You are on page 1of 28



inancial asset price volatility, and its and instability. Four case studies are exam-

F potential to undermine financial stabil-

ity, has been a subject of concern in
recent years. This chapter examines his-
torical volatility and correlations between asset
classes in the major mature markets. It dis-
ined: the Black Monday crash of 1987; the
bursting of the Japanese bubble in 1990; the
Long-Term Capital Management (LTCM)
crisis of 1998; and market conditions follow-
ing the bursting of the recent technology,
cusses the links between volatility and instabil- media, and telecommunications (TMT)
ity, some of the policy lessons that have been equity bubble.
learned during various crises, and the implica- From these cases, policymakers and market
tions those lessons have today. The chapter participants could learn lessons about how
focuses mostly on equity prices, as these have volatility can become amplified in a crisis and
been unusually volatile in recent years, but how to control factors such as leverage, short-
also considers their relationship to the wider age of liquidity, and lack of transparency that
financial markets. can turn volatility into instability. This is
Asset price volatility is unavoidable and is inevitably an ongoing process, with lessons
not necessarily undesirable, since it reflects from each crisis and subsequent innovations
the process of pricing and transferring risk as by the market and by policymakers. An impor-
underlying circumstances change. Indeed, if tant continuing policy question is how to
financial markets do not react to changing avoid creating circumstances where, in a crisis,
underlying conditions in the markets (policy participants’ attempts to control their own
changes or shocks, for example), misalloca- risk by selling into falling markets make the
tion of financial resources will occur. But if overall system unstable in new ways.
volatility leads to financial instability that too
can impose real costs. Examination of past
crises indicates that the biggest dangers to Concepts: Financial Market Volatility and
financial stability seem to have come not so Financial System Instability
much from a sustained high level of volatility Since the terms “market volatility” and
as from sudden increases in volatility. This “financial instability” are often used inter-
suggests that policymakers and market partici- changeably in the public debate, it may be
pants should focus more on reducing the useful first to define and distinguish these
instability that surrounds unexpectedly strong concepts. Volatility, simply put, refers to the
turbulence than on controlling the general degree to which prices vary over a certain
level of volatility. length of time. (This chapter limits itself to
The empirical work in the chapter will discussing volatility of prices, rather than
show that most periods of high volatility in volatility of capital flows.) Most commonly—
equity prices have been associated with nega- and this convention will be followed here—
tive shocks to the real economy. But there are price volatility is defined as the standard
several instances where the volatility was deviation of changes in the log of asset prices.
rooted more in financial market disturbances Although there is no generally accepted
instead. These instances provide opportuni- definition of financial system instability or sys-
ties to look more specifically at the financial temic risk, the following definition, which
sector causes and consequences of volatility incorporates many of the elements in defini-


tions put forward by other authors, may be cause participants to reevaluate the future
useful:1 value of, and the risks embodied in, assets or
their perception of counterparty risks. There
Periods of financial system instability entail
are generally two types of shocks: those that
severe market disruptions that—by impairing the
are broad or systematic, affecting large seg-
system’s ability to provide payment services, to
ments of the financial system, and those that
price and transfer risks, and/or to allocate credit
are idiosyncratic, affecting the health of spe-
and liquidity—have the potential to cause a
cific institutions or the price movements in
reduction in real activity.
specific markets. Broad shocks are often
Financial system instability is often linked to related to large changes in one or more coun-
concerns about key financial institutions tries’ prospective macroeconomic perform-
becoming illiquid or failing, although con- ance, while examples of idiosyncratic shocks
cerns about the overall liquidity and infrastruc- are a sudden drop in the prices of certain key
ture of financial markets can also play a role. assets—sometimes stemming from a correc-
Although financial instability has the potential tion of an earlier asset price misalignment
to damage the real economy, it will not always (or bubble)—or the failure of a financial
lead to an actual reduction in economic activ- institution.
ity. Policy reactions by the authorities, for The degree to which shocks to the financial
instance, may avert economic problems. system are amplified and propagated across
Periods of financial instability are nearly markets or across institutions is a key element
always accompanied by greater market volatil- of financial system instability. Because idiosyn-
ity. However, market volatility need not imply cratic shocks originate in one part of the mar-
financial instability (see Schwartz, 1985; and ket and could spread to others, they can often
Crockett, 1997a). Volatility will often have prove particularly useful case studies of the
benign consequences and need not be a con- vulnerability of the financial system. Broad
cern to authorities. In efficient markets, where shocks, on the other hand, tend to affect the
prices embody all available information, asset financial system in several areas simultane-
price volatility will reflect the volatility of eco- ously, making it more difficult to isolate indi-
nomic fundamentals and is an inherent part vidual systemic weaknesses. The four case
of a well-functioning financial system. Even studies presented later in this chapter there-
relatively large short-term volatility can be the fore look at idiosyncratic financial shocks.
result of a rational reaction by market partici-
pants to rapidly changing events and
increased uncertainty about future returns. It Factors That Can Turn Volatility into Instability
is only when volatility becomes extreme (often Among the factors that can amplify price
referred to as “tail events”), is a potential volatility and turn it into instability are the
source of strains on key financial institutions following:
or markets, or results in self-perpetuating con-
tagious price falls, that it is associated with Incentive Structures
financial instability and should be a concern Peer-group performance measures or index-
for the authorities. tracking can encourage herding and short-
The financial system is continually subject termism among institutional investors,
to shocks (related to news or events) that leading to amplified or self-perpetuating price

1See Crockett (1997a and b), Davis (2002), and De Bandt and Hartmann (2000) and the references therein for

various definitions of financial system stability and systemic risk.


movements. Pressures to meet short-term tions as market participants try to discern the
earnings targets, for instance, or structures facts and assess the implications amid partial
that reward staff at intermediaries according information and rumors. Market uncertainty
to volume of business rather than risk- over the solvency of individual firms, and con-
adjusted return can lead to underestimation cerns (whether justified or not) about others
of long-term risk and imprudent leveraging. that share some of the same characteristics,
Conflicts of interest at intermediaries can also can impair the allocation of credit and func-
lead to insufficient disclosure of risks to tioning of payment systems.
investors. Sudden changes in herd sentiment,
amplified by any increase in leverage, could Market Infrastructure Weaknesses
then create instability through contagious Payment, clearing, or settlement systems
price falls and difficulty in repricing risks. may not be adequate to allow participants to
cope with large margin calls, doubts over
Lack of Robust Risk Management counterparty risk, or heavy volumes of busi-
Leverage increases the sensitivity of finan- ness. This could cause illiquidity and pay-
cial institutions and the system as a whole to ments difficulties to spread rapidly through
economic downturns and to asset price the system.
declines more generally. Rare events and The appropriate balance between market
regime shifts that may not be factored into discipline and regulation needs to be found.
risk measurement models or stress tests may Otherwise deregulation can lead to an exces-
be sources of unappreciated risk. Currency sive buildup of debt as new investors in the
mismatches can lead to systemic risks, espe- market underestimate the risks in the newly
cially under pegged exchange rate regimes deregulated segment of a market, while new
where the possibility of a regime change may regulatory and supervisory systems may not
not be fully taken into account in risk man- have been sufficiently calibrated to withstand
agement. Certain hedging strategies (delta an economic downturn or a burst of negative
hedging or “portfolio insurance”) may lead to news. Alternatively, regulations that tighten
feedback mechanisms that amplify price risk limits during times of market instability
movements. The unwinding of a concentra- can have procyclical effects that amplify mar-
tion of leveraged positions (relating perhaps ket volatility. Regulation could also be exces-
to a popular “carry trade” or asset bubble) can sive, hampering market innovation. All these
similarly increase volatility. A combination of are challenges that authorities unavoidably
extreme price movements and sudden realiza- face and therefore need to be prepared to
tion of previously unappreciated market and address.
credit risks could lead to heavy losses at key The potential sources of instability just men-
institutions and disruptions to market tioned are illustrated by the case studies dis-
pricing. cussed later.

