currency and Financial Crisis? Part II: The Policy Debate,Japan and the World Economy, Vol. 11,No. 3, pp. 30573.

      Dell’Ariccia, Giovanni, IsabelGödde, and JerominZettelmeyer,2000, “Moral Hazard and International Crisis Lending: A Test,paper presented at the first IMF Research Conference,Washington,November.

      Dollar, David, and MaryHallward-Driemeier,2000, “Crisis, Adjustment, and Reform in Thailand’s Industrial Firms,World Bank Research Observer (International), Vol. 15 (February),No. 1, pp. 122.

      Domaç, Ilker, and GiovanniFerri,1998, “The Real Impact of Financial Shocks: Evidence from Korea,Policy Research Working Paper No. 2010 (Washington:World Bank).

      Dornbusch, Rudiger, and AlejandroWerner,1994, “Mexico: Stabilization, Reform, and No Growth,Brookings Papers on Economic Activity: 1, Brookings Institution, pp. 253315.

      Drazen, Allan, and VittorioGrilli,1993, “Benefits of Crises for Economic Reforms,American Economic Review, Vol. 83 (June), pp. 598607.

      Edwards, Sebastian,2000, “Contagion,The World Economy, Vol. 23 (July), pp. 873900.

      Eichengreen, Barry,1999, “Is Greater Private Sector Burden Sharing Impossible?Finance & Development, Vol. 36 (September), pp. 1619.

      Fane, George, and Ross H.McLeod,1999, “Lessons for Monetary and Banking Policies from the 1997–98 Economic Crises in Indonesia and Thailand,Journal of Asian Economics, Vol. 10 (Fall),No. 3, pp. 39513.

      Feldstein, Martin,1998, “Refocusing the IMF,Foreign Affairs, Vol. 77 (March–April), pp. 2033.

      Feldstein, Martin,1999, “A Self–Help Guide for Emerging Markets,Foreign Affairs, Vol. 78 (March–April), pp. 93109.

      Flood, Robert P., and Nancy P.Marion,1998, “Perspectives on Currency Crisis Literature,IMF Working Paper 98/130 (Washington:International Monetary Fund).

      Frankel, Jeffrey A.,1999, “The Balance Between Adjustment and Financing,paper presented at the International Monetary Fund seminar,Key Issues in Reform of the International Monetary System,Washington,May.

      Furman, Jason, and Joseph E.Stiglitz,1998, “Economic Crises: Evidence and Insights from East Asia,Brookings Papers on Economic Activity: 2, Brookings Institution, pp. 135.

      Ghosh, Swati R., and Atish R.Ghosh,2000, “East Asia in the Aftermath: Was There a Crunch?” inEuropean Monetary Union, Emerging Markets and Econometric Issues in International Finance,ed. byCornelius, Gottschling, and Kreuter (Frankfurt, Campus Verlag).

      Ghosh, Swati R., and Atish R.Ghosh,2001, “Structural Factors in Currency Crises” (unpublished; Washington:International Monetary Fund).

      Giavazzi, Francesco, and MarcoPagano,1990, “Can Severe Fiscal Contractions Be Expansionary? Tales of Two Small European Countries,NBER Working Paper No. 3372 (Cambridge, Massachusetts:National Bureau of Economic Research).

      Goldfajn, Ilan, and TaimurBaig,1998, “Monetary Policy in the Aftermath of Currency Crises: The Case of Asia,IMF Working Paper 98/170 (Washington:International Monetary Fund).

      Goldfajn, Ilan, and PoonamGupta,1999, “Does Monetary Policy Stabilize the Exchange Rate Following a Currency Crisis?IMF Working Paper 99/42 (Washington:International Monetary Fund).

      Goldstein, Morris, 2001, “IMF Structural Policy Conditionality: How Much Is Too Much?IIE Working Paper No. 01–4 (Washington:Institute for International Economics).

      Gould, David M., and Steven B.Kamin,2000, “The Impact of Monetary Policy on Exchange Rates During Financial Crises,U.S. International Finance Discussion Papers No. 669 (Washington:U.S. Board of Governors of the Federal Reserve System, International Finance Division).

      Guitián, Manuel, 1981, Fund Conditionality: Evolution of Principles and Practices, IMF Pamphlet Series, No. 38 (Washington:International Monetary Fund).

      Hutchison, Michael M., 2001, “A Cure Worse Than the Disease? Currency Crises and the Output Costs of IMF-Supported Stabilization Programs,NBER Working Paper No. 8305 (Cambridge, Massachusetts:National Bureau of Economic Research).

      Institute of International Finance,1999, Report of the Working Group on Financial Crises in Emerging Markets (Washington).

      International Monetary Fund,1999, Policy Development and Review Department,Involving the Private Sector in Forestalling and Resolving Financial Crises” (Washington).Available via the Internet:

      International Monetary Fund,2000a, Policy Development and Review Department,Review of Fund Facilities—Preliminary Considerations” (Washington).Available via the Internet:

      International Monetary Fund,2000b, Public Information Notice 00/37,Debt and Reserve-Related Indicators of External Vulnerability.Available via the Internet:

      International Monetary Fund,2001a, Policy Development and Review Department,Conditionality in Fund-Supported Programs—Overview” (Washington).Available via the Internet:

      International Monetary Fund,2001b, Policy Development and Review Department,Conditionality in Fund–Supported Programs—Policy Issues” (Washington).Available via the Internet:

      International Monetary Fund,, 2001c, Policy Development and Review Department,Structural Conditionality in Fund-Supported Programs” (Washington).Available via the Internet:

      Kaminsky, Graciela, and Carmen M.Reinhart,1999, “The Twin Crises: The Causes of Banking and Balance of Payments Problems,American Economic Review, Vol. 89 (June), pp. 473500.

      Kaplan, Ethan, and DaniRodrik,2001, “Did the Malaysian Capital Controls Work?NBER Working Paper No. 8142 (Cambridge, Massachusetts:National Bureau of Economic Research).

      Kenen, Peter B.,2000, “Financial-Sector Reform in Emerging-Market Countries: Getting the Incentives Right,paper prepared for the Per Jakobsson Foundation panel discussion,Strengthening the Resilience of Global Financial Markets,Lucerne, Switzerland,June.

