What Have We Learned, So Far, From the Asian Financial Crisis?1
Two years have now passed since Thailand began to exhaust its foreign exchange reserves
in a futile defense of the baht, setting in motion a chain of events that developed into an Asian
financial crisis, and then later into a global financial crisis. During the past several months, a
modicum of calm has returned to the international financial markets, providing some hope that the
worst of the crisis is over. South Korea seems to be rebounding, with improvements in credit
ratings and some initial signs of a return of foreign capital. Thailand, too, despite enormous
remaining problems, appears to be recovering. Japan has taken some steps to redress its internal
problems, and is promising to take more. Interest rate cuts in the United States and several
European countries in September and October 1998 brought about some return of confidence and
stopped the immediate threat of a further spread of the crisis. The crisis countries are still far from
out of the woods, however, with difficult tasks of financial and corporate restructuring still in
their early phases, and another year of recession nearly certain in many countries. Tremendous
risks remain, as well in Indonesia, Russia, and Brazil, so it is much too early to declare the crisis
over in any true sense.
This brief note is aimed at deciphering what happened and why, and assessing the major
policy lessons that follow from the onset of the crisis. It begins by describing the major
hypotheses that have been put forth about the origins of the crisis, and assessing the extent of the
validity of each. Was the crisis the result of extreme vulnerabilities and weaknesses in the Asian
economies? Was it the predictable consequence of the "moral hazard" of excessive risks taken
by investors in the belief that they would be bailed out in the event of a crisis? Was it the result of
exchange rate devaluations, as some observers has alleged? Was it a case of a creditor panic?
The paper then summarizes some of the main policy lessons of the crisis, in areas including
financial sector liberalization, exchange rate policy, management of capital flows, and crisis
prevention and management by the international financial institutions. We do not address the
important issue of appropriate strategies for the Asian countries for bank reorganization and
corporate restructuring, which will be dealt with in detail in a forthcoming paper.
1
Partially sponsored by the Office of Emerging Markets, Economic Growth Center,
Bureau for Global Programs, Field Support and Research, U.S. Agency for International
Development under the Consulting Assistance on Economic Reform (CAER) II Project (Contract
PCE-Q-00-95-00016-00, Delivery Order no. 16, "Next Steps in the Asian Financial Crisis").
The views and interpretations in this paper are those of the authors and should not be attributed to
USAID.
2
The Origins of the Crisis
The essence of the crisis was a huge, sudden reversal of capital flows. Economies that had
been attracting large amounts of foreign capital suddenly became subject to withdrawals of shortterm lines of credit, an exodus of portfolio capital, and offshore flight by domestic investors.
Table 1 shows that net private capital flows into the five most affected Asian economies (South
Korea, Indonesia, Thailand, Malaysia, and the Philippines) jumped from $37.9 billion in 1994 to
$97.1 billion in 1996. The bulk of these new inflows came as loans from private creditors
(commercial banks plus non-bank creditors, such as bond-holders), which tripled in just two years
from $25.8 billion to $78.4 billion. But in the last half of 1997, these inflows suddenly reversed
themselves, with net private capital flows turning to an outflow of $11.9 billion. This turnaround
of $109 billion in one year (actually just six months), from an inflow of $97 billion to an outflow
of $12 billion, is equivalent to about 10% of the pre-crisis GDP of these five countries.
One reason that such a large amount of capital was able to leave so quickly was that a
substantial portion was structured with very short-term maturities. In each of the severely-hit
economies, short-term foreign exchange liabilities of the economy grew in excess of short-term
foreign exchange assets of the economy, leaving the economy vulnerable to liquidity problems in
the event of a sudden withdrawal of foreign capital. Presumably, foreign lenders (mainly banks)
had made short-term loans under the assumption that they would routinely roll over such loans in
the future. In the event, they pulled these loans abruptly in the second half of 1997.