Lack of Transparency
Lack of disclosure by individual firms makes Empirical Evidence on Volatility,
risk management by others under volatile con- Correlations Between Markets, and
ditions more difficult. Inadequate initial dis- Macroeconomic Factors
closure of the true scale of positions or The empirical work that follows assesses his-
financial condition can lead to sudden torical trends in financial market volatility and
changes in market sentiment when the exis- aims to separate episodes of high volatility
tence of large exposures or weaknesses that reflect macroeconomic factors from those
becomes known and to extreme price reac- that stem more from financial shocks. The


data examined relate to equity prices, foreign

exchange rates, and bond returns in Germany, Figure 3.1. Equity Market Volatility
Japan, the United Kingdom, and the United (In percent)
States, representing the four major financial 80
S&P 500 1
centers. Volatility is measured by the historical 70
standard deviation of price changes, calcu- 60
lated as the moving average over a rolling 50
sample. 40
Developments during the past 30 years sug- 30
gest that equity volatility has recently picked 20
up, while recent bond and foreign exchange 10
volatility have remained within their typical
historical bands (and indeed in a number of 1970 73 76 79 82 85 88 91 94 97 2000 03

cases show less volatility than periods in, for 80

example, the 1980s). The evidence also indi- 70
cates that the major mature equity markets 60
have become more integrated. 50
Econometric estimates suggest that, apart 40
from in Germany, the connection between 30
equity market volatility and domestic reces- 20
sions is fairly close.2 However, the periods of 10
our four case studies are exceptions where
volatility is elevated with little or no direct link 1970 73 76 79 82 85 88 91 94 97 2000 03

to domestic recessions. 80
Historical Trends in Financial Market Volatility
Equity price volatility has trended up since 40
the mid-1990s.3 Equity volatility has been par- 30
ticularly high since 2000, except in Japan, as
the TMT bubble burst, followed by shocks
such as the events of September 11, 2001, the
Enron and WorldCom accounting scandals, 1970 73 76 79 82 85 88 91 94 97 2000 03
and geopolitical uncertainty (Figure 3.1). 80
This pattern is consistent with an asymmetric Nikkei
“feedback” or “leverage effect” generally 60
observed: equity volatility tends to rise when 50
asset prices fall (Campbell, Lo, and

is important to note that these estimates examine
the correlation between volatility and recessions, but 10
do not attempt to test the causality between them. 0
3Volatility is calculated as the annualized standard 1970 73 76 79 82 85 88 91 94 97 2000 03
deviation of percentage returns over a rolling sample.
The standard deviations are calculated from an expo- Sources: Datastream; and IMF staff estimates.
1The following figures are outside the scale of this figure: 94 percent on October 5,
nentially weighted moving average of past squared
1987; and 91 percent on November 1, 1987.
returns, where the weights decay by a factor of 0.94 for
daily returns and 0.92 in the case of monthly data.


MacKinlay, 1997, p. 497). All four equity mar-

kets analyzed exhibit brief intense spikes in
Figure 3.2. Bond Market Volatility 1
(In percent) volatility during periods of financial stress,
such as the October 1987 crash and the
United States LTCM crisis in 1998. Except for the 1987
crash, equity volatility in the United States
20 and the United Kingdom until the mid-1990s
had remained generally lower than during
the oil crisis in the mid-1970s. Equity market
10 volatility in Japan surged in the early 1990s
5 following the bursting of the equity bubble,
and in Germany volatility jumped at the time
1980 83 86 89 92 95 98 2001 of reunification.
The volatility of returns on an index of 7 to
United Kingdom 10 year government bonds in the United
States, the United Kingdom, and Germany has
20 moved in a relatively stable range since the
1987 crash (Figure 3.2), and for the United
States and United Kingdom has remained
10 considerably lower than during the high infla-
5 tion of the early 1980s. Some simultaneous
spikes in volatility can be identified in all four
1980 83 86 89 92 95 98 2001 markets, including in 1994 when the U.S.
Federal Reserve reversed its interest rate pol-
Germany icy and the 1998 LTCM episode, but in gen-
25 eral spikes are much less pronounced than for
20 equities.
Like bonds, foreign exchange volatility
does not show any rising trend (Figure 3.3).
10 Foreign exchange volatility between the dol-
lar, yen, pound, and euro has been high only
at specific moments of policy uncertainty,
1980 83 86 89 92 95 98 2001 most notably around the 1985 Plaza Agree-
ment and the 1992 Exchange Rate Mechanism
Japan crisis. Since the early 1990s, the volatility of
25 the dollar vis-à-vis the euro and pound has
20 declined, with a peak in mid-2000 when the
euro reversed its decline. The yen-dollar
volatility jumped in the fall of 1998 when
10 investors reduced their yen carry trades and
associated hedging positions.
Extreme daily price changes (so-called tail
1980 83 86 89 92 95 98 2001 events) have become more frequent for equity
markets, while less frequent in bond markets
Sources: Datastream; and IMF staff estimates. and stayed close to average frequencies in for-
1Based on 7 to 10 year bond index returns.

eign exchange markets (Table 3.1). Since

October 1997, the percentage of days in


which equity prices moved more than 3 per-

cent was two to three times higher than in the
overall period since 1970. By contrast, the
number of large daily movements declined
sharply in bond markets while it remained at
about normal in foreign exchange markets. Figure 3.3. Historical Foreign Exchange Volatility
(In percent)
The frequency of tail events is a useful meas-
ure of market instability because standard 30
U.S. dollar/Euro 1
deviation measures of volatility are a form of 25
averaging that may mask occasional large
price movements that can impose strains on 20

the system.
Large equity tail events—though recently
more frequent than average—have not been 10
unusually common compared with past
episodes of financial stress.4 Monthly U.S.
equity data that includes the Great Depression 0
1973 76 79 82 85 88 91 94 97 2000
show how limited recent tail event counts
have been by comparison with some other 30
U.S. dollar/British pound
periods (Table 3.2).5 For example, the
1973–74 recession, oil shocks, and the end of
the Bretton Woods regime created deep 20
uncertainty and a period of much more fre-
quent large price moves.6
Correlations between national markets 10
have been rising for equities and in some
cases for bonds. As financial markets and
underlying economies become increasingly 0
1973 76 79 82 85 88 91 94 97 2000
integrated and companies’ operations
become more multinational, correlations 30
Yen/U.S. dollar
would be expected to rise.7 Indeed, correla-
tions between national equity returns have
risen substantially in several cases, generally 20


4Following the 1987 crash, U.S. stock markets intro- 10

duced circuit breakers that cause trading to halt after
an equity price decline reaches a certain threshold. 5
However, these have been triggered only once and so
have not directly significantly reduced the recent tail 0
event count. 1973 76 79 82 85 88 91 94 97 2000
5Jorion (2002) comes to the same conclusion using

similar data and technique. Sources: Datastream; European Central Bank; and IMF staff estimates.
1Prior to 1999, data refer to European Currency Unit.
6See Davis (2003), who compares the 1973–74 bear

market in equities to the bear market that began in

7See Bordo, Eichengreen, and Irwin (1999), who

show that, since the mid-1970s, globalization has led

economies and financial markets to be more inte-


Table 3.1. Frequency of Tail Events1

(In percent)

Sample Standard
Equity 2000s 1973–74 1970–Sep. 1997 Oct. 1997–2003 Full Sample Deviation
S&P 500 5.7 1.9 0.6 3.4 1.1 1.0
DAX 16.7 0.7 1.7 10.2 2.7 1.3
FTSE 4.4 4.4 1.3 2.6 1.5 1.0
Nikkei 9.7 9.12 2.2 5.9 2.9 1.2

Sample Standard
Bond returns 2000s 1990–92 1994 Oct. 1997–2003 Full Sample Deviation
United States 1.4 1.3 1.2 0.9 1.9 0.5
Germany 2.0 1.7 3.8 1.1 1.5 0.3
United Kingdom 1.4 1.5 1.5 0.9 1.8 0.5
Japan 0.9 2.9 3.4 0.5 1.9 0.3

Sample Standard
Foreign exchange 2000s 1990–92 1973–Sep. 1997 Oct. 1997–2003 Full Sample Deviation
Euro 0.1 0.4 0.2 0.1 0.2 0.6
Sterling 0.0 0.0 0.1 0.0 0.1 0.6
Yen 0.7 0.0 0.3 0.4 0.3 0.7
1For equity and foreign exchange, the frequency is calculated as the number of trading sessions with 3 percent or greater returns as a percent-
age of the total number of trading sessions during the relevant period. For bonds the cut-off is calculated as 3 times the full sample standard
deviation for each series of bond returns.
2Sample period is 1990 to 1992 for comparison purposes with the Japanese bursting bubble period.

involving a greater comovement with the S&P Kingdom, and Germany has generally, and
500. An average of these correlations has var- perhaps ominously (see below), been declin-
ied substantially, but reached a new high in ing (Figure 3.6).9
2002 (Figure 3.4).8 Cross-country bond
return correlations between the United
States, United Kingdom, and Germany have Macroeconomic Factors and Equity
become increasingly positive recently, in line Market Volatility
with increasingly integrated fixed-income While the level of asset prices is
markets as well as the convergence in busi- related to macroeconomic activity, the rela-
ness cycles. Only Japanese bond returns tionship between asset return volatility and
exhibited slightly declining correlation with macroeconomic conditions is not so straight-
those abroad, reflecting an increasingly iso- forward. Although studies have found that
lated domestic financial system (Figure 3.5). stock market volatility rises during eco-
The correlation of bond and equity returns nomic contractions,10 the explanations put
within the United States, the United forward for this empirical observation have

8Like the volatility measures, correlations are calculated using exponential weights with a decay factor of 0.94.
9One criticism of the correlation estimates used here is that they are biased upward during periods in which
returns are more volatile (Forbes and Rigobon, 2001). However, Chakrabarti and Roll (2002) argue that correla-
tions are not necessarily biased if the crisis is characterized by sharp asset price declines, which happen also to
coincide with heightened volatility.
10Studies of U.S. equity market volatility and the business cycle date back to Officer (1973). Schwert (1989) shows

that recessions are the single most important explanatory factor for volatility. Hamilton and Lin (1996) show that
recessions account for about 60 percent of the variation in volatility, while Campbell and others (2001) find that
volatility increases by a factor of two to three during recessions. There is also some limited empirical evidence that
cross-country stock market correlations rise during recessions (see Erb, Harvey, and Viskanta, 1994).