      Kopits, George,2000, “How Can Fiscal Policy Avert Currency Crises?IMF Working Paper 00/185 (Washington:International Monetary Fund).

      Kraay, Aart,2000, “Do High Interest Rates Defend Currencies During Speculative Attacks?Policy Research Working Paper No. 2267 (Washington: World Bank).

      Krueger, Anne, and AaronTornell,1999, “The Role of Bank Restructuring in Recovering from Crises: Mexico 1995–98,NBER Working Paper No. 7042 (Cambridge, Massachusetts:National Bureau of Economic Research), pp. 28–34.

      Krugman, Paul,1979, “A Model of Balance-of-Payments Crises,Journal of Money, Credit and Banking, Vol. 11 (August), pp. 31125.

      Krugman, Paul,1998, “What Happened to Asia?” (unpublished; Cambridge, Massachusetts:Massachusetts Institute of Technology).

      Krugman,2001, “Crises: The Next Generation?” (unpublished; Princeton: Princeton University).

      Lane, Timothy D.,Atish R.Ghosh, JavierHamann, StevenPhillips, MarianneSchulze-Ghattas, and TsidiTsikata,1999, IMF-Supported Programs in Indonesia, Korea, and Thailand: A Preliminary Assessment, IMF Occasional Paper No. 178 (Washington:International Monetary Fund).

      Lane, Timothy D., and StevenPhillips,2000, “Does IMF Financing Result in Moral Hazard?IMF Working Paper 00/168 (Washington:International Monetary Fund).

      Lindgren, Carl-Johan, Thomas J.T.Balino, CharlesEnoch, Anne-MarieGuide, MarcQuintyn, and LeslieTeo,1999, Financial Sector Crisis and Restructuring: Lessons from Asia, IMF Occasional Paper No. 188 (Washington:International Monetary Fund).

      McBrady, Matthew R., and Mark S.Seasholes,2000, “Bailing-In” (unpublished; Berkeley:University of California at Berkley).

      Milesi-Ferretti, Gian Maria, and AssafRazin,1996, “Current Account Stability,Princeton Studies in International Finance No. 81 (Princeton, New Jersey:International Finance Section, Department of Economics, Princeton University).

      Mulder, Christian, RobertoPerelli, and ManuelRocha,2001, “The Role of Corporate, Legal, and Macro Balance Sheet Indicators in Crisis Detection and Prevention” (unpublished; Washington:International Monetary Fund).

      Musso, Michael, and Miguel A.Savastano,1999, “The IMF Approach to Economic Stabilization,IMF Working Paper 99/104 (Washington:International Monetary Fund).

      Mussa, Alberto, and StevenPhillips,2001, “Comparing Projections and Outcomes of IMF-Supported Programs,IMF Working Paper 01/45 (Washington:International Monetary Fund).

      Nunnenkamp, Peter, 1999, “The Moral Hazard of IMF Lending: Making a Fuss About a Minor Problem?Kiel Discussion Papers, March (Kiel:Institut für Weltwirtschaft).

      Olivier, Jeanne, and JerominZettelmeyer,2001, “International Bailouts, Moral Hazard, and Conditionally,paper prepared for the 33rd Economic Policy Panel meeting, Stockholm,April.

      Oxford Analytica,2001, “Russia: Internal Debt (E),Oxford Analytica Brief (November 9, 2001, Part 1).

      Özatay, F.,2000, “The 1994 Currency Crisis in Turkey,Journal of Policy Reform, Vol. 3, pp. 32752.

      Radelet, Steven, and Jeffrey D.Sachs,1998, “The East Asian Financial Crisis: Diagnosis, Remedies, Prospects,Brookings Papers on Economic Activity: 1, Brookings Institution, pp. 190.

      Rodrik, Dani, 1999, “The New Global Economy and Developing Countries: Making Openness Work,Policy Essay No. 24 (Washington:Overseas Development Council; Baltimore,Maryland:Johns Hopkins University Press).

      Sachs, Jeffrey D., AaronTornell, and AndrésVelasco,1996, “The Collapse of the Mexican Peso: What Have We Learned?Economic Policy, No. 22 (April), pp. 1563.

      Sachs, Jeffrey D., AaronTornell, and AndrésVelasco,1996, “Financial Crises in Emerging Markets: The Lessons from 1995,NBER Working Paper No. 5576 (Cambridge, Massachusetts: National Bureau of Economic Research).

      Sargent, Thomas J., and NeilWallace,1985, “Some Unpleasant Monetarist Arithmetic,Quarterly Review, U.S. Federal Reserve Bank of Minneapolis, Vol. 9 (Winter), pp. 1531.

      Summers, Lawrence H.,2000, “International Financial Crises: Causes, Prevention, and Cures,American Economic Review, Papers and Proceedings (U.S.), Vol. 90 (May), pp. 116.

      Tanner, Evan,2000, “Exchange Market Pressure and Monetary Policy: Asia and Latin America in the 1990s,Staff Papers, International Monetary Fund, Vol. 47,No. 3, pp. 31133.

      Üçer, E., Murat, C., Caroline vanRijckeghem, and R.Yolalan,1998, “Leading Indicators of Currency Crises: A Brief Literature Survey and an Application to Turkey,Yapi Credit Economic Review, Vol. 9 (December),No. 2.

      West, Kenneth D.,1990, “The Sources of Fluctuations in Aggregate Inventories and GNP,Quarterly Journal of Economics, Vol. 105 (Winter), pp. 93971.

      Willett, Thomas D.,2000, “Understanding the IMF Debate,School of Politics and Economics Working Paper (Claremont, California:Claremont Graduate University).

      Zhang, Xiaoming Alan,1999, “Testing for ‘Moral Hazard’ in Emerging Markets Lending,IIF Research Paper No. 99–1 (Washington:Institute of International Finance).

    Recent Occasional Papers of the International Monetary Fund

    210. IMF-Supported Programs in Capital Account Crises, by Atish Ghosh, Timothy Lane, Marianne Schulze-Ghattas, Aleš Bulíř, Javier Hamann, and Alex Mourmouras. 2002.

