Managers, Investors, and Crises:
Mutual Fund Strategies in Emerging Markets
Graciela Kaminsky♦
Richard Lyons
Sergio Schmukler
This draft: 9 August 1999
Abstract
This paper addresses the emerging-market strategies of an important class of
investor—mutual funds. Data on individual portfolios allow us to characterize
strategies in ways not possible using more aggregate data. The results provide
answers to several key questions. First, do international funds use momentum
strategies, as has been documented for domestic funds? Yes, their strategies exhibit
positive momentum—they systematically buy winners and sell losers. Momentum is
present both at the investor level—through redemptions/inflows—and at the fund
manager level. Second, do funds use momentum strategies in both crisis and noncrisis periods? Yes, but contemporaneous momentum (buying current winners and
selling current losers) is stronger during crises, whereas lagged momentum (buying
past winners and selling past losers) is stronger during non-crisis periods. Third, is
momentum equally strong on the buy and sell sides? No, contemporaneous
momentum is stronger on the sell side, whereas lagged momentum is stronger on the
buy side. Fourth, is momentum equally strong across different crises? No, in Latin
America, momentum was much stronger during the 1994 Mexican Crisis. Finally, do
funds use contagion strategies, i.e., do they sell assets from one country when crisis
hits another? Yes.
JEL Classification Codes: F3, G1, G2
Keywords: mutual funds; managers; investors; trading strategies; emerging markets;
momentum; feedback trading; crisis; contagion.
♦
Respective affiliations are George Washington University, UC Berkeley and NBER, and the World Bank.
Correspondence to Sergio Schmukler, The World Bank, 1818 H Street NW, Washington, D.C., 20433, Tel:
202-458-4167, [email protected]. We thank the following for valuable comments: Michael
Gavin, Federico Sturzenegger, and participants at the World Bank/Universidad Torcuato Di Tella
conference on Integration and Contagion (June 1999) and the Cancun Meeting of the Econometric Society
(August 1999). For help with data we thank the World Bank (East Asia and Pacific Region), Eric Sirri from
the SEC, Konstantinos Tsatsaronis from the BIS, and Ian Wilson from Emerging Market Funds Research.
For excellent research assistance we thank Jon Tong, Sergio Kurlat, Cicilia Harun, Jose Pineda, and Allen
Cheung. (The efforts of Sergio Kurlat and especially Jon Tong were prodigious.) For financial support we
thank the NSF and the World Bank (Latin American Regional Studies Program and Research Support
Budget).
I. Introduction
Financial crisis in 1997 engulfed not only Asia, it also spread to countries as
distant as South Africa, the Czech Republic, and Brazil. Why did crisis spread so far, and
so fast? Trade links are a possible explanation.1 But in many crisis episodes, trade links—
even indirect links—are negligible (e.g., the trade links between Latin America and
Mexico in 1994). Financial links are another, complementary explanation. There are
many ways, however, that financial links can help spread crisis. The key is to identify
which links matter most. This requires a high-resolution picture of the financial activities
involved.
A recent literature has begun to develop this high-resolution picture. There is
evidence that banks, for example, were important in spreading the 1997 Asian crisis. The
transmission channel was lending: countries were exposed to the same banks (Kaminsky
and Reinhart 1999). Portfolio investors have also been scrutinized, particularly
institutional investors, such as hedge funds, pension funds, and mutual funds (Brown et
al. 1998, Eichengreen and Mathieson 1998, Kim and Wei 1999, Frankel and Schmukler
1996, among many others). This scrutiny has led several authors to conclude that
institutional investors sometimes panic, disregarding fundamentals, and spreading chaos
even to countries with strong fundamentals. The literature notes that individual investors,
too, can contribute to this panic by fleeing from funds—particularly mutual funds—
forcing managers to sell when fundamentals do not warrant selling.
This paper contributes to this growing literature on financial links by examining
the strategies of an important class of investor: U.S. mutual funds. Surprisingly,
systematic analysis of mutual funds’ international strategies does not yet exist.2 This lack
of systematic analysis applies to funds’ behavior during crises as well. Though there is
some evidence that funds help spread crisis, that evidence is indirect and highly
aggregative. Frankel and Schmukler (1998b), for example, use closed-end mutual funds
1
See, for example, Eichengreen, Rose, and Wyplosz (1996), Corsetti, Pesenti, Roubini, and Tille (1998),
and Glick and Rose (1998).
2
Funds’ domestic (U.S.) strategies have been analyzed extensively, however. See Grinblatt et al. (1995),
Warther (1995), and Wermers (1999), among others.
1
to show that the Mexican crisis in 1994 was not transmitted to Asia directly, but
indirectly, via New York, where the funds are traded. The opposite view—that mutual
funds do not spread crisis—also has some support in aggregate data. For example, net
redemption by mutual-fund investors during crisis periods is not large, and outflows that
do occur tend to be small and short-lived (at least during Mexico’s crisis—see Marcis et
al. 1995 and Rea 1996). Froot et al. (1998) present a similar picture based on flows that
consolidate mutual funds with other types of international investor. In particular, they find
that net inflows during the Mexican and Asian crises decreased, but there is little
evidence of net outflows.
Our data on individual portfolios allow us to examine funds' trading strategies at
much higher resolution. The data include the quarterly holdings of 13 mutual funds from
April 1993 to January 1999. All 13 of these funds are dedicated Latin America funds (and
currently, only 16 dedicated Latin America funds exist, so we cover most of this market).
We use these data to address two sets of questions. The first set relates to funds’ use of
momentum strategies—systematically buying winning stocks and selling losing stocks
(Jegadeesh and Titman 1993, Grinblatt et al. 1995). The second set of questions relates to
funds’ use of contagion strategies—systematically selling assets from one country when
crisis hits another. This direct link between contagion and trading strategies has not yet
been addressed in the literature.
We find that international funds do use momentum strategies. Their strategies
exhibit positive momentum—they systematically buy winners and sell losers. This is due
to momentum both at the investor level—through redemptions/inflows—and at the fund
manager level. Funds also use momentum strategies in both crisis and non-crisis periods,
though contemporaneous momentum (buying current winners and selling current losers)
is stronger during crises, whereas lagged momentum (buying past winners and selling
past losers) is stronger during non-crisis periods. We also find that contemporaneous
momentum is stronger on the sell side, whereas lagged momentum is stronger on the buy