Table 3.2. United States: Frequency of Monthly

Equity Returns Greater Than 8 Percent
(In percent)

Periods S&P 500

1871–1899 2.0
1900s 3.3
1910s 0.8
1920s 5.0
1930s 22.3
1940s 2.5
1950s 0.0
1960s 0.8
1970s 4.1
1980s 5.7
1990s 3.3
2000s 8.8
1871–2002 4.3

Periods S&P 500

Oct. 1997–2002 6.6
Oct. 1997–Dec. 1999 3.7 Figure 3.4. Average Cross-Country Stock Market
1973–1974 10.8 Correlations 1
1980–1982 10.8
Data Source: Robert Shiller’s website:
~shiller/data.htm. 0.7
The frequency is calculated as the number of 8 percent or greater
monthly returns as a percentage of the total number of months dur- 0.6
ing the relevant period.

received only weak support.11 Recent
research, however, has shown that larger 0.3

investor uncertainty about asset fundamen- 0.2

tals tends to increase volatility (and cor- 0.1
relations) of asset returns and that this
investor uncertainty in principle rises during 1973 76 79 82 85 88 91 94 97 2000
recessions. This could explain the positive cor-
relation between equity volatility and reces- Sources: Datastream; and IMF staff estimates.
1Average of the bilateral correlations between S&P 500, FTSE, DAX, and Nikkei.
sions that has been observed. On the other
hand, periods of high market volatility that
are unrelated to economic recessions may
tend to indicate increases in investor uncer-
tainty related to instability in the financial sys-
tem rather than to macroeconomic factors.
Asset price volatility could increase even if
the fundamentals themselves do not become
more volatile. This could happen if investors
become more uncertain about underlying

11One explanation for this is that firms become riskier

during recessions because they tend to be more finan-

cially levered and as a result their share prices fluctuate
more. Yet Schwert (1989) finds that U.S. recessions still
explain a substantial part of U.S. equity market volatility
even after controlling for firm leverage.


Table 3.3. Correlations Between Historical

Volatility and Recessions
Market Own Recessions U.S. Recessions
S&P 500 0.281
FTSE 0.201 0.171
DAX –0.271 0.00
Nikkei 0.241 –0.05
1Indicates estimates that are significant at the 5 percent level.

Figure 3.5. Bilateral Bond Market Correlations

long-term economic and financial growth
0.8 United States and United States and Germany 0.8
rates and trends and therefore attach large
United Kingdom
0.6 0.6 significance to relatively small pieces of news.
0.4 0.4 This may explain why the volatility of macro-
0.2 0.2 economic variables per se explains only a
0 0
small amount of asset price volatility. (In the
G-7 there has been a general decline in the
–0.2 –0.2
volatility of many macroeconomic variables
–0.4 –0.4
1983 86 89 92 95 98 2001 1983 86 89 92 95 98 2001 such as GDP growth or inflation during the
1990s, and yet there is no evidence that asset
0.8 United States and Japan United Kingdom and Japan 0.8
price volatility has declined concurrently.)
0.6 0.6 The behavior of return volatility in the vari-
0.4 0.4 ous equity markets during business cycles sug-
0.2 0.2 gests an interesting pattern (Table 3.3).12
There is a fairly close positive correlation
0 0
between equity market volatility and domestic
–0.2 –0.2
recessions—except in Germany, where the
–0.4 –0.4 correlation was negative. Meanwhile, the
1983 86 89 92 95 98 2001 1983 86 89 92 95 98 2001
volatility in the FTSE was as almost as strongly
0.8 United Kingdom and Germany 0.8
Germany and Japan correlated with U.S. recessions as with U.K.
0.6 0.6 recessions.
0.4 0.4 High equity market volatility and domestic
0.2 0.2
recessions were particularly closely synchro-
nized in the United States and the United
0 0
Kingdom (Table 3.4) when measured by a
–0.2 –0.2 concordance statistic, which, unlike correla-
1983 86 89 92 95 98 2001 1983 86 89 92 95 98 2001
–0.4 tions, is not biased by a few large events. To

Sources: Datastream; and IMF staff estimates. 12To time recessions, for the United States, the

National Bureau of Economic Research (NBER) reces-

sion dates are used. For the other countries, the reces-
sions are dated based on the analysis presented in
Chapter III of the April 2002 World Economic Outlook
(IMF, 2002). There, business cycle turning points are
identified based on peaks and troughs in real eco-
nomic activity. Since the World Economic Outlook dates
are at a quarterly frequency, while the analysis in this
chapter is based on monthly data, we assume that the
economy is in recession during all three months of a
recession quarter.


Table 3.4. Concordance Statistics for High Equity

Volatility Regimes and Recessions1
Figure 3.6. Correlations Between Stock and Bond Returns
High U.S.
High Volatility Own U.S. Volatility
Regime Recessions Recessions Regimes 0.75
United States
United States 0.872 0.50
United Kingdom 0.822 0.832 0.802
Germany 0.53 0.622 0.652 0.25
Japan 0.48 0.41 0.50
1The concordance statistic determines the number of periods, as 0
a proportion of the number of periods in the sample, during which
the two relevant variables are in the same state. –0.25
2Indicates estimates that are significant at the 5 percent level,
implying that the respective regimes statistically coincide. –0.50

1984 87 90 93 96 99 2002
that end, an econometric model with two
equity-volatility regimes—a high-volatility and United Kingdom
a low-volatility regime—was used to estimate 0.50

the probabilities that the observed equity 0.25

returns fall into the high volatility regime
(Figure 3.7).13 Using this measure, German
and U.K. volatility appear even more closely –0.25
synchronized with U.S. recessions than with
recessions in their own countries. By contrast,
equity market volatility in Japan is relatively –0.75
1984 87 90 93 96 99 2002
detached from domestic and international
economic cycles. Japan
U.S. recessions overlap with all but three 0.50
periods when the model suggests that U.S.
equity markets were in the high-volatility
regime. The three episodes unrelated to reces- 0

sions coincided with the 1987 stock market –0.25

crash, the autumn of 1998, and the second
volatility spike in 2002, and were likely trig-
gered by financial stability concerns rather –0.75
1984 87 90 93 96 99 2002
than macroeconomic factors. Meanwhile, the
sustained period of the high-volatility regime 0.75
in Japan begins when the 1990 bubble bursts 0.50
and precedes recession by several years. These
are our four case studies (see the Appendix to
this chapter for details). 0
The correlations between equity markets
rise during U.S. recessions (Table 3.5). These
results suggest that global fundamental –0.50

1984 87 90 93 96 99 2002
13We use a Markov-switching regime econometric

model, where recurring persistent regimes of height- Sources: Datastream; and IMF staff estimates.
ened volatility are identified endogenously (see
Hamilton, 1994, for details).


Table 3.5. Equity Market Correlations During

U.S. Recessions and Expansions
United United
States Germany Kingdom Japan
United States 0.51 0.55 0.34
Germany 0.64 0.44 0.30
United Kingdom 0.69 0.58 0.29
Japan 0.60 0.71 0.54
Bottom part of matrix reports the estimated correlation coeffi-
cients during recession periods; top, during expansions.

Figure 3.7. Probability of High Equity Volatility State and

Recession Dates 1 uncertainty—proxied by U.S. recessionary
S&P 500 DAX periods—has not only an impact on the
1.0 1.0
volatility of equity returns but also their corre-
lation across countries.
0.8 0.8
U.S. recessions also overlap with all the peri-
ods when correlations between equity markets
0.6 0.6
surged abruptly except the same three
0.4 0.4
episodes identified above in the case of U.S.
high-volatility regimes. These non-recession-
0.2 0.2 related periods of heightened stock market
volatility generally corresponded to times of
0 0 greater systemic risk where flight-to-quality
1970 80 90 2000 1970 80 90 2000
dynamics were prevalent, as described in the
FTSE Nikkei
1.0 1.0
following section.

0.8 0.8
Episodes of Negative Correlation Between
0.6 0.6
Bonds and Equities
While correlations between bond returns
0.4 0.4 and equity returns in each country have typi-
cally been positive since the early 1980s, the
0.2 0.2 correlations sometimes turn negative during
periods of equity market volatility, suggesting
0 0
1970 80 90 2000 1970 80 90 2000 flight-to-quality. The three episodes in this
period coincide with the three U.S. high-
Sources: Datastream; and IMF staff estimates.
1Shaded areas show recession periods for each country.
volatility regimes identified above as not coin-
ciding with recessions (Figure 3.8). As such,
episodes of negative stock-bond correlations
tend to coincide with, and can be a signal of,
financial instability in mature markets, but
generally do not arise in periods when high
stock market volatility is related to economic
Negative correlations of equity and bond
returns also tend to coincide with sharp
increases in implied volatility in U.S. and


Table 3.6. Regime-Switching Model for Bond-

Equity Correlations: Coefficient Estimates1
(In percent)

United States Germany

β0 0.30 0.10
β1 –0.10 –0.10
Correlation 0.53 0.64
The coefficient related to the negative (positive) regime is β1 (β0)
for the U.S. and Germany. The negative values for β1 imply a nega-
tive relationship between stock and bond returns when in this
regime, and thus represent the flight-to-quality periods. The bottom
row is the correlation estimate between estimated probability of
being in the flight-to-quality regime and the implied volatility
1All estimates are significant at the 5 percent level.