    209. Methodology for Current Account and Exchange Rate Assessments, by Peter Isard, Hamid Faruqee, G. Russell Kincaid, and Martin Fetherston. 2001.

    208. Yemen in the 1990s; From Unification to Economic Reform, by Klaus Enders, Sherwyn Williams, Nada Choueiri, Yuri Sobolev, and Jan Walliser. 2001.

    207. Malaysia: From Crisis to Recovery, by Kanitta Meesook, II Houng Lee, Olin Liu, Yougesh Khatri, Natalia Tamirisa, Michael Moore, and Mark H. Krysl. 2001.

    206. The Dominican Republic: Stabilization, Structural Reform, and Economic Growth, by Alessandro Giustiniani. Werner C. Keller, and Randa E. Sab. 2001.

    205. Stabilization and Savings Funds for Nonrenewable Resources, by Jeffrey Davis, Rolando Ossowski, James Daniel, and Steven Barnett. 2001.

    204. Monetary Union in West Africa (ECOWAS): Is It Desirable and How Could It Be Achieved? by Paul Masson and Catherine Pattillo. 2001.

    203. Modern Banking and OTC Derivatives Markets: The Transformation of Global Finance and Its Implications for Systemic Risk, by Garry J. Schinasi. R. Sean Craig, Burkhard Drees, and Charles Kramer. 2000.

    202. Adopting Inflation Targeting: Practical Issues for Emerging Market Countries, by Andrea Schaechter, Mark R. Stone, and Mark Zelmer. 2000.

    201. Developments and Challenges in the Caribbean Region, by Samuel Itam, Simon Cueva, Erik Lundback, Janet Stotsky, and Stephen Tokarick. 2000.

    200. Pension Reform in the Baltics: Issues and Prospects, by Jerald Schiff, Niko Hobdari, Axel Schimmelpfen-nig, and Roman Zytek. 2000.

    199. Ghana: Economic Development in a Democratic Environment, by Sérgio Pereira Leite, Anthony Pelle-chio, Luisa Zanforlin, Girma Begashaw, Stefania Fabrizio, and Joachim Hamack. 2000.

    198. Setting Up Treasuries in the Baltics, Russia, and Other Countries of the Former Soviet Union: An Assessment of IMF Technical Assistance, by Barry H. Potter and Jack Diamond. 2000.

    197. Deposit Insurance: Actual and Good Practices, by Gillian G.H, Garcia. 2000.

    196. Trade and Trade Policies in Eastern and Southern Africa, by a staff team led by Arvind Subramanian, with Enrique Gelbard, Richard Harmsen, Katrin Elborgh-Woytek, and Piroska Nagy. 2000.

    195. The Eastern Caribbean Currency Union—Institutions, Performance, and Policy Issues, by Frits van Beek, José Roberto Resales, Mayra Zermeño, Ruby Randall, and Jorge Shepherd. 2000.

    194. Fiscal and Macroeconomic Impact of Privatization, by Jeffrey Davis, Rolando Ossowski, Thomas Richardson, and Steven Barnett. 2000.

    193. Exchange Rate Regimes in an Increasingly Integrated World Economy, by Michael Mussa, Paul Masson, Alexander Swoboda, Esteban Jadresic, Paolo Mauro, and Andy Berg. 2000.

    192. Macroprudential Indicators of Financial System Soundness, by a staff team led by Owen Evans, Alfredo M. Leone, Mahinder Gill, and Paul Hilbers. 2000.

    191. Social Issues in IMF-Supported Programs, by Sanjeev Gupta, Louis Dicks-Mireaux, Ritha Khemani, Calvin McDonald, and Marijn Verhoeven. 2000.

    190. Capital Controls: Country Experiences with Their Use and Liberalization, by Akira Ariyoshi, Karl Haber-meier, Bernard Laurens, Inci Ötker-Robe, Jorge Iván Canales Kriljenko, and Andrei Kirilenko. 2000.

    189. Current Account and External Sustainability in the Baltics, Russia, and Other Countries of the Former Soviet Union, by Donal McGettigan. 2000.

    188. Financial Sector Crisis and Restructuring: Lessons from Asia, by Carl-Johan Lindgren, Tomás J.T Balino, Charles Enoch, Anne-Marie Guide, Marc Quiniyn, and Leslie Teo. 1999.

    187. Philippines: Toward Sustainable and Rapid Growth, Recent Developments and the Agenda Ahead, by Markus Rodlauer, Prakash Loungani, Vivek Arora, Charalambos Christofides, Enrique G. De la Piedra, Piyabha Kongsamut, Kristina Kostial, Victoria Summers, and Athanasios Vamvakidis. 2000.

    186. Anticipating Balance of Payments Crises: The Role of Early Warning Systems, by Andrew Berg, Eduardo Borensztein, Gian Maria Milesi-Ferretti, and Catherine Pattillo. 1999.

    185. Oman Beyond the Oil Horizon: Policies Toward Sustainable Growth, edited by Ahsan Mansur and Volker Treichel. 1999.

    184. Growth Experience in Transition Countries, 1990–98, by Oleh Havrylyshyn, Thomas Wolf, Julian Berengaut, Marta Castello-Branco, Ron van Rooden, and Valerie Mercer-Blackman. 1999.

    183. Economic Reforms in Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan, and Uzbekistan, by Emine Gürgen, Harry Snoek, Jon Craig, Jimmy McHugh, Ivailo Izvorski, and Ron van Rooden. 1999.

    182. Tax Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union, by a staff team led by Liam Ebrill and Oleh Havrylyshyn. 1999.

    181. The Netherlands: Transforming a Market Economy, by C. Maxwell Watson, Bas B. Bakker, Jan Kees Martijn, and loannis Halikias. 1999.

    180. Revenue Implications of Trade Liberalization, by Liam Ebrill, Janet Stotsky, and Reint Gropp. 1999.

    179. Disinflation in Transition: 1993–97, by Carlo Cottarelli and Peter Doyle. 1999.

    178. IMF-Supported Programs in Indonesia, Korea, and Thailand: A Preliminary Assessment, by Timothy Lane, Atish Ghosh, Javier Hamann, Steven Phillips, Marianne Schulze-Ghattas, and Tsidi Tsikata. 1999.