side. Finally, we also find that funds use contagion strategies—they systematically sell
assets from one country when a crisis hits another.
2
The paper is organized as follows. The next section outlines our approach to
measuring momentum and contagion. Section III describes our data. Section IV presents
our momentum and contagion results. Section V addresses whether return autocorrelation
within Latin America can rationalize our section-IV results. Section VI concludes.
II. Trading Strategies: Momentum and Contagion
This section presents our approach to testing whether emerging-market funds
employ momentum and contagion strategies. Momentum strategies—also called positive
feedback strategies—involve systematic purchase of stocks that have performed well, and
sale of stocks that have performed poorly (“winners” and “losers”). Contagion strategies
involve the selling of assets from one country when crisis hits another. Contagion is thus
a cross-country phenomenon, in contrast to momentum, which is a within-country
phenomenon. (This type of cross-country analysis is not possible using recent singlecountry data sets, such as those of Kim and Wei 1999 and Choe, Kho, and Stulz 1999.)
First, we review the existing finance literature on momentum strategies. Second
we present our approach to testing for momentum, an approach that draws from this
earlier literature. Then we turn to contagion strategies, presenting first a brief review of
the "contagion" literature, followed by our approach to testing for contagion. The
approach we adopt in testing for contagion is in the same spirit as our test for momentum.
Finally, we introduce a regression-based approach that accommodates several aspects of
trading strategy—including momentum and contagion—in a unified framework.
II.1. Introduction to Momentum
The literature on momentum includes two lines of work, one based in asset
pricing and the other based in international finance. The asset-pricing line begins with the
finding that a strategy of buying past winners and selling past losers generates significant
positive returns over 3- to 12-month holding periods (Jegadeesh and Titman 1993, Asness
3
et al. 1997, Rouwenhorst 1998).3 Once established, this result inspired work on whether
investors actually follow momentum strategies. Grinblatt et al. (1995), for example,
examine the domestic strategies of US mutual funds and find that they do systematically
buy past winners. They do not systematically sell past losers, however. They also find that
funds using momentum strategies realize significantly better performance. Evaluation of
performance is in fact a central theme for all the papers in this asset-pricing line of the
literature.
The second line of work on momentum is based in international finance. Its
organizing theme is the link between returns and international capital flows. At the center
of this literature is the positive contemporaneous correlation between inflows and returns.
Early work establishes this correlation using data aggregated over both time and types of
market participant (Tesar and Werner 1994, Bohn and Tesar 1996). Later work relaxes
the aggregation over time to address whether the contemporaneous correlation in
quarterly data is truly contemporaneous (Froot et al 1998, Choe et al 1999, Kim and Wei
1999). Higher frequency data can distinguish three possibilities. Returns may precede
flows, indicating positive feedback trading (which is not necessarily irrational, per the
asset-pricing literature noted above). Returns and flows may be truly contemporaneous,
indicating that order flow itself may be driving prices.4 And returns may lag flows,
indicating flows’ ability to predict returns. Using high-frequency data aggregated across
types of market participant, Froot et al. (1998) find evidence of all three, with the first—
positive feedback trading—being the most important for explaining quarterly correlation.
3
The return “continuations” that are implied by this result are not inconsistent with the return “reversals”
documented elsewhere in the literature. Horizon length is the key to understanding this: continuations
appear at mid-range horizons, 3 to 12 months. Return reversals, in contrast, appear at short horizons (up to
1 month, see Jegadeesh 1990 and Lehmann 1990) and at long horizons (3 to 5 years, see De Bondt and
Thaler 1985). Reversals call for “contrarian” (or negative feedback) trading strategies. Parenthetically, all
these time-series anomalies are distinct from the cross-sectional anomalies that have received much
attention in the asset-pricing literature recently (e.g., size and book-to-market effects).
4
The literature on microstructure finance provides three channels for truly contemporaneous price impact.
The first is information—if the buyer has superior information about a security’s payoffs, then the purchase
acts as a signal of that information, thereby increasing price. The second is incomplete risk-sharing at the
marketmaker level—the buyer’s purchase temporarily disturbs the marketmaker’s position (“inventory”),
which requires the buyer to pay compensation in the form of a higher price. The third is imperfect
substitutability—the buyer’s purchase may represent a large enough marketwide portfolio shift that
permanently higher price is required to clear the market (even if it is common knowledge the buyer does not
have superior information about the security’s payoffs).
4
Choe et al. (1999) and Kim and Wei (1999) use high-frequency data from Korea to
examine positive feedback trading around the 1997 currency crisis. Choe et al. find that
foreign investors as a group engage in positive feedback trading before the crisis, but
during the crisis feedback trading mostly disappears. Kim and Wei examine foreign
institutional investors separately and find that they engage in positive feedback trading at
all times—before, during, and after the crisis.
Our analysis is related to, and borrows from, both of these lines of the literature.
Like the work in international finance, we are more concerned about international flows
and crisis transmission than portfolio performance. Like the work in asset pricing,
however, we maintain a direct link to investment strategy and its measurement. In
particular, we focus on a specific class of international investor—mutual funds. A benefit
of focusing on a specific investor class is that we can characterize the evolution of actual
portfolios, and how that evolution relates to returns in various countries. Another benefit
is that our data allow us to analyze jointly the behavior of fund managers and their
underlying investors. On the cost side, focusing on a specific investor class means that we
lose resolution in terms of data frequency: our data are quarterly.
II.2. Measuring Momentum
We use a momentum measure akin to that used to analyze mutual funds’ domestic
strategies (e.g., Grinblatt et al. 1995). The measure captures the relation between security
purchases/sales and returns over a benchmark period. It is based on individual
observations of the variable
 Q j ,t . Q j ,t . 1 
 R j ,t . k ,
M j ,t > 