German equity markets, as measured by the

volatility indexes VIX and VDAX.14 Based on a
regime-switching econometric model, the syn- Figure 3.8. Bond and Stock Return Correlations 1
chronization is measured by the correlation 0.75

between periods of negative bond-equity cor-

relations, on the one hand, and periods of 0.50
high or low implied volatility, on the other
hand (Table 3.6).15 The results suggest a close 0.25
mapping between the “flight-to-quality” peri-
ods and high levels of the VIX or VDAX
(Figure 3.9). The flight-to-quality regimes
coincided in the United States with the 1987
crash, the 1998 LTCM crisis, and the period
since mid-2000. For Germany, flight-to-quality
dynamics have been observed more or less –0.50
1985 87 89 91 93 95 97 99 2001
since 1998.
Overall, the flight-to-quality analysis sup- Sources: Datastream; and IMF staff estimates.
1Average of the stock and bond correlations for each country. Shaded areas indicate high
ports the hypothesis that periods of high equity volatility dates for the United States.
equity market volatility that are unrelated to
economic recessions tend to coincide with
heightened perception of risk by market par-
ticipants in response to increases in global
financial instability. The period since 2000—
when negative bond-equity correlations over-

14In Whaley (2000) the VIX index is referred to as

the “Investor Fear Gauge” because it tends to spike

during times of market turmoil.
15Following Stivers, Sun, and Connolly (2002), an

econometric Markov-switching model was estimated.

Bond returns were regressed on stock returns (plus
lagged bond returns and a regime-dependent con-
stant), and the coefficient on the stock returns was
allowed to take on one of two values—depending on
the positive or negative correlation regime.


lapped with a mild recession and high equity

volatility—is an exception, presenting a
“hybrid” case where both recessionary and
financial factors seem to have been at play.

Case Study Analysis of Periods of

Recessionless Financial Stress
Figure 3.9. “Flight-to-Quality” Probabilities and Although many spikes in financial asset
Implied Volatility 1
price volatility are related to periods of stress
70 1.0 in economic cycles, volatility can also spike at
United States

other times. For example, major market inno-
Flight-to-quality probabilities
(right scale)
0.8 vation, deregulation, or other structural
50 changes can lead to financial bubbles that
0.6 create volatility when they eventually burst. At
VIX 2 the outset of the bubble, new business oppor-
(in percent; left scale)
0.4 tunities can prompt a sudden rise in risk
appetite in financial markets, which is often
accompanied by a buildup of leverage (whether
10 explicitly, through direct borrowing, or implic-
itly, such as through use of derivatives).
0 0
1986 89 92 95 98 2001 Unrealistic assumptions about long-term finan-
60 1.0
cial returns and beliefs in stable relationships
Germany in markets, combined with weak risk manage-
50 ment, can encourage excessive risk-taking.
Flight-to-quality probabilities When market participants—in reaction to
(right scale)
40 exogenous events—reevaluate underlying
assumptions and curb their risk appetite, they
30 VDAX 3 start to unwind their financial positions.
(in percent; left scale) 0.4
Those exogenous events may be the proxi-
mate causes of the bursting of the bubble, but
0.2 are not necessarily the underlying causes, par-
ticularly if the market dynamics were unsus-
0 0 tainable in the long run; if the particular
1986 89 92 95 98 2001
events had not occurred, some other event in
Sources: Datastream; and IMF staff estimates. due course would likely have led to a similar
1“Flight-to-quality” probability represents the probability of being in a period when bond

and stock returns are negatively correlated, based on a regime-switching model.

reevaluation. Once the market decline begins,
2VIX is Chicago Board Options Exchange volatility index. This index is calculated by taking
leverage heightens financial stability risks: it
a weighted average of implied volatility for the eight S&P 100 calls and puts.
3VDAX represents the implied volatility of the DAX. increases investors’ losses from the falling
asset prices; it tends to raise counterparty
exposures; and it can force them to liquidate
positions quickly. These sorts of factors can
amplify the price declines.
Four particular episodes that involved
spikes in volatility provide some lessons for
financial stability and are discussed as case
studies in the Appendix at the end of this


chapter. These episodes were not accompa- markets a rise in volatility of asset prices and
nied by recessions, and so appear to have returns has only been evident in equity mar-
been less related to fundamental uncertainty kets and not in other markets such as bonds
about macroeconomic conditions. The four or foreign exchange.17 But in episodes of high
events, which all led to major concerns about equity market volatility, significant strains and
financial instability, are: flows have emerged in other markets as well.
• The Black Monday stock market crash of Although many of the details of the case stud-
1987; ies have been specific to equity markets, the
• The bursting of the Japanese equity and policy lessons are more widely applicable
real estate bubble in 1990; across the financial system.
• The LTCM crisis of 1998; and The current period of high equity volatility,
• Market conditions following the collapse of which includes the period following the col-
the TMT equity bubble in 2000.16 lapse of the TMT equity bubble, is unusual for
A sharp reduction in risk appetite in a cri- its length rather than its height. Most periods
sis, uncertainties over asset valuations, and the of volatility in recent decades have been short-
complex web of interlocking counterparty lived spikes that corresponded to sharp share
exposures may make it difficult for market price falls followed by a steady return to stabil-
participants to coordinate an orderly unwind- ity. However, the current period of higher
ing of positions without official intervention. volatility has lasted much longer than previous
These four financial instability cases suggest episodes.
that financial authorities, particularly central The unusual nature of the current period
banks, played a crucial role in restoring calm of volatility therefore makes it difficult to say
to the markets. The case studies focus less on whether it could evolve into financial instabil-
the run-up to the crisis and more on the ity. Previous crises have often arisen from peri-
period of the crisis itself and its unwinding. ods of relatively modest volatility. Arguably,
Typically, asset price volatility is particularly market participants became complacent about
high during and after the crisis, rather than in market risks, assuming for instance that exist-
the run-up, and the factors that determine ing exchange rate relationships would remain
whether volatility leads to financial instability stable or that sustained asset price rallies
can often be seen most clearly at that point. would continue. An extended period of high
In some ways, the periods of high volatility in volatility could, in fact, be less threatening to
the case studies are very different; some took financial stability than one where volatility is
place over days and others over years. Yet the low because a risk is not recognized by
lessons learned still show similarities. investors or because market mechanisms artifi-
cially dampen volatility. When volatility is in
plain sight to market participants and to regu-
Policy Implications lators, the awareness for risk management is
sharpened, more likely guarding institutions
Is the Current Period of Market Volatility a Cause and the system itself against potential finan-
for Concern? cial instability.
Although it is often stated that volatility has Nevertheless, periods of high volatility
increased in recent years, within the mature always argue for enhanced caution. First, mar-

16Part of the period following the TMT bubble coincided with a U.S. recession, but the high volatility persisted

after the recession ended.

17Although in recent months equity market volatility has fallen (see Chapter II), the average volatility over the

last three years remains high.


kets may have adjusted to the risk arising Breaking the Cycle of Amplifying Volatility
from the existing level of volatility but may Most of the case studies showed that, once a
not be prepared for a further increase. crisis had begun, the provision of liquidity by
Second, risk management systems may ade- central banks was a key factor in easing the
quately protect intermediaries from solvency funding constraints that were amplifying
and liquidity problems, but perhaps at the volatility. Liquidity injections allowed transac-
cost of lower levels of financing for the econ- tions to be settled smoothly and boosted the
omy than would be the case at lower volatility confidence of market participants that the
levels or of inefficient allocation of capital as authorities would proactively address the
intermediaries pursue profit opportunities wider crisis. They also helped to improve the
arising from the volatility itself rather than relative yield return of other assets compared
from long-run investment. Third, the volatil- with cash. Conversely, in Japan, even after the
ity may itself be an indicator of underlying asset bubble had burst, high interest rates
market weaknesses, which can be harbingers were maintained for wider policy reasons and
of instability. monetary policy thus could not soften the
Policy measures should not aim at reducing impact of falling asset prices.
asset price volatility for its own sake, but As another important step, officials and
should instead attempt: market participants can establish a forum for
• to avoid conditions where excessive vul- finding collective means to resolve short-term
nerabilities to volatility build up (e.g., liquidity problems. The agreement brokered
through excessive leverage or risk expo- by the Federal Reserve Bank of New York, for
sures); and example, permitted creditors to unwind
• to prevent volatility from triggering finan- LTCM’s positions in an orderly fashion, with-
cial instability (if, for instance, there are out the official sector providing liquidity. In
market features that, during a crisis, would other cases, private sector groupings—such as
tend to artificially amplify volatility, put stock exchanges, clearinghouses, or more
payments or settlement systems under informal crisis groups—may be able to reach
strain, or induce the bankruptcy of a key similar agreements.
intermediary). Features of the market structure can also
The policy implications therefore often aim to stop the market’s fall. Following Black
involve measures to reduce the weaknesses in Monday, circuit breakers were devised to slow
behavior of institutions and systems that can the transmission mechanisms between equity
lead to forced sales or otherwise amplify price and futures markets once a market fall begins.
volatility, rather than to directly control price If circuit breakers, however, are not well
volatility itself. designed, they could themselves be a source
The case studies indicate policy lessons of amplified volatility.
from past periods of financial stress aimed at In principle, and if possible, policy meas-
limiting the effects of volatility by: ures to avoid the amplification of volatility
• breaking the cycle of amplifying volatility; should best be taken before a crisis happens,
• strengthening risk management practices; so as to address underlying causes rather than
• aligning incentive structures; symptoms. The remaining policy lessons
• enhancing transparency; address aspects that are more preventive.
• improving market infrastructure; and However, finding the right balance is not
• finding the balance between leaving risk always easy. In particular, the debate remains
control to market discipline and unresolved as to how to strike an appropriate
regulation. balance between two important goals for con-
These topics are discussed in turn below. trolling the effects of volatility:


• setting rigorous and consistent standards during the LTCM crisis. The need to adjust
for limiting participants’ exposures and dis- exposures rapidly (such as on swap spreads
closing information on mark-to-market posi- and on options) exaggerated the breaking
tions, thereby avoiding the buildup of down of the normal price relationships
leverage and potentially unsustainable posi- between instruments, thus increasing losses
tions that amplify volatility; and and the need for participants to close posi-
• preventing these standards from simply tions at fire-sale prices. Strict Value-at-Risk
amplifying volatility in another way, for exposure limits and simple stop-loss rules also
example, by forcing or encouraging asset tend to provoke sales in a price-insensitive
sales into falling markets at fire-sale prices manner, and this experience has led some risk
to control risks. managers to reassess the need for flexibility in
There are a number of areas, described below, the application of such rules (or at least in
where this policy dilemma exists. their timing).18
The control of counterparty exposures can
exacerbate developments during a crisis.
Strengthening Risk Management Black Monday focused attention on counter-
Striking this balance is particularly perti- party exposures in equity markets and
nent in risk management, both for regulators exchange-traded futures contracts, as well as
and for the market itself. in bank clearing systems. It helped launch ini-
The degree of leverage is a crucial factor in tiatives for wider use of collateral and netting.
the extent to which volatility turns into insta- Meanwhile, in the LTCM crisis, counterparty
bility, as it can increase both market risk and exposure problems surfaced in a new range of
counterparty risk. Even a small number of markets, such as over-the-counter (OTC)
leveraged players can cause major problems derivatives and in transactions with hedge
for the market as a whole, as the portfolio funds. This has led to tighter collateral and
insurers of Black Monday, the hedge funds netting practices, such as larger haircuts, and
and other arbitrageurs of the LTCM crisis, greater emphasis on “know-your-customer”
and the telecom and energy firms of the TMT procedures. It is important not to use collat-
equity bubble showed. Their leverage creates eral as the only safeguard; in Japan, the wide-
the potential for large margin calls and even spread use of real estate and equity collateral,
for insolvency and can greatly accentuate the on the assumption that valuations were
original price fall as they attempt to rapidly robust, gave false comfort.
close out their large and sometimes highly Notwithstanding improvements in risk man-
risky positions. Continually more sophisticated agement, several questions are unresolved,
measurement of leverage—including leverage carrying the potential to amplify volatilities
embedded in off-balance-sheet exposures—is during crises:
needed as new financial instruments and • Banks and other financial institutions
strategies evolve. (including particularly large and complex
During Black Monday, the severe limitations institutions) have greatly strengthened the
of portfolio insurance in coping with tail measurement and management of consoli-
events of extreme volatility were exposed. dated counterparty and other credit expo-
While this kind of formalized computer trad- sures, including their monitoring of hedge
ing was better controlled afterwards, the risks funds. But the official sector needs to con-
associated with arbitrage were exposed again tinue to identify remaining gaps (such as in

18For a dissenting view, see Jorion (2002).


consolidated supervision of banking and sophistication of risk management else-

insurance operations), and vulnerabilities, where in the system. Some have suggested
some of which can result from differences that insurance companies have taken on
or lack of coordination and information- credit risk from banks because, by using dif-
sharing between national supervisory ferent risk methodologies, insurers estimate
systems. credit risks as being lower than banks do,
• The highly concentrated nature of the OTC and because their regulatory capital
derivatives business exposes the market to requirements for investment risks may be
the risk of failure of a major dealer, less demanding. A buildup of credit risk
although market participants contend that leverage in the insurance and other sectors
collateralization and netting agreements could amplify volatility in the event of a
cover most of the risk. In the absence of rapid reevaluation of risks—these concerns
public information about derivatives expo- are related to the debate about fair-value
sures, it is unclear how quickly exposures accounting (see below).
could grow in the event of a major market
• In the current low-yield environment, his- Aligning Incentive Structures
torical volatilities of fixed-income returns The bursting of the TMT equity bubble
have been relatively modest, and partici- demonstrated the importance of aligning mar-
pants may have been tempted to move to ket participants’ incentives with the goals of
riskier assets to improve yield. Market risk stable and efficient markets and avoiding
measurement, including through Value at short-termism. Compensation packages for
Risk (VaR), has become much more sophis- corporate managers often encouraged short-
ticated. But participants must not rely too termism, including bonuses and stock options
heavily on historical relationships, such as tied to near-term performance. Practices are
volatilities and correlations, for risk manage- now changing (partly because of changes in
ment, because a sudden shift to a higher- accounting treatment). For instance, some
yield environment is unlikely to follow companies have started instead to issue shares
historical statistical patterns. Appropriate with long-term lock-up provisions to execu-
stress tests should be conducted because, if tives. Possible conflicts of interest by stock
VaR limits are rigidly applied, many partici- market analysts and other participants
pants using similar VaR techniques could undoubtedly accentuated the bubble and the
simultaneously try to close their positions in resulting crash and have contributed to the
a falling market. lingering uncertainty about underlying com-
• The focus on internal and external ratings pany performance that is helping to keep
in the Basel II proposals, while generally equity volatility high. The corporate gover-
helpful, carries the risk of procyclical nance issues this raised have started to be
increases in lending during a boom and addressed. Index-tracking by institutional
reductions in lending if the credit environ- investors and the short-term focus on meeting
ment deteriorates. As with market risk meas- quarterly earnings targets by corporate man-
ures, too abrupt an implementation of agers, analysts, and fund managers can lead to
tighter limits risks increasing volatility dur- herd behavior, leading investors not to ques-
ing a downturn. tion the majority market view during a boom
• While banks and securities firms have and thus heightening the risk of an abrupt
improved their risk management, including change in market views.
dispersing risks by selling them to others, Looking ahead, a number of issues still
there are potential questions about the need to be addressed:


• More needs to be done to encourage • It is now better recognized that conflicts of

longer-term incentive structures for corpo- interest within investment banks can
rate managers. For instance, greater use of amplify volatilities by encouraging invest-
executive compensation packages that are ment booms and hampering full risk assess-
vested only after, say, a three-year perform- ment. The public attention suggests that
ance record would help to reduce short-ter- conflicts of interest will be dampened at
mism. But the underlying tendency for least for a while, not only through regula-
markets to focus excessively on quarterly tion but through banks’ desire to protect
earnings figures remains a difficulty. their reputations. But standards have by no
• While corporate governance is being means been raised uniformly and the risk
strengthened in the wake of the TMT equity that these conflicts could shift to less heavily
bubble, the process of agreeing new stan- regulated companies exists.
dards both within countries and internation-
ally will inevitably be complex (especially
when relating to accounting) and will last a Enhancing Transparency
number of years. The sharper focus on The need for transparency was a particular
underlying earnings, removing some of the lesson from the 1998 crisis. Globally this was
distortions of profit-smoothing, and recogniz- reflected in the new international financial
ing previously hidden factors such as stock architecture, and of particular importance to
options and pension fund valuation changes, the mature markets were topics such as
will be helpful. However, a balance needs to increased disclosure by hedge funds, at least
be found between avoiding artificial smooth- to their counterparties. The other episodes
ing and creating spurious volatility through also raised transparency issues. The Japanese
rigid application of fair-value accounting.19 and TMT equity bubbles highlighted the need
• The prevalence of index-tracking and for bank and corporate sector balance sheet
benchmarking among portfolio managers transparency and accuracy, not just so that
could be seen as reducing the risk of ampli- counterparties and analysts have meaningful
fying sales into falling markets, by leading information but also so that the reporting
investors to continue to hold their positions institutions themselves operate under the
during downturns. However, it could also right economic incentives.
amplify volatility. First, it could lead institu- Transparency could be further strength-
tional investors not to conduct due dili- ened in several areas:
gence during market rallies (for instance, • Measuring risk concentrations and leverage
the sharp gains of TMT stocks forced index- during normal market times reduces the
trackers to hold heavy weightings in those danger that a sudden realization of the
sectors). Second, there could be a sudden scale of positions during a crisis could lead
shift away from pure index-tracking when to destructive simultaneous attempts to
the market turns down, for instance if unwind exposures. While reporting and dis-
investors simultaneously shift portfolios into closure in OTC markets are being
cash, reinforced by fund managers trying to improved, more needs to be done to the
match asset allocations in their peer group. market’s ability to assess aggregate levels of
It remains unclear, however, whether there exposures in the related areas of derivatives,
is much the official sector can (or should) offshore centers, Special Purpose Vehicles,
do to address this. and hedge funds.