    177. Perspectives on Regional Unemployment in Europe, by Paolo Mauro, Eswar Prasad, and Antonio Spilimbergo. 1999.

    176. Back to the Future: Postwar Reconstruction and Stabilization in Lebanon, edited by Sena Eken and Thomas Helbling. 1999.

    175. Macroeconomic Developments in the Baltics, Russia, and Other Countries of the Former Soviet Union, 1992–97, by Luis M. Valdivieso. 1998.

    174. Impact of EMU on Selected Non–European Union Countries, by R. Feldman, K. Nashashibi, R. Nord, P. Allum, D. Desruelle, K. Enders, R. Kahn, and H. Temprano-Arroyo. 1998.

    173. The Baltic Countries: From Economic Stabilization to EU Accession, by Julian Berengaut, Augusto Lopez-Claros, Francoise Le Gall, Dennis Jones, Richard Stern, Ann-Margret Westin, Effie Psalida, Pietro Garibaldi. 1998.

    172. Capital Account Liberalization: Theoretical and Practical Aspects, by a staff team led by Barry Eichen-green and Michael Mussa, with Giovanni Dell’Ariccia, Enrica Detragiache, Gian Maria Milesi-Ferretti, and Andrew Tweedie. 1998.

    171. Monetary Policy in Dollarized Economies, by Tomás Baliño, Adam Bennett, and Eduardo Borensztein. 1998.

    170. The West African Economic and Monetary Union: Recent Developments and Policy Issues, by a staff team led by Ernesto Hernández-Catá and comprising Christian A. Francois, Paul Masson, Pascal Bouvier, Patrick Peroz, Dominique Desruelle, and Athanasios Vamvakidis. 1998.

    169. Financial Sector Development in Sub-Saharan African Countries, by Hassanali Mehran, Piero Ugolini, Jean Phillipe Briffaux, George Iden, Tonny Lybek, Stephen Swaray, and Peter Hayward. 1998.

    168. Exit Strategies: Policy Options for Countries Seeking Greater Exchange Rate Flexibility, by a staff team led by Barry Eichengreen and Paul Masson with Hugh Bredenkamp, Barry Johnston, Javier Hamann, Esteban Jadresic, and Inci Ötker. 1998.

    167. Exchange Rate Assessment: Extensions of the Macroeconomic Balance Approach, edited by Peter Isard and Hamid Faruqee. 1998

    166. Hedge Funds and Financial Market Dynamics, by a staff team led by Barry Eichengreen and Donald Mathieson with Banktm Chadha, Anne Jansen, Laura Kodres, and Sunil Sharma. 1998.

    165. Algeria: Stabilization and Transition to the Market, by Karim Nashashibi, Patricia Alonso-Gamo, Stefania Bazzoni, Alain Féler, Nicole Laframboise, and Sebastian Paris Horvitz. 1998.

    164. MULTIMOD Mark III: The Core Dynamic and Steady-State Model, by Douglas Laxton, Peter Isard, Hamid Faruqee, Eswar Prasad, and Bart Turtelboom. 1998.

    Note: For information on the title and availability of Occasional Papers not listed, please consult the IMF Publications Catalog or contact IMF Publication Services.

    Capital outflows, of course, are a general characteristic of balance of payments crises, such as those in the heavily indebted countries of the 1980s. However, the magnitudes involved in the capital account crisis countries, stemming from stock imbalance and vulnerabilities, made their nature qualitatively different. It would also be lair to say that, within the economics profession, there has often been very little consensus on the nature of the crises, or the appropriate policies for dealing with them. Krugman (2001) discusses the evolution of thinking about such crises.

    Dates refer to the year of original program approval, except in the cases of Argentina and the Philippines, Argentina (1995) refers to the Ninth Review and Extension of the extended arrangement approved in March 1992; Philippines (1997) refers to the Fourth Review and Extension of the extended arrangement approved in June 1994.

    For a general discussion of solvency and current account sustainability, see Milesi-Ferretti, Maria, and Razin (1996); for an application to pre-crisis conditions in Mexico, see Calvo (1996).

    Although some observers have pointed to the role of unsustainable current account deficits and associated real exchange rate misalignment in the lead-up to the Mexican crisis (see Dornbusch and Werner (1994)), there is a growing consensus that traditional flow disequilibria cannot explain the virulence of recent capital account crises and that other types of vulnerabilities triggering self-fulfilling expectations on the part of investors were a key factor in these crises. Views differ as to the link between these vulnerabilities and shifts in expectations, and the element of randomness in the latter. See Calvo and Mendoza (1996) and Sachs, Tornell, and Velasco (1996) for analyses of the crisis in Mexico; and Corsetti, Pesenti, and Roubini (1998), Lane and others (1999), Boorman and others (2000), Radelet and Sachs (1998), and Krugman (1998), for discussions of the Asian crisis.

    Calvo and Mendoza (1996) emphasize the distinction between vulnerabilities related to traditional flow disequilibria and increasingly important vulnerabilities arising from stock disequilibria.

    This argument suggests that certain types of weak fundamentals not only make it possible that the economy moves to a bad equilibrium, they make it likely. For a similar view, see Corsetti, Pesenti, and Roubini (1998). Calvo (1996) develops a similar argument in the context of a formal model, which accommodates multiple equilibria but, through additional equations, produces a unique equilibrium that is highly sensitive to parameter changes.

    In Turkey and Mexico, there were also weaknesses in the financial sector, particularly related to inadequate prudential regulation and supervision.

    In Korea, direct reserve losses were limited, but the central bank effectively lost usable reserves by shifting a large part of its foreign exchange holdings to foreign branches and subsidiaries of Korean banks, which faced difficulties rolling over their short-term liabilities.

    As noted above, two of the countries had arrangements in place at the time of the crisis that had either lapsed (Argentina) or were about to expire (Philippines).

    Korea did not have a formal peg but had pursued a policy of keeping the real effective exchange rate relatively stable.