Q
j
,
t


(1)
where Qj,t is the number of shares of stock j held at time t in a given portfolio, Q j, t is
(Qj,t+Qj,t-1)/2, and Rj,t-k is the return on security j from t-k-1 to t-k. When k=0, this
measure captures the contemporaneous relation between trades and returns—referred to
as lag-zero momentum (L0M). When k=1, the measure captures the lagged response of
5
trades to returns (feedback trading), and is referred to as lag-one momentum (L1M).
Under the null hypothesis of no momentum, the mean of the observations Mj,t is zero.
This measure of momentum has two important advantages. First, it is not
contaminated by “passive price momentum.” Passive price momentum arises in
momentum measures—like those of Grinblatt et al.—where the term in brackets is a
change in portfolio weight rather than a percentage quantity adjustment. When using a
portfolio weight, a price increase in one stock (relative to prices of other holdings)
produces a positive relation between weights and returns that has nothing to do with
investment strategy. (A similar positive relation arises for losing stocks.) The second
advantage of our measure over one based on portfolio weights is that our measure is not
contaminated by another passive effect— “passive quantity momentum.” When using
portfolio weights, a large trade in one stock can have substantial effects on the weights of
holdings that involve no transactions. Our main concern here—as in the rest of the
international-finance-based literature on momentum—is the relation between returns and
transaction flows.5 Accordingly, we want our realizations of Mj,t to reflect actual
transactions—the buying and selling of winners and losers.
Statistical Inference
The link between quantity changes and statistical inference warrants further
attention. In particular, percentage quantity changes—the term in brackets in equation
(1)—may have fund-specific volatilities. Two factors could account for differing
volatilities at the fund level. First, there is considerable cross-sectional difference in fund
size, and size can affect trading strategies. Second, trading strategies can differ
substantially across funds in ways that are distinct from size, such as turnover ratios,
redemption penalties, and other factors. Below, we test for heteroskedasticity at the fund
level, and after finding it, we correct for it.6
5
This contrasts with the asset-pricing-based literature on momentum, whose main concern is portfolio
performance, in which case it is necessary to consider the return impact of all portfolio positions.
6
Because our heteroskedasticity correction affects only standard errors, each observation of Mj,t gets equal
weight in the calculation of a momentum measure’s mean. Our correction for heteroskedasticity therefore
does not alter the fact that funds with more observations have more effective weight. We are currently
exploring whether funds differ appreciably in the intensity of their momentum trading. As for
6
Whole-Fund Momentum vs Manager-Only Momentum
An important issue in the context of mutual-fund strategies is the effect of net
redemptions. Many funds experience substantial redemptions during crisis periods. If, on
average, funds sell shares to meet redemptions when Rj,t-k is negative, then our
momentum measures will be positive. This result is not spurious. But it does reflect
strategies of underlying investors, rather than strategies of the fund manager.
To supplement our results, we control for this redemption effect by measuring the
quantity transacted in each stock relative to a benchmark. This benchmark reflects the
quantity that would be transacted if a fund's net flows from investors produced
proportional adjustment in all stocks. Specifically, to isolate the manager's contribution to
momentum we calculate individual observations of:

 Q j ,t . Q j ,t . 1
M ≤j ,t > 
Q j ,t


 )Q
.
j
j ,t

j
. Q j ,t . 1 ∗Pj ,t 

 R j ,t . k ,
Q j ,t Pj ,t


(2)
where Pj,t is the price of security j at time t, and Pj, t is (Pj,t+Pj,t-1)/2. The second term in
brackets is a term specific to each fund that captures the percent increase in portfolio size
due to net inflows. The overall momentum measure in equation (2) therefore reflects the
degree to which a manager buys winners and sells losers beyond any effect from
inflows/outflows. Under the null hypothesis of no momentum at the manager level, the
mean of the observations M ≤j ,t is zero. Henceforth, we refer to momentum statistics
calculated from equation (1) as whole-fund momentum statistics, and momentum
statistics calculated from equation (2) as manager-only momentum statistics.
Conditional Momentum
heteroskedasticity in the time-series dimension, our sample partition into crisis and non-crisis periods
accounts for the most obvious correction.
7
In addition to the overall momentum measures L0M and L1M, we are also
interested in two conditional momentum measures. The first conditional measure
involves splitting our sample into sub-periods: crisis and non-crisis. The crisis portion of
our full sample (April 1993 to January 1999) includes four sub-periods: December 1994
to June 1995 (Mexico), July 1997 to March 1998 (Asia), August 1998 to October 1998
(Russia), and January 1999 (Brazil).
For our second conditional momentum measure, we split our sample into buys
and sells (as in Grinblatt et al 1995). Overall measures like L0M and L1M make no
distinction between buying and selling. But buying past winners and selling past losers
need not be symmetric. In fact, though Grinblatt et al. find positive L0M and L1M
overall, this result comes primarily from funds buying past winners—funds did not
systematically sell past losers. This distinction has particular significance for our analysis
given the possibility of one-sided effects from crisis. To implement this, we calculate a
Buy L0M statistic using those observations with an increase in shares held, and a Sell
L0M statistic from those observations with a decrease in shares held. That is:

0 ,

 Q j ,t . Q j ,t . 1  ~
 R j ,t . k . R j
Buy L0M > Mean 


Q

j ,t

∗
 Q j ,t . Q j ,t . 1 

?


Q
j ,t


 Q j ,t . Q j ,t . 1  ~
 R j ,t . k . R j
Sell L0M > Mean 


Q

j ,t

∗

 Q j ,t . Q j ,t . 1 

 = 0 .