19A Banque de France discussion paper (2001) suggested that full fair-value accounting, in particular of banking

books, would further amplify credit cycles.


• More broadly, the process of making corpo- of payments and settlement systems in stock
rate balance sheets more transparent and markets, exchange-traded derivatives markets,
meaningful involves complex issues. One and banking systems. By 1998, similar issues
area where difficult judgments need to be were highlighted in the OTC international
made is “fair-value” accounting, and particu- bond and derivative markets, resulting in
larly how it relates to longer-term invest- tightening of practices and contractual stan-
ments by financial institutions such as dards. By contrast (or perhaps, rather, as a
insurance firms and pension funds. It is consequence) these topics were less of an
important to give the public a transparent issue in the aftermath of the TMT equity bub-
measure of institutions’ financial situations ble. Currently work continues in such areas as
in existing market conditions, while avoid- derivatives documentation, refinement of pay-
ing excessive focus on the balance sheet ments and settlement systems, and central
impact from short-term volatility. Moving to clearinghouses.
fair-value accounting for insurance compa-
nies could likely harden minimum capital
requirements, for example, and could risk Finding the Balance Between Market Discipline
amplifying volatility. and Regulation
• There may be scope for some middle ground In many respects markets functioned rea-
in the fair-value accounting debate to sonably well during the case studies illus-
achieve an appropriate level of transparency, trated. Indeed it could be argued that the
while smoothing the more extreme effects of financial instability in mature markets in the
marking to market. This could avoid unwar- 1987 stock market crash and the LTCM crisis
ranted market reactions from disclosures or was encouragingly short-lived. In the Japanese
premature supervisory requirements to sell and TMT equity bubbles, it was perhaps not
assets during market downturns. Ways could the speed but the size of the market fall that
be sought to make “fair values” more stable, caused the main problems.
help analysts interpret the sensitivity of the In considering the degree to which new pol-
results to market values, or use appropriately icy efforts are needed, it is important to strike
gradual periods for adjusting holdings to a balance between regulation and allowing
stay within regulatory standards. For market forces to work. The predisposition
instance, market prices could be averaged should perhaps be not to impose extra restric-
over a relatively short period, supplemental tions or requirements unless a solid case is
accounting information could illustrate the made that there is a market failure to be
dependence of headline data on the addressed. But the markets will continue to
assumptions made—particularly on the lia- innovate, and regulators need to innovate
bility side—or regulatory limits could use with them. Some innovations will be direct
more stable valuation measures or appropri- responses by participants seeking less regu-
ately long adjustment periods. lated alternatives as regulators become more
sophisticated in monitoring existing markets
and controlling leverage and risk. The chal-
Improving Market Infrastructure lenge for regulators is to reach the optimum
Lessons about financial infrastructure have trade-off between regulation and market disci-
tended to progress from formal, centralized, pline. Experience shows that in many areas,
markets to less formal markets, such as over- self-regulation is not enough. Participants are
the-counter transactions. The 1987 crash and often too close to events and insufficiently
Japanese bubble highlighted the importance independent to be able to see what is needed
of collateralization, netting, and other aspects for the big picture of stability. At the same


time, regulators need to work with partici- U.S. economic activity, led to increased confi-
pants to think through the likely changes in dence in U.S. financial assets, which fueled
market behavior that would result from new the stock market boom. Leveraged M&A activ-
regulations. ity led to stock retirements and takeover pre-
miums, which strongly promoted the upsurge
in stock prices. At the same time, however, the
Future Work United States was running increasingly large
Of all the areas of debate described above, trade and fiscal deficits. Financial deregula-
the question of “fair-value” accounting per- tion in other countries, especially Japan,
haps best crystallizes the need to balance the helped finance the U.S. trade deficit. In the
requirement for continuously updated risk first half of 1987, foreign institutions bought
measurement and control against not induc- as large a volume of U.S. equities as domestic
ing price-insensitive sales of positions to stay institutions. Many of these foreign investors
within limits during a crisis. There are no easy had weak risk management capabilities and
answers, but policymakers and market partici- relied on U.S. institutions to manage their
pants should find a solution that considers the funds.
systemic need to avoid amplifying market
volatility, while still keeping close and timely Crisis Trigger
control of risks at individual institutions. It In early October 1987 a disagreement
would be preferable to learn the lessons on between G-5 authorities on the appropriate
finding this middle ground from past finan- stance of monetary policy unsettled markets
cial crises rather than from the next one. and led to market speculation that the
Future editions of the GFSR will return to Louvre Accord was breaking down. On
other aspects of volatility and the policy October 14, 1987, the announcement of the
reform agenda. Potential topics for examina- unexpectedly large August trade deficit
tion include: depressed the dollar and sent U.S. bond
• the volatility of flows in mature markets, to yields up. Equities thus became less attractive
complement this analysis of price volatility; to foreign investors and also less attractive rel-
• the balance between regulation and market ative to bonds. On the same day, legislation
discipline, and possible trade-offs between was filed in Congress to eliminate tax benefits
transparency of mark-to-market values and from the financing of corporate takeovers. In
volatility; and response, arbitrage traders started to sell
• the implications of these subjects for the shares in takeover candidates, which had led
current reform agenda, including potential the earlier market rally.
procyclical effects associated with Basel II
and with “fair-value” accounting for the Market Price Reaction
insurance and pension fund industry. In the seven days after October 14 the Dow
Jones Industrial Average fell by 31 percent,
including 23 percent on October 19, 1987,
Appendix: Case Studies the largest one-day fall in its history. The cor-
relation between U.S. bond and stock prices
The “Black Monday” Stock Market Crash of 1987 turned suddenly negative amid a flight to
quality. Bid-ask spreads widened, and at times
Initial Macroeconomic and Business Conditions liquidity evaporated altogether. The equity
A dollar stabilization policy set out by the price falls and overall volatility rapidly spread
Plaza Accord in 1985 and Louvre Accord in around the world, as correlations between
early 1987, combined with steady growth in national stock markets rose sharply.


Amplifying Factors worked as a conduit for the Fed to coordinate

The use of portfolio insurance strategies by a orderly securities clearings. As a result, market
number of major institutional investors ampli- functions were recovered rapidly. Neverthe-
fied the speed of stock price falls. Portfolio less, the “flight to quality” shift of investments
insurance uses computer models to protect from stocks to bonds persisted for some time
equity portfolio values in a falling market by after the crash. Authorities in other countries
selling stock index futures automatically. This also supplied short-term liquidity in response
selling drove stock index futures prices down to the spillover to their own financial systems,
and created price gaps between futures and but in more limited fashion than in the
the underlying stocks, which gave index arbi- United States. Continental European central
trageurs an opportunity to profit by simulta- banks, in particular, kept monetary policy on a
neously buying futures and selling stocks. more even keel.
This arbitrage transferred the selling pressure Large investors moved away from computer-
from the futures market back to the stock generated portfolio insurance as a hedging
market. The ensuing stock price falls trig- tool, as they learned of its limitations during
gered further programmed selling of index large market movements.
futures, with additional pressure on spot The Fed improved payment systems and
equity prices. Only a handful of large market stocks, futures, and option clearing systems
players were responsible for much of the sell- were integrated, introducing delivery versus
ing pressure. payment and the use of collateral. Since then,
Foreign investors also amplified the market market participants as well as official bodies
decline as the dollar’s fall prompted them to have developed more extensive collateraliza-
close U.S. equity positions. tion and netting systems throughout the
Complexity and fragmentation of clearing systems financial markets that could reduce the need
for stocks, futures, and options created delay for large margin calls in the midst of market
and confusion over payments of margin calls turbulence. The Fed was also empowered to
triggered by stock price falls, raising concern lend directly to securities brokers in case of
over the solvency of securities brokers and the emergency.
ability of exchange clearinghouses to make The securities regulators introduced circuit
payments. Banks quickly restricted lending to breaker mechanisms such as price limits, posi-
brokers. The consequent illiquidity and wor- tion limits, volume limits, and trading halts.
ries that participants would make forced sales Recommendations for greater disclosure
to meet margin payments further amplified focused on payment systems positions.
the market price falls and increased the flight Although portfolio insurance standing orders
to quality. had been large and undisclosed, there was no
real move to try to encourage extra disclosure
Responses by the Market and by of participants’ positions.
the Official Sector Although market confidence was tem-
In response to mounting fear of a systemic porarily damaged, the steady recovery in
breakdown, the Federal Reserve announced equity prices after the crash (within two years
that it was ready to provide ample liquidity to the Dow Jones index was back above its pre-
the U.S. financial system. The Fed’s action crash level) restored many institutional
helped restore banks’ confidence and thus investors’ belief that equities were the highest
maintain the supply of funding to brokers and returning asset in the long run. Incentives,
market makers and avoid payments failures. based on past performance, to weight long-
Banks, which had little direct exposure to term portfolios toward equities therefore
equities and therefore remained strong, remained in place, especially in the United