    One hypothesis is that there exists a threshold level of Official financing below which confidence is not restored, and that official financing (in the cases reviewed here) never reached that threshold. Such a judgment would have to relate the site of the financing packages to the size of the stock imbalances underlying the crises. In any event, the absence of a strong catalytic effect in any of the programs reviewed here makes it difficult either to prove or disprove this hypothesis.

    Of course, different combinations of these would have very different ramifications for the evolution of the crisis and the real economy.

    The adequacy of official financing and private sector involvement must, of course, also be gauged in the context of the overall program, including the design and implementation of macroeconomic and structural policies; these are presaged in the following section, and taken up more fully in Chapter V.

    Thus, another methodology for assessing current account deficits uses a comparison with the current account-to-GDP ratio that would stabilize the ratio of net foreign liabilities at some specified level, such as 40 percent (see IMF, 2000b). Examining the implications of such indicators is left for future work.

    In Brazil and Mexico, although the programmed current account balances fell short of the “required balances,” these countries expected foreign direct investment (FDD inflows of about 2 percent of GDP.

    As discussed in Chapter IV below, the forced current account corrections, in turn, implied wrenching macroeconomic adjustment and sharp declines in economic activity.

    The main exception is Brazil, where capital outflows were larger than projected, but were financed by a rundown of reserves rather than by current account adjustment. It should also be noted that capital outflows from Korea in 1998 amounted to about 4½ percent of GDP, so that much of the current account surplus partly reflected the authorities’ deliberate policy of resisting a reappreciation of the nominal exchange rate and accumulating foreign exchange reserves during 1998 as well as a sharp contraction in domestic demand.

    To focus on projection errors of the flows, the program current and capital accounts are expressed in terms of actual GDP (not program GDP).

    Most theoretical frameworks model capital flows responding to current and (via exchange rate expectations) prospective interest differentials and an overall “risk premium.”

    For instance, although short-term debt is usually considered most “at risk,” it is often assumed that trade credit lines, which are collateralized by goods, may be relatively immune. In the case of Korea, however, a large component of the capital outflow in early 1998 was the drying-up of trade credits. A further complication arises from large stocks of derivative instruments (e.g., forwards and swaps in Thailand).

    If these projections were unbiased and unconditional forecasts, outturns might be expected to be better than projected in about one-half of the cases.

    In addition, as discussed in Box 3.2, there were attempts at private sector involvement after the initial program failed to stem capital outflows.

    In some cases, such as Brazil, the original IMF-supported programs sought to maintain formal or informal exchange rate pegs that the markets perceived as lacking credibility.

    One exception is Argentina, where the reelection of President Menem in May 1995 gave a boost to confidence as his campaign had been based on the need to maintain the currency board arrangement.

    For a fuller discussion of the Asian experience with the so-called second lines of defense, see Lane and others (1999).

    See, for example, Guitián (1981).

    In Argentina, the other case in which the program sought to maintain the exchange rate peg (in the form of a currency hoard), the floor on net international reserves was made an indicative target. There was also a performance criterion on “free” international reserves—gross international reserves minus currency issued and legal reserve deposits at the central bank—which allowed the central hank to use its reserves pursuant to its commitments under the currency board.

    In some cases, such as Thailand, there were also explicit limits on the amount of sterilized intervention that could be undertaken under the program.

    A case in point is Korea, where the original program envisaged a buildup of net international reserves from about $6 billion at the outset of the program in early December 1997, to $11 billion by end-December, with an automatic adjustor for shortfalls in balance of payments support. In the event, not all of the expected amount was disbursed and the NIR floor was automatically adjusted to 53 billion, which was met.

    See, For instance, Lane and Phillips (2000).

    This could, in principle, have been done in several ways—either separately or in combination. These include moral suasion to induce creditor banks to negotiate a restructuring of” credit lines or provide new money, negotiated restructuring of bond debt, payment standstills enforced by exchange controls, or default.

    Moreover, to the extent that private sector involvement imposes losses on creditors—which was not the case in either Korea or Brazil—its use could also help alleviate moral hazard.

    Theoretical arguments for exchange controls typically appeal to inefficiencies in the operation of capital markets due to, among other things, asymmetric information combined within appropriate deposit insurance; mismatches between financial intermediaries’ long-term assets and short-term liabilities that leave them vulnerable to runs; principal-agent problems that result in herd behavior: the dependence of asset values on expectations, which generates bubbles and peso problems; and problems associated within completeness of contingent markets and bounded rationality (Rodrik, 1999).

    Any form of negotiated private sector involvement requires both coordinating different creditors and confronting them with a credible threat of default; to be credible, such a threat has to he acted on in some instances. Moreover, any reforms that would make it easy for debtors to restructure bonds or other forms of debt would run counter to the economic rationale for the existence of such debt, which is based on asymmetric information.

    Thus, in Brazil, generally considered the most successful case of large-scale private sector involvement among the countries reviewed here, the projection error on the capital account still amounted to some 3½ percent of GDP.

    Standard intertemporal models of the current account suggest that a country should finance a temporary shock but adjust to a permanent shock. Since the stock of external debt is finite, capital outflows may be viewed as a temporary shock (unless the country never regains the same level of market access), which would be appropriate to finance with official resources. However, if a wide class of domestic liabilities—in the limiting case, a large fraction of the M2 monetary aggregate—is subject to outflows, this might require implausible levels of official financing.

    Strictly speaking, this argument holds even if official financing is enabling greater capital outflows as long as the offset is not one-for-one. In such cases, however, it would be very difficult to justify greater official financing without simultaneously attempting private sector involvement as well.

    This also assumes that the full financing package would be used to offset capital outflows in 1998. In fact, capital outflows in 1997 had already amounted to some 8 percent of GDP.

    A further consideration concerns the source of the crisis: when the crisis results purely or primarily from contagion, there may be a stronger argument for providing large, official financing (and the CCL facility goes part of the way in providing such insurance). But, in practice, it is often difficult to distinguish between domestically driven and pure contagion crises; notably, the initial crisis in Indonesia in 1997 was largely viewed as being driven by regional contagion.