Q j ,t



)
)
(3)
~
Here, we subtract mean returns R j from individual stock returns R j ,t . k so that these
measures asymptotically approach zero under the null hypothesis that momentum
investing is not present. (We use the local-market index return as a proxy for these mean
returns.)
II.3. Introduction to Contagion
8
The financial crises of the 1990s in Europe, Mexico, Asia, Russia, and Brazil
spread rapidly across countries, including countries with diverse market fundamentals.7
These events spawned a literature to make sense of the seeming "contagion." From the
outset, however, it was clear that authors were using that term quite differently. Presently,
the literature on contagion identifies three types: fundamental-spillover contagion,
common-cause contagion, and non-fundamental contagion. Fundamental-spillover
contagion is the rapid transmission of an inside disturbance to multiple, economically
interdependent countries. Common-cause contagion is the rapid transmission of an
outside disturbance to multiple countries (e.g., a fall in commodity prices, or learning
about common fundamental factors). Fundamental disturbances underlie both of these
first two types. The third type—non-fundamental contagion—is the rapid transmission of
a disturbance to multiple countries beyond what is warranted by fundamentals (or after
controlling for fundamentals). This third type is sometimes referred to as pure or true
contagion.
Many authors focus on the two types of contagion driven by fundamentals. For
example, Eichengreen, Rose, and Wyplosz (1996) examine whether contagion is more
prevalent among countries with either important trade links or similar market
fundamentals. In the first case, devaluation in one country reduces competitiveness in
partner-countries, prompting devaluations to restore competitiveness (fundamentalspillover contagion). In the second case, devaluation acts like a wake-up call: investors
seeing one country collapsing learn about the fragility of “similar” countries, and
speculate against those countries' currencies (common-cause contagion). The Eichengreen
et al. evidence points in the direction of trade links rather than similar fundamentals.
Corsetti et al. (1998) also claim that trade links drive the strong spillovers during the
Asian crisis. Kaminsky and Reinhart (1999) focus instead on financial-sector links. In
7
Witness Indonesia in 1997. Nobody can disagree that there were signs of weakness in the Indonesian
economy at the outset of the Asian crisis: the banking sector was fragile, the economy was not growing, and
there was a current account deficit. Still, these problems were not insurmountable. Kaminsky (1998), for
example, estimates that the probabilities of crisis in Indonesia by June 1997 amounted to only 20 percent.
This probability stands in sharp contrast to the likelihood of a currency crisis in Thailand, which
skyrocketed to 100 percent at the beginning of 1997. Still, the Indonesian rupiah collapsed only weeks after
the floating of the Thai baht.
9
particular, they examine the role of common bank lenders and the effect of cross-market
hedging (a type of common-cause contagion). They find that common lenders were
central to the spreading of the Asian crisis (as they were to the spreading of the Debt
Crisis of 1982).
The non-fundamental category of contagion has attracted more attention than the
two fundamentals-driven categories. On the theoretical side of this portion of the
literature, the focus is on models of rational herding. For example, in the model of Calvo
and Mendoza (1998), the costs of gathering country-specific information induce rational
investors to follow the herd. In the model of Calvo (1999), uninformed investors replicate
selling by liquidity-squeezed informed investors because the uninformed mistakenly (but
rationally) believe these sales are signaling worsening fundamentals. Kodres and Pritsker
(1999) focus on investors who engage in cross-market hedging of macroeconomic risks.
In that paper, international market comovement can occur in the absence of any relevant
information, and even in the absence of direct common factors across countries. For
example, a negative shock to one country can lead informed investors to sell that
country’s assets and buy assets of another country, increasing their exposure to the
idiosyncratic factor of the second country. Investors then hedge this new position by
selling the assets of a third country, completing the chain of contagion from the first
country to the third.
The literature on non-fundamental contagion also has an empirical branch.
Kaminsky and Schmukler (1999) find that spillover effects unrelated to market
fundamentals are quite common, and spread quickly across countries within a region.
Valdes (1998) examines the degree to which comovement of Brady-bond prices is
unexplained by fundamentals. Interestingly, contagion in his paper is symmetric, applying
both on the downside during crises and on the upside during periods of rapid capital
inflow. To examine whether crises are spread by a particular investor group, it is
necessary to have information on quantities—in particular, transaction quantities. Several
recent papers have undertaken this task. For example, Choe, Kho, and Stulz (1998) use
daily data on equity transactions in Korea to examine whether foreign investors might
destabilize prices. They find evidence of herding by foreign investors before Korea’s
10
economic crisis in late 1997, but these effects disappear during the peak of the crisis, and
there is no evidence of destabilization. Since their data only include transactions on the
Korean Stock Exchange, these authors cannot examine the transmission of crisis across
countries.
II.4. Measuring Contagion
Our approach to testing for contagion is different from the literature reviewed
above. Our data on individual portfolios allow us to address contagion in a new way—
from the trading-strategy perspective. We will use the term contagion strategy to mean
the systematic selling (buying) of stocks in one country when the stock market falls
(rises) in another.8
To do this we introduce a new measure—a contagion measure. Our contagion
measure is based on the same methodology outlined above for measuring momentum. We
calculate individual observations of the variable:
 Q j ,t . Q j ,t . 1 
 Rc , t .
C j ,t > 


Q
j ,t


(4)
The difference here it that instead of testing for a relation between quantity changes and
own-stock returns, our contagion measure tests for a relation between quantity changes
and foreign-country returns. In effect, we are testing for what might be called "crosscountry momentum." Here, Rc,t is the return on the foreign-country index c from t-1 to t,
with c ∉{Brazil, Mexico, U.S., Asia, Russia}. Naturally, when calculating the contagion
measure for a Brazilian stock j, we do not include the Brazilian index return in Rc,t
8
Notice that this definition does not take account of the fundamental-versus-non-fundamental distinction
introduced above. The next section introduces our regression-based approach that will allow us to test for
contagion at the same time we control for various fundamental factors.
11
(similarly for Mexico). Under the null hypothesis of no contagion, the mean of the
observations Cj,t is zero.
Our contagion measure in equation (4) allows us to address many of the issues we
address with our momentum measure. For example, we can separate contagion at the
investor and manager levels by using the correction in equation (2). Similarly, we can
examine crisis versus non-crisis sub-samples, and partition the crisis sub-sample further
to isolate a particular crisis.
II.5. An Integrative, Regression-Based Approach
The bivariate relations examined via our equations (1) through (4) draw from, and
therefore allow direct comparison with, past empirical work on momentum. But these
bivariate relations provide no means of testing joint significance. Is lag-one momentum
still significant after controlling for contemporaneous price effects via lag-zero
momentum? Is cross-country contagion still significant after controlling for own-price
effects via lag-zero and lag-one momentum? Are these relations robust to the inclusion of
local-market index returns? The US-market index return?
To address these questions we need a more integrative model. A regression-based
approach provides a natural framework. All of the questions of the previous paragraph
can be addressed by estimating the following model:
 Q j ,t . Q j ,t . 1 

 > β , χ 1 R j ,t , χ 2 R j ,t . 1 , χ 3 R LA,t , χ 4 R LM ,t , χ 5 RUS ,t , φ j ,t .