States, United Kingdom, and a number of cent, and continued to drift down in the
other countries. decade that followed. Neither bond yields nor
any cross-market correlations responded
immediately. Land prices continued to rise for
Bursting of Japan’s Equity and Real Estate a while, but reacted sharply to the lending
Bubble in 1990 limits on real-estate-related industries set by
the Ministry of Finance in April 1990. By the
Initial Macroeconomic and Business Conditions fall of 1990, land prices were falling nation-
In the aftermath of the Louvre Accord, the wide. Bond-equity correlations remained posi-
Bank of Japan kept interest rates down to sup- tive until 1993. Lack of liquidity and
port the value of the dollar and to boost infrequent settlement cycles, as well as infla-
Japan’s domestic economy, stimulating demand tion concerns, inhibited the use of govern-
for equities. Easy monetary conditions encour- ment bonds as a safe haven.
aged leveraged investment, aggressive equity
financing, and excessive borrowing based on Amplifying Factors
inflated land collateral. Restrictions on land The stock market falls were amplified by
sales limited the supply of land and drove up portfolio insurance products and by arbitrage
land prices, and banks took greater risks, mostly activities between stock and futures markets—
through real-estate-related lending. Rapid bank the same mechanism as in Black Monday—as
credit expansion, supported by bank equity well as by unwinding of margin trading.
issues that increased lending capacity and by Lending based on land and, to a lesser
unrealized gains from banks’ stockholdings, extent, equities as collateral amplified Japan’s
further fueled the stock and real-estate market financial bubble and the subsequent burst.
boom. Cross-shareholdings (i.e., double- When equity prices began falling, initially
gearing), historical cost accounting, and insuffi- investors shifted their funds out of the stock
cient disclosure contributed to weakening market into land investments and bank
market discipline in an atmosphere of wide- deposits, which boosted banks’ lending
spread optimism. Starting in May 1989, con- against land collateral. The “land myth” that
cerns over inflation led the Bank of Japan to land prices would never fall and “bank myth”
progressively increase the official discount rate. that banks would never fail created a wide-
spread false belief that land and banks were a
Crisis Trigger safe haven, even after the stock market col-
Excessive price-earnings ratios and the suc- lapse began.
cessive official discount rate rises during 1989 Financial risks started to accumulate in
started to concern the equity market. As long- banks’ balance sheets. Due to long-term rela-
term interest rates spiked up in early 1990, tionships, banks did not wind down stock-
and equity futures began to fall, arbitrage holdings or, after land prices began falling,
between cash stocks and futures transmitted loans collateralized on land. Historical cost
the downward pressure to the stock market. accounting and inadequate disclosure
allowed banks to defer losses stemming from
Market Price Reaction stock falls and recognition of nonperforming
From February 20 to April 5, 1990, the loans. Nevertheless, the continued slide in
Nikkei index dropped 23 percent, even land and stock prices gradually eroded banks’
though the S&P and European indices rose, economic capital. Ineffective unwinding of
then fell further, this time in line with other impaired assets aggravated the crisis by lead-
markets. From December 31, 1989 to its low ing to credit contraction and contributing to
in October 1990, the index fell almost 50 per- recession and deflation.


Responses by the Market and by a new age of high productivity growth, finan-
the Official Sector cial globalization and the successful process
Initially, the continued strong economic toward EMU, and continued flows of funds
and monetary growth led the Bank of Japan into the United States and other mature
to continue tightening monetary policy even equity and bond markets supported a long-
though stock prices were collapsing. The Bank lasting appreciation of asset prices. However,
of Japan eventually began easing monetary weakening counterparty credit standards,
policy in August 1991 but a substantial complacent risk management, and lack of
amount of funds flowed into the government disclosure by hedge funds allowed firms such
bond market for safety. Continued land and as LTCM to build up highly leveraged posi-
stock price declines further weakened the bal- tions that were not appreciated by the market
ance sheets of the banks and corporations and that in some areas amplified the asset
despite further monetary easing and fiscal price appreciation. Instead of controlling
expansion. Eventually in February 1999, to the size of their positions with hedge funds,
abate deflationary pressures, the Bank of counterparties relied heavily on collaterali-
Japan adopted the zero interest rate policy. zation of mark-to-market exposures to
On the structural front, a series of deregula- control risks.
tions was introduced to improve the efficiency
Crisis Trigger
of the financial system and the government
promoted financial consolidation. Mark-to- In August 1998, Russia’s unilateral debt
market accounting was introduced and several restructuring triggered a global reversal of the
agencies were established by the government excessive narrowing in credit spreads.
to purchase nonperforming loans (NPLs) and Unwinding convergence plays put selling pres-
shares held by banks. sures on mature market securities that had
But, amid weak capital and low profitability, been used as collateral in leveraged positions
low interest rates and deposit guarantees in GKOs and other emerging market asset
allowed banks to delay costly debt restructur- positions. By mid-September, the rapidly
ing. Delays in debt restructuring created more mounting margin requirements pushed
NPLs than banks’ operating profits can LTCM to the brink of collapse.
absorb. Cross-shareholdings also made it diffi-
Market Price Reaction
cult for banks to sell devalued stocks, and thus
left banks highly vulnerable to equity prices. Market stories of LTCM’s weakness con-
Consequently, the financial system became tributed to the swap spread widening in the
more fragile to the point that some banks week of August 17 and equity option volatility
required injections of public capital. increases in the week of August 24. Spreads
between older (“off-the-run”) and benchmark
treasuries widened by up to 35 basis points as
Failure of LTCM in 199820 the sell-off of off-the-run issues caused their
liquidity to evaporate, while there was a flight-
Initial Macroeconomic and Business Conditions to-quality into benchmark bonds. U.S. and
In the mid-to-late 1990s, most mature other government yields dropped from
economies, especially the United States, grew September 29 to October 6. The principal
steadily in a low inflationary environment. equity markets sold off jointly and bond-equity
The belief that the U.S. economy had entered correlations turned negative in the United

20See IMF (1998) for more details.


States, United Kingdom, and Germany, its counterparties, and other market partici-
reflecting further flight-to-quality. As margin pants, took on similar leveraged positions and
calls spread to other hedge fund positions, the also faced selling pressures.
dollar dropped by 17 percent against the yen
from October 6 to 8. Responses by the Market and by
the Official Sector
Amplifying Factors Concerned that a forced liquidation of
The key amplifier in the LTCM episode was LTCM’s complex positions could produce
leverage. LTCM engaged in credit spread major market disruptions and possible coun-
plays based on the leveraging of on- and off- terparty failures among systemically important
balance-sheet positions (though reportedly institutions, the Federal Reserve orchestrated
later also took some directional positions, a coordinated resolution of LTCM by its credi-
particularly on equity volatility). LTCM lev- tors. Fourteen major creditors and counter-
ered up its positions by short-selling lower- parties of LTCM agreed to take over its
yielding high-quality assets and using the management and inject $3.6 billion to man-
proceeds to take long positions in riskier age its orderly unwinding. This coordinated
assets (mortgage-backed securities, mature effort prevented a chain reaction of distressed
market junk bonds). It also repoed assets and sales of positions and possible failures that
invested the proceeds in other relatively high- could have further disrupted U.S. and interna-
yield assets, including derivative contracts. tional capital markets. The Fed did not con-
LTCM’s balance sheet positions totaled about tribute funds to LTCM’s resolution, and
$120 billion at the beginning of 1998, com- instead provided liquidity to the wider money
pared with a capital base of $4.8 billion. At market to ensure orderly clearing of securities
the same time, LTCM held $1.3 trillion gross transactions and deter panic sales.
notional value of off-balance-sheet derivative Learning from these lessons, financial
positions. supervisors in the United States and elsewhere
Major counterparties, because of competi- put more emphasis on internal risk controls
tive pressures, did not require initial margins and risk assessment, and encouraged banks to
for derivative contracts and took no haircut intensify monitoring of their borrowers’ finan-
on repo transactions, and this allowed LTCM cial status (see IMF, 1999). Many mature mar-
to build up high leverage with relatively little ket supervisors have intensified market
capital. Lack of transparency about hedge surveillance. Due to the global repercussions
fund activities and failure by many other mar- of the LTCM incident and related problems
ket participants to adequately monitor coun- from the financial crisis, the G-7 established
terparty and market risks further allowed the Financial Stability Forum to improve cross-
LTCM and others to build up leverage. border and cross-market cooperation of offi-
Once the crisis began, LTCM’s attempts to cial agencies in identifying incipient
unwind its positions amplified the volatility. vulnerabilities. The Basel Committee on Bank
The Russian crisis, at first, widened credit Supervision published guidance on sound
spreads. LTCM responded to the resulting practice for banks’ interaction with highly
margin calls by liquidating some of its most leveraged institutions (HLIs). Internationally
liquid positions. However, the selling pressure active banks strengthened monitoring of HLIs
pushed down the prices of underlying assets and improved counterparty risk and collateral
and widened credit spreads further. This spi- management. The growing understanding of
ral gradually forced LTCM to liquidate less liq- the need to diversify credit risks also spurred
uid positions at losses. The unwinding process the growth of new financial products, such as
was also accentuated by the fact that many of credit derivatives.