    The IMF’s thinking on coordinated private sector involvement has evolved in light of experience. Thus, as part of the reformulated Brazil program, the authorities reached a voluntary agreement with commercial bank creditors to maintain their exposure to Brazilian banks at their end-February 1999 levels. In the more recent programs with Turkey (1999) and Argentina (2000), some form of coordinated private sector involvement has been an integral part of program design from the outset. Moreover, the technical capacity to undertake private sector involvement has also improved. At the outset of the program in Korea, for instance, basic information on the extent and terms of credit lines to banks was lacking, thereby complicating the logistics of private sector involvement; such information systems have now been established in many emerging market countries.

    See Feldstein (1999) on the need for self-help.

    Kenen (2000) proposes tying prevention efforts—such as the implementation of international standards—to the terms of IMF support in the event of crisis.

    The pattern of program projection errors in these countries differed markedly from most IMF-supported programs. See Mussa and Phillips (2001).

    Estimates of short-run activity elasticities on imports are some-what smaller in East Asia than in Latin America, so that a given decline in imports is associated with a larger decline in teal GDP.

    This feature was not exclusive to IMF forecasts. For instance, the IMF’s program projections are compared with commercial “consensus forecasts” made for the Asian crisis countries; the former were not systematically more pessimistic (Lane and others, 1999).

    The charts graph the contribution of gross fixed investment. The actual real GDP decline was greater than the sum of the components shown because of the negative contribution of inventories.

    See below on the debate regarding the “credit crunch” in East Asia. Calvo (1999) develops a model in which the sudden withdrawal of foreign saving cascades through the economy via inter-firm financial and production relationships, causing a disproportionate negative supply response.

    Note that the shocks sum to the actual change in real GDP Growth. Thus, the econometric technique decomposes the observed change in real GDP Growth into a component associated with a movement of the aggregate supply function and a component associated with the movement of the aggregate demand function.

    For Thailand, the decomposition suggests an even larger share of aggregate supply shocks. However, the very short time series of quarterly data available probably makes the estimates unreliable. Note that the predominance of supply shocks begins with the very sharp output contractions. For the run-up to the crisis (early to mid-1997 in Thailand and Korea), the Blanchard—Quah decomposition identifies negative demand shocks, especially in Thailand, perhaps reflecting falling external demand for these countries’ exports.

    In these models, a negative demand shock would be associated with an increase in inventories as firms “smooth production” by not cutting production by the full amount of the demand decline. On the other hand, faced by a shock to costs—or other disruption to production—firms reduce production and meet demand out of existing inventories, leading to a decumulation of inventories; see West (1990). It bears emphasizing, however, that such patterns are very sensitive to the precise timing of the shocks.

    For country-specific definitions of fiscal balances, see Annex III.

    Excluding carrying cost of financial sector restructuring.

    In the original program, the public sector borrowing requirement (for the first half of 1999) was a performance criterion and the primary balance was an indicative target. this implied that in the event of higher interest rates than assumed in the program, a larger adjustment in the primary balance would be needed. However, the indicative target for the primary balance may have signaled that the projected adjustment was, indeed, all that was required. At the same time, an unintended implication of the performance criterion for the public sector borrowing requirement is that it gave the authorities an incentive to refrain from raisins interest rates.

    See, for example, Chalk and Hemming (2000) for a discussion.

    The required minimum primary surplus is given by: p=(iπg)bπgv where (i -π) is the real interest rate, b is the ratio to GDP at which the level of public debt is to be stabilized, g is the real Growth rate of GDP. and v is velocity of base money (nominal GDP divided by base money). The required primary surplus is thus increasing in the real interest rate, decreasing in the Growth rate of the economy, and, for a given money velocity, decreasing in the inflation rate. Note that this includes seignorage as revenue, whereas part of base money will generally reflect increases in NIR.

    For each program, the required primary surplus as of the year-end closest to the program approval date is reported; Appendix III provides details of the calculations. By aggregating domestic and external public debt, the calculations implicitly assume that there will be no further teal depreciation (which would raise the debt ratio and thus the required primary surplus) nor any real appreciation (which would lower the debt ratio and the required primary surplus correspondingly).

    Moreover, historical Growth rates may have been unsustainably high.

    Preannounced treasury bill auctions actually had to be canceled.

    For instance, to lower the debt-to-GDP ratio by 5 percentage points and achieve an inflation rate of 60 percent roughly the inflation rate prevailing prior to the currency crisis) would have required a primary surplus of 4 percent of GDP.

    External obligations accounted for over half of total public debt.

    Typically, only the carry costs of the present value of the (implicit or explicit) fiscal burden of banking sector restructuring costs are included in the overall deficit. This is consistent with the logic of stabilizing the public debt ratio, albeit at the higher debt level implied by the additional stock of debt so created. Likewise, privatization receipts should be excluded from the overall balance and treated as a financing item.

    Giavazzi and Pagano (1990) explore the possibility that fiscal contractions can have expansionary effects on the economy.

    Some authors have, in fact, argued that prospective fiscal deficits associated with the weaknesses in the financial sector were at the core of the Asian crisis (see Burnside, Eichenbaum, and Rebelo (l999)).

    Since the impact of fiscal policy on aggregate demand is likely to be multifaceted, any summary indicator, such as the fiscal impulse, inevitably suffers from some shortcomings; see, for instance, Chand (1992). Numerically, the fiscal impulse equals the change in the fiscal stance, which, in turn, equals the difference between the actual and cyclically neutral fiscal balances. In terms of stimulus to economic activity, the overall deficit is of relevance. In terms of adjustment effort, the primary balance is a better indicator. Corresponding calculations for the primary balance are given in Annex III; in general, these differ little from the figures reported here, except those for Brazil and Indonesia, where the primary fiscal impulse turned out to be about –2 percent of GDP, and Mexico, where the primary impulse was about –4½ percent of GDP.

    Following the abandonment of the real’s currency peg and the sharp depreciation of the exchange rate in January 1999, the outlook was believed to have deteriorated significantly and Growth projections were lowered at the time of the first review. With growing evidence that these fears were unlikely to materialize, Growth projections were subsequently raised again.