Q
j ,t


(5)
Here, Rj,t and Rj,t-1 are own-stock returns, as before. These variables capture lag-zero and
lag-one momentum, respectively. The variable RLA,t is the contemporaneous return on
other countries within Latin American (i.e., excluding the local-market return). This
variable captures the cross-country contagion of the previous section. The fourth variable,
RLM,t, is the local-market index return. This variable does not enter the analysis
introduced in the previous sections, and is intended here as a control for country-level
systematic factors. The last variable, RUS,t, is the US-market index return. This variable
12
also does not enter in the previous sections, and is intended here as a control for
systematic US factors, which have well established effects on emerging equity markets.
13
III. Data
Our data on mutual-fund holdings come from two sources. The first source is the
US Securities and Exchange Commission (SEC). Mutual funds are required to report
holdings to the SEC twice a year. The second source is Morningstar. Morningstar
conducts surveys of mutual fund holdings at a higher frequency: quarterly surveys are the
norm for most funds. For our purposes, quarterly data are available from Morningstar for
about 50% of the funds we examine.
Our sample includes the holdings of 13 Latin America equity funds (open-end)
from April 1993 to January 1999 (24 quarters). Those funds are (1) Fidelity Latin
America, (2) Morgan Stanley Dean Witter Institutional Latin America, (3) Van Kampen
Latin America (formerly Morgan Stanley), (4) BT Investment Latin America Equity, (5)
TCW Galileo Latin America Equity, (6) TCW/Dean Witter Latin America Growth, (7)
Excelsior Latin America, (8) Govett Latin America, (9) Ivy South America, (10) Scudder
Latin America, (11) T. Rowe Price Latin America, (12) Merrill Lynch Latin America, and
(13) Templeton Latin America. Not all of these funds existed from the beginning of our
sample; on average we have about 10 quarters of data (out of a possible 24) per fund.
Our third source of data is Bloomberg. Bloomberg provides monthly price series
for all equities held by the 13 funds, including ADRs. (The need for monthly price data
arises in our analysis of lag-one momentum.) These price series are corrected for splits
and dividends. All return data are expressed in percent.
14
IV. Results: Momentum and Contagion
We present our results in four parts. First, we present some aggregate evidence
from our sample regarding the trades of mutual funds in times of crisis. Next we present
results on within-country momentum. We follow these with results on cross-country
contagion. Finally, we present regression-based results that integrate momentum and
contagion with other determinants of trading strategy.
IV.1. Aggregate Evidence Mutual Funds in Crisis Episodes
Our sample shows that during the fourth quarter of 1997—the peak of the Asian
crisis—Latin American funds suffered large outflows (Figure 1).9 The reversal from
inflows to outflows during the Asian and Russian crises is more severe than that during
the Mexican crisis in December 1994. In the Mexican crisis, mutual funds tended to pull
out of Mexico, Argentina, and Brazil, all of which are relatively liquid; funds tended not
to pull out from more illiquid markets, such as Colombia. Moreover, the Mexico-induced
pullout was temporary—by the third quarter of 1995 mutual fund inflows to Latin
America had resumed (consistent with the findings of Marcis et al. 1995 and Rea 1996).
Relative to the Mexican crisis, the Asian and Russian crises of 1997 and 1998 were more
broad-based and persistent. In those crises the retreat from Latin America was more
indiscriminate, with heavy sales reaching even the most illiquid markets. On average, net
sales in 1998 were about 32 percent. Note that this result is different from that of Froot et
al. (1998), who find little evidence of net outflows during the Asian crisis. A possible
explanation is that the aggregated data used by Froot et al. include institution types, such
as hedge funds, that counteract the clear net selling by mutual funds.
IV.2. Momentum Results
In our full sample, we find strong evidence of lag-zero momentum, both at the
9
Net selling in Figure 1 is calculated as the change in number of shares—as a percentage of average shares
held during the quarter—valued at the price at the beginning of each quarter. The average number of shares
held during the quarter is the mean of the beginning- and end-of-quarter holdings.
15
whole-fund level and at the manager-only level (Table 1).10 Moreover, our point estimate
for whole-fund momentum is larger than that for manager-only momentum, which is
consistent with underlying investors following positive-momentum strategies. Significant
lag-one momentum comes through only in the non-crisis portion of our sample.
Interestingly, it is the contemporaneous momentum that comes through especially
strongly during crises.
Tables 2 and 3 show that Sell L0M is positive and significant, while Buy L0M is
insignificant. This asymmetry holds for the full sample as well as for each of the two subsamples. When we move from L0M to L1M the asymmetry reverses—Buy L1M tends to
be positive and significant, whereas Sell L1M is insignificant at the five-percent level.
Grinblatt et al. (1995) also find an asymmetric result, though theirs is the same for L0M
and L1M—for both measures their Buy side is positive and significant but their Sell side
is insignificant.
Table 4 presents results for our momentum measures within separate crisis-period
sub-samples. The interesting question here is whether is momentum equally strong across
different crises. We find that the answer is no. Within our Latin American sample, we
find that positive momentum was strongest during the 1994 Mexican Crisis.
IV.3. Contagion Results
Tables 5 and 6 present our contagion results. In general, we find more significance
at the whole-fund level than at the manager-only level, an indication of contagion
strategies on the part of underlying fund investors. Of the five different return
benchmarks, Russia and Mexico appear to be the strongest—funds are systematically
buying equities in other countries when these countries’ returns are high, and vice versa.
Notice too the negative contemporaneous relation with U.S. equity returns.
IV.4. Regression-Based Results
10
For robustness we estimate each cell based only on observations within three standard deviations of the
mean. Using all observations increases both point estimates and t-statistics.
16
Recall that our regression-based approach provides a mean of testing the joint
significance of the momentum and contagion strategies we document in tables 1 through
6. Tables 7 through 10 present OLS estimation of the model introduced in Section II:
 Q j ,t . Q j ,t . 1 

 > β , χ 1 R j ,t , χ 2 R j ,t . 1 , χ 3 R LA,t , χ 4 R LM ,t , χ 5 RUS ,t , φ j ,t ,