Table 2 shows data on one type of foreign liability, short-term debts (with maturity of one
year or less) owed to foreign commercial banks, and one major type of foreign asset, the official
foreign exchange reserves of the central bank. In Thailand, Indonesia, and Korea, short-term debt
exceeded available foreign exchange reserves just before the crisis hit. The ratio was slightly
lower (but still high) for Malaysia and the Philippines, two other Asian economies that were hit
badly, but not so severely, by the crisis. Short-term debt far exceeded reserves in Mexico and
Argentina just before those two countries were hit by financial crisis. The ratio of short term
foreign debt to reserves hit 1.7 in Mexico and 1.3 in Argentina in June 1994. The pattern also
held for Russia in mid-1998, when it faced intense balance of payments pressures. In the case of
Brazil, the data show that in June 1998, its reserves slightly exceeded its short-term debt.
However, by late September, Brazil=s reserves had dwindled to $45 billion, below the level of its
short-term debt, and the real was under intense attack.
It is important to keep in mind that short-term debts owed to foreign banks are but just
one type of short-term foreign liability. Portfolio capital, bank deposits held by foreign nonbanks, long-term loans with conversion covenants, and hedging instruments can all be withdrawn
very quickly, putting further pressure on foreign exchange reserves and the exchange rate. At the
same time, there may be other forms of foreign exchange assets in addition to official reserves that
can be drawn upon in the event of a foreign creditor panic. A priority for future research should
be to measure in a more comprehensive manner the short-term cross-border assets and liabilities
3
facing emerging market economies, and the role of various types of financial claims in the onset of
financial panic.
It is also important to note that for countries operating on pegged exchange rates, a panic
by domestic investors can also deplete foreign exchange reserves, and thereby precipitate a
financial crisis. For example, holders of sight deposits in the banking system, or domestic holders
of treasury bills, might decide suddenly to convert their domestic assets into foreign exchange,
thereby draining the foreign exchange reserves at the central bank (such, after all, is the classic
framework for understanding a balance of payments crisis). The Brazilian government, for
example, owes about $250 in domestic debt (with an average maturity of about seven months),
and redemptions of these notes have added to the pressure on the real. In earlier studies (e.g.
Sachs, Tornell, and Velasco, 1995), the ratio of M2/Reserves was used as an indicator of the
vulnerability to such a crisis. In fact, our reading of the current round of panics is that they all
occurred in circumstances with very high levels of cross-border exposure. Thus, we find that the
ratio of short-term foreign debts to reserves is a more sensitive indicator of vulnerability than the
ratio of M2 to reserves. Future crises, however, may be triggered by domestic investors rather
than foreign investors. Once again, a priority for future research is an exploration of the relative
propensities of foreign and domestic investors to financial panic, as a step in creating a better
early-warning system.
In the recent crises, the key question is why did all this capital suddenly leave? One year
ago, as the capital withdrawals swept across Thailand, the Philippines, Malaysia, Indonesia, and
later, Korea, there were four main culprits identified as causes of the crisis: (i) weaknesses within
the Asian economies, especially poor financial, industrial, and exchange rate policies; (ii) overinvestment in dubious activities resulting from the moral hazard of implicit guarantees, corruption,
and anticipated bailouts; (iii) financial panic, in that what began as moderately-sized capital
withdrawals cascaded into a panic because of weaknesses in the structure of international capital
markets and early mismanagement of the crisis, and (iv) exchange rate devaluations in mid-1997
in Thailand (and late in the year in Korea), that may have plunged these countries into panic.
Early in the crisis, almost all analysts (led by the IMF) pointed towards weaknesses in the
Asian economies and corruption-cum moral hazard, and a few (notably the editorial page of the
Wall Street Journal) laid the blame mainly on the initial devaluations. Panic and weaknesses in
international capital markets were not the explanations of choice, though we favored such a view
in our first assessment of the crisis (Radelet and Sachs, 1998a). Now, one year later, the
underlying weaknesses in international financial markets are much more widely recognized, with
widespread calls for changes in the international financial architecture. Some individual analysts,
such as Paul Krugman, have substantially changed their point of view on the causes of the crisis.
His initial analysis (Krugman, 1998) argued that problems within the Asian economies, combined
with corruption and moral hazard, led to wild over-investment and a boom-bust cycle largely
anticipated by rational market participants. A more recent analysis (Krugman, 1999) argues that
such weaknesses cannot explain the depth and severity