    At the same time, the primary balance rather than the public sector borrowing requirement became a performance criterion.

    The stimulus was relatively small in Thailand, where the fiscal deficit fumed out significantly smaller than programmed.

    In Brazil, pan of the positive stimulus reflects the cost of higher foreign currency denominated debt interest payments.

    However, part of the positive impulse in the fourth quarter of 1997 may have reflected World Bank and Asian Development Bank loans which were channeled through the budget.

    Turkey is also the only case in which the Blanchard-Quah decomposition discussed in the previous section finds a large negative demand component to the output contraction.

    The use of such policy rules is discussed in Kopits (2000).

    Flood and Marion (1998) provide a useful survey of these models. A “third-generation” model in which corporate balance sheet vulnerabilities make the currency susceptible to a crisis is presented by Aghion, Baccehetta, and Banerjee (2000).

    In the Philippines and Indonesia, there was a ceiling on base money instead of central bank NDA. this allowed for large scale sterilized intervention in support of the currency—quite intentionally, in the original Indonesia program, as the weakness of the exchange rate was viewed as being mostly the result of contagion, and therefore to be resisted. But it may have also contributed to capital outflows because the automatic lightening from the capital outflows was vitiated by expansions in central bank credit. In the event, in Indonesia, the increase in central bank liquidity (in support of the collapsing banking system) eventually breached the base money ceiling anyway.

    In Mexico, there was a significant overrun of central bank NDA al end-1995 associated with late disbursements of agricultural subsidies and the early payment of minimum wage increases. In Indonesia, the end-1997 stock of base money was some 15 percent above the program ceiling, reflecting massive liquidity support to banks in the midst of the banking crisis, not all of which had been “sterilized” by reserve losses.

    For instance, as widely reported in the press, in the Korean program there was an understanding in early 199K that the authorities would not reduce interest rates until the exchange rate had substantially appreciated back to 1,400 won per U.S. dollar, but there was no specific commitment to raise rates further if necessary to achieve such appreciation.

    See, for instance, Furman and Stiglitz (1998), for a summary of the evidence on the interest rate-exchange rate relationship, see Lane and others (1999), p. 46. One issue concerns the initial exchange rate overvaluation. To the extent that real exchange rates are viewed as being overvalued, high dollar rates of return may no longer be effective in stemming capital outflows as devaluation expectations lake hold. As discussed above, however, the very sharp nominal and real depreciations experienced at the onset of these crises meant that any initial overvaluation (which had, by most measures, been greater in Latin America and Turkey than in the Asian countries) was soon eliminated.

    For instance, in Indonesia the Balanced Budget Law had contributed to stifling the development of a domestic government bond market, leaving relatively few instruments with which to lure investors into domestic currency assets.

    Comparing capital flows and ex post U.S. dollar returns on domestic assets is somewhat problematic because the change in the exchange rate is, of course, influenced by the capital flows.

    Of the Asian countries, Indonesia is the only one with substantial onshore dollar deposits.

    These real interest rates arc deflated by the consumer price index. Since producer prices generally rose by more, wholesale price index-deflated real interest rates were lower.

    The analysis of real money arid real credit Growth rates is complicated by a number of factors and caveats. In particular, sharp exchange rate movements can affect the behavior of these aggregates to the extent that foreign currency deposits or loans are important. In 1998, the share of foreign currency loans in total domestic credit amounted to about 7 percent in Korea, 25 percent in the Philippines, and less than 1 percent in Thailand.

    Although nominal Growth rates of money and credit are a better indication of the monetary policy stance, in terms of the impact on the real economy, presumably, it is the behavior of real aggregates that matters.

    The ordering of the vector autoregression (VAR) assumes that real money (real credit) is exogenous with respect to output Growth: in consequence, the VAR estimates may well overstate the effect of monetary policy on output Growth.

    Thus, in Argentina, real money Growth declined by about 13 percentage points in the first quarter of 1995, and the output decline was 4 percentage points in the second quarter of 1995, In Mexico, real money Growth declined by 21 percentage points (and real credit Growth by 23 percentage points in the second quarter of 1995), while real GDP Growth declined by 4 percentage points. In Turkey, real money Growth declined by 10 percentage points and real credit Growth by about 20 percentage points, while real GDP Growth fell by almost 10 percentage points.

    In Indonesia, real money and real credit Growth were strongly positive in the first and second quarters of 1998, reflecting liquidity support to the banking system. The very high nominal money Growth rates could not be supported, however, and the bout of subsequent inflation resulted in strongly negative real money/credit Growth rates in the second half of 1998.

    There is now a large body of literature on macroeconomic factors underlying currency crises; see, for example, Kaminsky and Reinhart (1999), and Berg and Pattillo (1999). Most of this literature, however, does not focus on structural factors as such. For a discussion of the role of structural factors in currency crises, see Mulder, Perrelli, and Rocha (2001); and Ghosh and Ghosh (2001).

    Since structural reforms may need time to be instituted, a key concern is the credibility of announcements about future policy intentions.

    The term “conditionality” is used broadly here to include prior actions (or conditions for completion of a review), structural performance criteria, and structural benchmarks. In addition, particularly in the Asian programs, structural reforms were often specified in policy matrices, which did not have formal status, although they may have been perceived as part of IMF conditionality; see “Structural Conditionality in Fund-Supported Programs” for a discussion.

    The question of whether blanket guarantees should be extended in the midst of crises is not uncontroversial, however. Such guarantees may help avert a full-blown run on the banking system by providing some of the fiscal sector’s strength to the banking system. If, however, the financial sector losses are sufficiently large to threaten the public balances, the guarantee may lack credibility. Blanket guarantees also raise important issues about possible moral hazard. Moreover, once such guarantees have been extended, it may be difficult to find an opportune moment to remove them. thus, the Turkish blanket guarantee, introduced as a temporary measure in 1994, remains in place as of 2001.