Q
j ,t


(5)
where Rj,t and Rj,t-1 are own-stock returns, RLA,t is the contemporaneous return on other
countries within Latin American, RLM,t is the local-market index return, and RUS,t is the
US-market index return.
At the whole-fund level, we find that the full sample generates positive and
significant coefficients on all of the first three variables. Put differently, L0M, L1M, and
L0C are all robust to moving from bivariate measures to multivariate measures, and
including controls for the overall local and US markets. Interestingly, the local-market
control is never significant. The US-market control, in contrast, is quite significant, and
negative. This squares with past empirical work that shows that US investors tend to
chase emerging markets when returns at home are low. When the sample is split into
crisis and non-crisis sub-periods, we find that the contagion strategy—L0C—is largely a
crisis-period phenomenon.
V. Rationalizing Momentum Trading: Return Autocorrelation
In an environment with positively autocorrelated returns, momentum trading is a
natural response. Section IV presented evidence of positive L0M and, at least during noncrisis periods, positive L1M. This raises the question of whether returns within Latin
America exhibit positive autocorrelation. One common way to test for return
autocorrelation is using variance ratios. If returns follow a random walk, then return
variance is a linear function of horizon length. That is, the variance of returns over k
periods is k times the variance of returns over one-period. If instead returns are positively
autocorrelated, the variance of k-period returns is larger than the sum of one-period
17
returns—variances grow faster than linearly. Thus, variance ratios larger than one are
consistent with rational positive momentum trading. Alternatively, when returns are
negatively autocorrelated, the variance of k-period returns is smaller than k times the
variance of one-period returns. Variance ratios smaller than one would call for negative
momentum (or contrarian) trading.
Table 11 reports the values of the variance-ratio test statistic at different horizons,
together with p-values, for seven Latin American countries. For comparison we also
provide results for the U.S. stock market.11 Interestingly, stock returns in several Latin
American markets are highly persistent (variance-ratio statistics larger than one), even at
three and four-year horizons. In sharp contrast, U.S. returns show no persistence at any
horizon. Though certainly not proof that the positive momentum trading we find in Latin
America is rational—after all, this persistence in returns is at the index level—these
results do point to the possibility of rationalizing our momentum results, at least for some
countries (e.g., Mexico, Chile, Colombia, and Venezuela).
VI. Conclusion
Discriminating among the various ways that financial markets can spread crisis
requires a sharper picture of actual activity in international financial markets. Who is
doing the trading? What are their trading strategies? These are questions we address in
this paper. In particular, we examine the portfolios of an important class of international
investor—US mutual funds. Many observers have claimed that fund strategies help
spread crisis. For example, some claim that collapse in Thailand induced funds to sell in
other Asian countries to meet redemptions, or to rebalance holdings across countries. This
type of spillover was discussed widely following the 1994 crisis in Mexico.
We address two sets of questions. The first relates to whether and when these
funds use momentum strategies—systematically buying winning stocks and selling losing
stocks. We find that international funds do use momentum strategies. Their strategies
exhibit positive momentum, due to momentum both at the investor level—through
11
See Campbell, Lo, and MacKinlay (1997) for the asymptotic distribution of the variance-ratio test.
18
redemptions/inflows—and at the fund manager level. Funds also use momentum
strategies in both crisis and non-crisis periods, though contemporaneous momentum
(buying current winners and selling current losers) is stronger during crises, whereas
lagged momentum (buying past winners and selling past losers) is stronger during noncrisis periods. We also find that contemporaneous momentum is stronger on the sell side,
whereas lagged momentum is stronger on the buy side.
The second set of questions we address relates to whether funds use contagion
strategies—selling assets from one country when crisis hits another. We find that funds
do use contagion strategies. This result is robust to controlling for own-stock returns, the
local-market factor, and the US-market factor. Strictly speaking, these controls are not
sufficient to conclude that this contagion is non-fundamental (or pure) contagion. But
these controls are quite sound on a theoretical basis. In our judgment, then, the inference
that the contagion we find is non-fundamental is on firmer footing than the inference that
the contagion is driven by fundamentals. In any event, we have uncovered several stylized
facts that are useful for evaluating hypotheses about the emerging-market crises and their
transmission.
Beyond these stylized facts, this paper also includes several methodological
innovations. For example, the distinction between manager-only and whole-fund
momentum is new to the literature, as is our method for distinguishing the two. Our
method of measuring contagion via transaction quantities is also new. Finally, our
regression-based approach to integrating momentum and contagion measures with
systematic return factors provides a more robust picture of the trading strategies we set
out to understand.
19
Figure 1: Mutual Funds' Net Buying/Selling of Stocks in Latin America
Argentina
Mexico
60.00
40.00
40.00
20.00
20.00
0.00
31-20.00Jan
-93
-40.00
15Jun
-94
28Oct
-95
11Ma
r97
24Jul
-98
6De
c99
0.00
31-20.00Jan
-93
-40.00
15Jun
-94
28Oct
-95
11Ma
r97
24Jul
-98
6De
c99
11Ma
r97
24Jul
-98
6De
c99
11Ma
r97
24Jul
-98
6De
c99
11Ma
r97
24Jul
-98
6De
c99
-60.00
-60.00
-80.00
-80.00
-100.00
-100.00
-120.00
Brazil
Peru
80.00
60.00
60.00
40.00
40.00
20.00
0.00
31Jan
-20.00-93
20.00
15Jun
-94
28Oct
-95
11Ma
r97
24Jul
-98
6De
c99
-40.00
0.00
31-20.00Jan
-93
-40.00
15Jun
-94
28Oct
-95
-60.00
-60.00
-80.00
-80.00
-100.00
Chile
Venezuela
80.00
40.00
60.00
20.00
40.00
0.00
31Jan
-20.00
-93
20.00
15Jun
-94
28Oct
-95
-40.00
0.00
31Jan
-20.00-93
15Jun
-94
28Oct
-95
11Ma
r97
24Jul
-98
6De
c99
-40.00
-60.00
-80.00
-100.00
-60.00
-120.00
Colombia
Latin America
60.00
30.00
20.00
40.00
10.00
20.00
0.00
0.00
-20.00
31
Ja
15
Ju
28
O
t
11
M
ar
24
Ju
6D
ec
31Jan
-10.00-93
15Jun
-94
28Oct
-95
-20.00
-40.00
-30.00
-60.00
-40.00
Source: our data. Net Buying/Selling is equal to the value-weighted percentage change in quarterly holdings of all funds in each country, where
the value weighting uses the beginning-of-period share price. All figures in percent.
Table 1
Lag-0 and Lag-1 Momentum
All Sample
Non-Crisis
Crisis
L0M
T-statistic
Observations
2.35***
5.64
4,925
0.97***
3.15
3,289
5.13***
4.55
1,636
L1M
T-statistic
Observations
0.20
1.56
4,853
0.25**
2.42
3,215
0.11
0.40
1,638
L0M
T-statistic
Observations
0.87**
2.88
4,931
0.30
1.32
3,289
2.01**
2.68
1,642
L1M
T-statistic
Observations
0.16
1.613
4,850
0.23**
2.20
3,211
0.11
0.61
1,639
Whole Fund
Manager Only
L0M is the point estimate for the mean lag-0 momentum measure. L1M is the point estimate for the mean lag-1 momentum measure