Market Conditions Following the TMT Equity mid-1970s) when equity volatilities peaked,
Bubble Collapse and credit spreads reached highs not seen in
over a decade. Bond-equity correlations in the
Macroeconomic and Business Conditions United States and the United Kingdom
The long period of global economic growth turned negative and remained so from early
in the 1990s supported strong investment and in 2000, reflecting flight-to-quality. In
consumption spending—financed to a large Germany and Japan bond-equity correlations
extent by debt—and the surge in equity turned sharply negative in the fall.
prices. Information technology (IT) innova-
tion led to euphoria about the “new econ- Amplifying Factors
omy,” strong sustained productivity gains, and Leverage taken on, particularly by energy
exuberant expectations of long-term growth and telecommunications companies, ampli-
in demand and profits, especially in the TMT fied the TMT equity bubble. Many issuers in
sector. Deregulated energy and communica- these newly deregulated sectors were able to
tions markets created opportunities for rapid remain highly rated and raise large amounts
business growth. The dotcom boom was also of debt. Meanwhile others were able to raise
fuelled by the prospect of lucrative initial large amounts in the high-yield market.
public offerings or takeovers by established Moreover, attempts were made by others in
companies. the corporate sector to match the apparent
equity results of high-tech sectors by financial
Crisis Trigger leverage, including venture capital invest-
A developing investment and inventory ments in dotcom companies and telecom
overhang and overcapacities, particularly in companies. Weak corporate governance and
the fast-rising telecom and IT industries, gave internal controls allowed many companies to
rise to a reassessment of business models and reward their managers with stock options and
of projections for long-term earnings. Against other benefits, sometimes tempting managers
this background, a sharp drop in profits for to manipulate short-term earnings. Conflicts
companies in these sectors in early 2002 com- of interest and governance problems at invest-
bined with increasing nervousness about valu- ment banks led to abuses, such as mislead-
ation levels of stocks led TMT stocks to begin ingly optimistic analyst reports and allocations
falling. of IPO stock to insiders.
During the boom, many insurance and pen-
Market Price Reaction sion fund investors tended to automatically
A far slower process of risk aversion has purchase equity and debt in proportion to the
emerged through the process of unwinding market to remain close to index weightings.
the TMT equity bubble. The NASDAQ fell 32 This helped to sustain the boom, although
percent from its open on March 27 to its close these investors were not highly leveraged and
on April 14, 2000, the start of a long slide that therefore did not come under pressure to sell
ultimately took this technology-related index quickly once the bubble burst.
down 78 percent from early 2000 to late 2002. Nevertheless, during the post-bubble
Deepening and widening interactions period, gradual sales of equities by insurers to
included a decline in the broader U.S. and preserve their capital strength and meet regu-
European indices starting in the second half latory requirements as their asset portfolio val-
of 2000. Successive equity lows created deeper ues fell contributed to equity market declines.
uncertainty, culminating in the equity lows of Bank lending began to decline, reflecting
mid-to-late 2002 (for the broader markets, the the shared assessment by syndicated lenders
largest cumulative equity decline since the in late 2000 that some lending had been


excessive. The commercial paper market con- sis on independent directors. The U.S. finan-
tracted sharply, cutting off new funding and cial supervisors now require financial con-
requiring repayments in response to market glomerates to separate research and
rumors, starting in 2001. Subsequently, head- investment banking.
line bankruptcies at Enron (2001) and
WorldCom (2002) led to large investor losses
and a loss of confidence in the accuracy of References
reported corporate results. During this later Banque de France, 2001, “Financial Cycle: Factors
period lower corporate investment and GDP of Amplification and Policy Implications,”
growth, combined with the events of Banque de France Bulletin No. 95 (Paris,
September 11, 2001, and the uncertainties November).
leading up to the Iraq war, kept the equity Bordo, Michael D., Barry Eichengreen, and
falls going. Douglas Irwin, 1999, “Is Globalization Today
Really Different Than Globalization a Hundred
Responses by the Market and by Years Ago?” NBER Working Paper No. 7195
the Official Sector (Cambridge, Mass.: National Bureau of
Economic Research).
The robust banking system worked as a con-
Campbell, John, Martin Lettau, Burton G. Malkiel,
duit of liquidity to securities brokers. and Yexiao Xu, 2001, “Have Individual Stocks
Although banks had facilitated corporate Become More Volatile? An Empirical
fund-raising, they had managed to control Explorations of Idiosyncratic Risk,” Journal of
their risks, including by taking a cautious atti- Finance, Vol. 56 (February), pp. 1–43.
tude toward equity investments, selling credit Campbell, John, Andrew Lo, and A. Craig
products on to other investors, or by other- MacKinlay, 1997, The Econometrics of Financial
wise reducing exposure through devices such Markets (Princeton: Princeton University Press).
as credit derivatives. Chakrabarti, Rajesh, and Richard Roll, 2002,
Businesses themselves, facing a cash “East Asia and Europe during the 1997 Asian
squeeze, began aggressively improving their Collapse: A Clinical Study of a Financial Crisis,”
Journal of Financial Markets, Vol. 5 (January),
cash flow starting in early 2001. Investment
pp. 1–30.
spending dropped precipitously, liquidity
Crockett, Andrew, 1997a, “The Theory and Practice
cushions were built up, and the maturity of of Financial Stability,” Essays in International
borrowing extended. Corporate bond markets Finance No. 203 (Princeton: Princeton
were willing to fund companies based on case- University, Department of Economics).
by-case examination of the names, resulting in ———, 1997b, “Why Is Financial Stability a Goal of
a surge in bond issues in 2001. Public Policy?” in Maintaining Financial Stability
As longer-term policy responses, authorities in a Global Economy: A Symposium Sponsored by the
in the United States and other financial cen- Federal Reserve Bank of Kansas City (Federal
ters took measures to strengthen corporate Reserve Bank of Kansas City), pp. 7–36.
governance and accounting and auditing stan- Davis, E. Philip, 2002, “A Typology of Financial
dards. The Sarbanes-Oxley Act of 2002 cre- Instability,” Financial Stability Review 2 (Vienna:
Osterreichische Nationalbank), pp. 92–110.
ated the Public Company Accounting
———, 2003, “Comparing Bear Markets—1973 and
Oversight Board, required new rules dealing
2000,” National Institute Economic Review, No. 183
with analyst conflicts of interest, strengthened (January), pp. 78–89.
corporate governance and disclosure, and lim- De Bandt, Olivier, and Philipp Hartmann, 2000,
ited insider transactions and loans to execu- “Systemic Risk: A Survey,” European Central
tives. The New York Stock Exchange and Bank Working Paper No. 35 (Frankfurt: ECB).
NASDAQ are taking steps to tighten corporate Erb, Claude, Campbell R. Harvey, and Tadas E.
governance standards and place more empha- Viskanta, 1994, “Forecasting International Equity


Correlations,” Financial Analysts Journal, Vol. 50 Jorion, Philippe, 2002, “Fallacies About the Effects
(November–December), pp. 32–45. of Market Risk Management Systems,” Financial
Forbes, Kristin, and Roberto Rigobon, 2001, Stability Review, Issue No. 13, December
“Measuring Contagion: Conceptual and (London: Bank of England).
Empirical Issues” in International Financial Officer, Robert R., 1973, “The Variability of the
Contagion, ed. by S. Claessens and K. Forbes Market Factor of the New York Stock Exchange,”
(Boston: Kluwer Academic Publishers). Journal of Business, Vol. 46, pp. 434–53.
Hamilton, James, 1994, Time Series Analysis Schwartz, Anna, 1985, “Real and Pseudo-Financial
(Princeton: Princeton University Press) . Crises,” in Financial Crises and the World Banking
———, and Gang Lin, 1996, “Stock Market System, ed. by Forrest Capie and Geoffrey Wood
Volatility and the Business Cycle,” Journal of (New York: St. Martin’s).
Applied Econometrics, Vol. 11 (Sept.–Oct.) pp. Schwert, G. William, 1989, “Why Does Stock Market
573–93. Volatility Change Over Time?” Journal of Finance,
IMF, 1998, World Economic Outlook and International Vol. 44 (December) pp. 1129–55.
Capital Markets Interim Assessment, World Stivers, Chris, Licheng Sun, and Robert Connolly,
Economic and Financial Surveys (Washington). 2002, “Stock Implied Volatility, Stock Turnover,
———, 1999, International Capital Markets: and the Stock-Bond Return Relation,” Federal
Developments, Prospects and Key Policy Issues, World Reserve Bank of Atlanta Working Paper No.
Economic and Financial Surveys (Washington). 2002–3a (Atlanta).
———, 2002, World Economic Outlook, World Whaley, Robert, 2000, “The Investor Fear Gauge”
Economic and Financial Surveys (Washington, Journal of Portfolio Management, Vol. 26 (Spring)
April). pp. 12–17.