    Nor was there any formal structural conditionally in the Argentina (1995) program, where the main structural vulnerability concerned the banking sector. By late-1994, the share of nonperforming loans exceeded 10 percent of banks’ portfolios. this made banks vulnerable both to contagion from the Mexican financial crisis and the monetary lightening needed to defend the currency board arrangement. The result was a liquidity squeeze on weaker banks that threatened to spread to apparently solvent institutions as well. The authorities responded to the growing banking crisis with a series of measures: reserve requirements were lowered and remunerated: swap and rediscount facilities were introduced or extended: trust funds were established to foster privatization of provincial banks and help with their capitalization and restructuring; a “safety net” was established to redistribute liquidity within the system: and a privately managed deposit insurance scheme was introduced. These measures proved sufficient to abate the banking crisis and restore confidence; by late-1995, risk premia had declined, monetary policy was gradually loosened, and by early-996 bank deposits had recovered to their precrisis levels.

    Anne Krueger and Aaron Tornell (1999) note on pp. 28–34 that “a major lesson from Mexico is that… delay (in cleaning up the bank’s portfolios) is likely both to prevent recovery … and to lead to an even larger problem.”

    See, for instance, Feldstein (1998) and Goldstein (2001).

    Lindgren and others (1999) provide a detailed discussion of the various financial sector reform strategies adopted in the Asian crisis countries.

    For instance, among the ancillary measures (that were not subject to formal conditionally), the Thai program envisaged “allow[ing] Thai citizens married to foreigners to own land” as part of the market opening effort. In a similar vein, the Korean program included measures on import liberalization (“phase out the Import Diversification Program, currently covering 16 items”), it should be noted, however, that the total number of measures that were largely irrelevant to the present crises was very small.

    At the start of the Indonesian program, 16 small banks were closed. Although their closure was needed, depositors were left with concerns that they might not be adequately protected in the event that other nonviable institutions were subsequently closed.

    The assumption here is, of course, that some unobservable “shock” hit these economies at some point before the crises, triggering a policy response from the national authorities. Traditionally, the perceived nature of the shock would dictate the optimal response: if it was temporary, it should be financed; if it was permanent, it should lead to adjustment. However, the applicability of this framework to a world of capital mobility and imperfect information has been questioned by Frankel (1999).

    This is the methodology employed by Sachs, Tornell, and Velasco (1995) in their analysis of the global effects of the Mexican financial crisis.

    The exchange market pressure index is available on monthly frequency, and the financial fragility index on a quarterly basis. In the latter case, the “program date” represents the actual quarter during which the program took place.

    In particular, in the case of Argentina, the estimated weights used in the construction of the index are extremely skewed, with exchange rate movements receiving a weight of more than 99 percent (reflecting the fact that the fluctuations in the peso/dollar exchange rate have been minimal during the period 1990–99). As a result, the index moves within a significantly smaller range of values than in the other countries.

    In the cases of Mexico and Thailand, these results are consistent with well-documented attempts by these countries to prevent initial capital outflows to be reflected in monetary contractions, which led to further losses in foreign reserves. See, for example, Fane and McLeod (1999) for Thailand; and Sachs, Tornell, and Velasco (1996), and Calvo and Mendoza (1996) for Mexico. In contrast, in Brazil, reserve losses were not sterilized and interbank rates rose significantly.

    For a description of the change in regime mentioned above and the motives behind it, see, for example, Agénor, McDermott, and Üçer (1997); Üçer, Van Rijckeghem, and Yolalan (1998); and Özatay (2000).

    This is not to say that the final trigger of the crises in those countries may have involved some form of self-fulfilling prophecy. But the fact remains that, according to the index examined above, the exchange rates of those countries were under pressure for several months before a crisis erupted and they agreed to a program with the IMF.

    Again, there is no presumption in this statement to the effect that these victims of contagion were “innocent” victims, as their vulnerability to contagion may have also resulted from the pursuit of inconsistent policies.

    Data on gross reserves were available on a quarterly basis, but data on short-term debt were interpolated from annual data extracted from the IMF World Economic Outlook database. Reserves were netted out of forward operations by the central bank in the case of Thailand, and central bank deposits at foreign branches of local banks in Korea’s case.

    Although a ratio of one emerges as a straightforward benchmark because it reflects complete reserve coverage of short-term debt, there is no economic reason to set one as a policy target for such a ratio. Not only is short-term debt an incomplete measure of a country’s gross borrowing needs, but the determination of a specific “critical” threshold for any measure of financial fragility is essentially an empirical matter.

    Notably, in all of these cases, public information about the current evolution of usable reserves (Mexico and Thailand) and short-term debt (Indonesia) was limited, a fact that may have prevented these countries’ crises from having erupted earlier.

    Thus, abstracting from variations in interest rates and GDP Growth rates, a primary balance equal to the “medium-term sustainable balance” plus the carry costs of financial sector restructuring would be consistent with stabilizing the public debt ratio at the higher level of debt (i.e., inclusive of the financial sector restructuring costs).

    Effective interest rates are computed as the ratio of interest payments on the average stock of debt. For the Latin American countries (and Turkey), explicit monetary corrections are used; these are not available for the Asian countries (where, in any case, preprogram inflation rates were generally much lower). this implies that (for the Asian countries), the required “medium-term sustainable” primary surplus could be even lower than those reported.

    If a single base period approach were adopted instead, different years might need to be used according to whether actual and potential real GDP were close in that particular quarter.

    The financial crisis that followed the December 1994 devaluation of the Mexican peso.

    The Philippines has had a long history of IMF programs. The Extended Fund Facility, with total access of SDR0.8 billion, had been in place from June 1994. At the first sign of the Thai crisis, the arrangement was extended for an additional nine months in July 1997 and the total amount approved was augmented. A new two-year Stand-By Arrangement, with total access of SDR1 billion, was approved in April 1999.

    The decline in both nominal and real exchange rates was the smallest of all Asian-crisis countries.

    The figures given here differ from those reported at the time because of subsequent changes to the coverage of public sector accounts and a comprehensive revision to the national accounts data (including the level of GNP).

    By 1998, the overall public sector deficit had grown to 16 percent of GNP and, by 1999, the deficit reached almost 24 percent of GNP.

      You are not logged in and do not have access to this content. Please login or, to subscribe to IMF eLibrary, please click here

      Other Resources Citing This Publication