(measured from return over the previous month). Whole-Fund momentum tests whether the mean of (ΕQjt/Qjt)Rjt-k is zero. ManagerOnly momentum controls for investor redemption effects as in equation (2). All t-statistics are corrected for heteroskedasticity across
funds. Full sample: quarterly data from April 1993 to January 1999. The crisis portion of the sample includes December 1994-June
1995, July 1997-March 1998, August 1998-October 1998, and January 1999. The non-crisis portion is the rest of the sample. The
total of roughly 4400 observations is13 funds times an average of about 35 stocks per fund, times an average of about 10 quarters of
available data per fund. For robustness, results in each cell are based only on observations within three standard deviations of the
mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
Table 2
Buy and Sell Momentum: Whole Fund
Statistic
All Sample
Non-Crisis
Crisis
Buy L0M
T-statistic
Observations
-0.24
-0.91
2,234
-0.46
-1.45
1,645
0.36
0.75
589
Sell L0M
T-statistic
Observations
2.74***
6.89
2,585
2.10***
4.35
1,555
3.70***
3.77
1,030
Buy L1M
T-statistic
Observations
0.56**
2.31
2,211
0.35
1.41
1,623
1.13*
1.85
588
Sell L1M
T-statistic
Observations
-0.25
-0.37
2,541
-0.02
-0.05
1,521
-0.58
-0.42
1,020
These results are "Whole Fund" in that they include no control for investor redemption effects as in equation (2). Buy L0M is the
point estimate for the mean of lag-0 momentum measured from purchases only. Sell L0M is the point estimate for the mean of lag-0
momentum measured from sales only. L1M denotes lag-1 momentum (measured from return over the previous month). T-statistics for
the test that the mean equals zero are corrected for heteroskedasticity across funds. Full sample: April 1993 to January 1999. The
crisis portion of the sample includes December 1994-June 1995, July 1997-March 1998, August 1998-October 1998, and January
1999. The non-crisis portion is the rest of the sample. For robustness, results in each cell are based only on observations within three
standard deviations of the mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
Table 3
Buy and Sell Momentum: Manager Only
Statistic
All Sample
Non-Crisis
Crisis
Buy L0M
T-statistic
Observations
-0.38*
-1.78
2,415
-0.49
-1.71
1,559
-0.17
-0.48
856
Sell L0M
T-statistic
Observations
2.78***
5.58
2,508
1.97***
4.27
1,708
4.51***
3.80
800
Buy L1M
T-statistic
Observations
0.54*
1.92.
2,387
0.40*
1.573
1,537
0.80
1.08
850
Sell L1M
T-statistic
Observations
-0.19
-0.41
2,461
0.03
0.075
1,666
-0.65
-0.54
795
These results are "Manger Only" in that they control for investor redemption effects as in equation (2). Buy L0M is the point estimate
for the mean of lag-0 momentum measured from purchases only. Sell L0M is the point estimate for the mean of lag-0 momentum
measured from sales only. L1M denotes lag-1 momentum (measured from return over the previous month). T-statistics for the test
that the mean equals zero are corrected for heteroskedasticity across funds. Full sample: April 1993 to January 1999. The crisis
portion of the sample includes December 1994-June 1995, July 1997-March 1998, August 1998-October 1998, and January 1999.
The non-crisis portion is the rest of the sample. For robustness, results in each cell are based only on observations within three
standard deviations of the mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
Table 4
Momentum Results by Crisis
Mexican Crisis
Asian Crisis
Russian Crisis
Whole Fund
L0M
T-statistic
Observations
12.11***
3.45
268
1.69***
2.97
920
8.26***
4.24
417
L1M
T-statistic
Observations
1.00*
1.82
297
-0.25
-0.69
898
0.22
0.57
413
L0M
T-statistic
Observations
6.56**
2.16
279
0.99**
2.32
920
1.04
1.15
412
L1M
T-statistic
Observations
1.00***
2.71
297
-0.17
-0.75
898
-0.04
-0.21
414
Manager Only
L0M is the point estimate for the mean lag-0 momentum measure. L1M is the point estimate for the mean lag-1 momentum measure
(measured from return over the previous month). Whole-Fund momentum tests whether the mean of (ΕQjt/Qjt)Rjt-k is zero. ManagerOnly momentum controls for investor redemption effects as in equation (2). All t-statistics are corrected for heteroskedasticity across
funds. The Mexican Crisis portion of the sample includes December 1994-June 1995. The Asian Crisis portion of the sample includes
July 1997-March 1998. The Russian Crisis portion of the sample includes August 1998-October 1998. For robustness, results in each
cell are based only on observations within three standard deviations of the mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
Table 5
Contagion Measure Results
Country/Regional Index
Brazil
Statistic
Whole Fund
L0C
T-statistic
Manager
Only
L0C
T-statistic
All
Sample
NonCrisis
Mexico
Crisis
All
Sample
NonCrisis
U.S.
Crisis
All
Sample
NonCrisis
Asia
Crisis
All
Sample
NonCrisis
Russia
Crisis
All
Sample
NonCrisis
Crisis
1.75***
3.77
0.60
1.11
4.01***
3.09
1.07**
2.55
0.42
1.00
2.37**
2.25
-0.58***
-2.81
-0.25
-1.07
-1.26***
-3.08
0.72**
2.23
0.38***
2.69
1.38
1.38
3.91***
3.06
2.49
1.39
6.18***
2.73
0.10
0.28
-0.56
-0.95
1.39***
3.15
0.19
0.80
-0.11
-0.48
0.77
1.36
-0.50***
-3.61
-0.50***
-2.89
-0.51***
-2.84
0.53**
2.20
0.11
0.90
1.36**
2.15
-0.59
-0.66
-1.25
-1.04
0.46
0.29
L0C denotes lag-0 contagion. Whole-Fund contagion tests whether the mean of (ΕQjt/Qjt)Rct is zero, where Rct is the return on country index c from t-1 to t, with c∉{Brazil, Mexico, U.S., Asia,
Russia}. Manager-Only contagion controls for investor redemption effects as in equation (2). All t-statistics are corrected for heteroskedasticity across funds. Full sample: April 1993 to January 1999.
The crisis portion of the sample includes December 1994-June 1995, July 1997-March 1998, August 1998-October 1998, and January 1999. The non-crisis portion is the rest of the sample. Asia is the
IFC Asia Stock Market Index. Note that Brazilian equities are excluded from the calculation of L0C for Brazil (similarly for Mexico).
Table 6
Contagion Measure Results by Crisis
Country/Regional Index
Brazil
Statistic
Whole-Fund
L0C
T-statistic
Manager
Only
L0C
T-statistic
Mexican
Crisis
Asian
Crisis
Mexico
Russian
Crisis
Mexican
Crisis
Asian
Crisis
U.S.
Russian
Crisis
Mexican
Crisis
Asian
Crisis
Asia
Russian
Crisis
Mexican
Crisis
Asian
Crisis
Russia
Russian
Crisis
Mexican
Crisis
Asian
Crisis
Russian
Crisis
11.25**
2.25
0.44
0.54
7.41***
5.04
10.58**
*
3.41
-0.64
-1.01
3.36**
2.53
-3.59***
-3.71
-0.90**
-2.23
-0.22
-0.33
4.84***
3.27
0.27
0.40
1.70
0.88
N/A
0.06
0.05
22.03***
3.55
4.56***
4.44
0.40
0.82
1.59
1.33
4.87***
5.69
-0.47
-1.02
0.61
1.11
-1.73***
-3.91
-0.42
-1.48
0.10
0.86
2.26***
3.37
1.66
1.35
0.45
0.63
N/A
-1.70**
-2.42
6.03
1.18
L0C denotes lag-0 contagion. Whole-Fund contagion tests whether the mean of (ΕQjt/Qjt)Rct is zero, where Rct is the return on country index c from t-1 to t, with c∉{Brazil, Mexico, U.S., Asia,
Russia}. Manager-Only contagion controls for investor redemption effects as in equation (2). All t-statistics are corrected for heteroskedasticity across funds. The Mexican Crisis portion of the sample
includes December 1994-June 1995. The Asian Crisis portion of the sample includes July 1997-March 1998. The Russian Crisis portion of the sample includes August 1998-October 1998. Asia is the
IFC Asia Stock Market Index. Note that Brazilian equities are excluded from the calculation of L0C for Brazil (similarly for Mexico).
Table 7
Regression Results: Whole Fund
 Q j ,t . Q j ,t . 1 

 > β , χ 1 R j ,t , χ 2 R j ,t . 1 , χ 3 R LA,t , χ 4 R LM ,t , χ 5 RUS ,t , φ j ,t


Q
j
,
t


Independent Variables
L0M (χ1)
All Sample
Non-Crisis
Crisis
0.0021***
5.34
0.0029***
4.04
0.0015***
2.76
0.0029***
3.15
0.0042***
5.501
0.0002
0.138
0.0035***
3.20
0.0021
1.49
0.0041***
2.62
0
-0.01
-0.0003
-0.32
0.0006
0.44
T-statistic
-0.0065***
-6.24
-0.0041*
-1.95
-0.0096***
-4.54
Constant
T-statistic
-0.0048
-0.24
-0.0086
-0.30
0.0026
0.05
4,842
0.05
3,223
0.03
1,619
0.06
T-statistic
L1M (χ2)
T-statistic
L0C (χ3)
T-statistic
Local Index Return (χ4)
T-statistic
US Return (χ5)
Observations
Adjusted R-squared
These results are "Whole Fund" in that they include no control for investor redemption effects as in equation (2). T-statistics are
corrected for heteroskedasticity across funds. Full sample: April 1993 to January 1999. The crisis portion of the sample includes
December 1994-June 1995, July 1997-March 1998, August 1998-October 1998, and January 1999. The non-crisis portion is the rest
of the sample. For robustness, results in each cell are based only on observations within three standard deviations of the mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
28
Table 8
Regression Results by Crisis: Whole Fund
 Q j ,t . Q j ,t . 1 

 > β , χ 1 R j ,t , χ 2 R j ,t . 1 , χ 3 R LA,t , χ 4 R LM ,t , χ 5 RUS ,t , φ j ,t


Q
j
,
t


Independent Variables
L0M (χ1)
Mexican Crisis
Asian Crisis
Russian Crisis
0.0030***
6.63
0.0016
1.22
0.0005
0.80
0.0022
0.67
0.0015
0.55
-0.0027*
-1.96
0.0112
1.74
0.002***
3.62
-0.0123***
-3.88
-0.0014
0.42
-0.0001
-0.03
0.0026
1.68
T-statistic
-0.0050
-0.46
-0.0160***
-4.02
0.0104
1.69
Constant
T-statistic
0.2972
0.86
0.0895***
3.28
-0.5570***
-4.64
Observations
Adjusted R-squared
297
0.13
885
0.05
406
0.05
T-statistic
L1M (χ2)
T-statistic
L0C (χ3)
T-statistic
Local Index Return (χ4)
T-statistic
US Return (χ5)
These results are "Whole Fund" in that they include no control for investor redemption effects as in equation (2). T-statistics are
corrected for heteroskedasticity across funds. The Mexican Crisis portion of the sample includes December 1994-June 1995. The
Asian Crisis portion of the sample includes July 1997-March 1998. The Russian Crisis portion of the sample includes August 1998October 1998. For robustness, results in each cell are based only on observations within three standard deviations of the mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
29
Table 9
Regression Results: Manger Only

 Q j ,t . Q j ,t . 1
.

Q j ,t


 )Q
j
Independent Variables
L0M (χ1)
j ,t

j
. Q j ,t . 1 ∗Pj ,t 

 > β , χ 1 R j ,t , χ 2 R j ,t . 1 , χ 3 R LA,t , χ 4 R LM ,t , χ 5 RUS ,t , φ j ,t
Q j , t Pj , t


All Sample
Non-Crisis
Crisis
0.0042***
5.56
0.0052***
3.76
0.0033***
2.76
0.0049***
2.78
0.0071***
4.04
0.0004
0.17
0.0014
1.17
-0.0001
-0.44
0.004***
3.46
-0.0004
-0.328
-0.0011
-0.55
0.0011
0.48
T-statistic
-0.0115***
-6.54
-0.0063***
-2.90
-0.0196***
-7.018
Constant
T-statistic
0.0108
0.326
-0.0155
-0.40
0.0970**
2.53
Observations
Adjusted R-squared
4,942
0.03
3,274
0.02
1,668
0.05
T-statistic
L1M (χ2)
T-statistic
L0C (χ3)
T-statistic
Local Index Return (χ4)
T-statistic
US Return (χ5)
These results are "Manger Only" in that they control for investor redemption effects as in equation (2). T-statistics are corrected for
heteroskedasticity across funds. Full sample: April 1993 to January 1999. The crisis portion of the sample includes December 1994June 1995, July 1997-March 1998, August 1998-October 1998, and January 1999. The non-crisis portion is the rest of the sample.
For robustness, results in each cell are based only on observations within three standard deviations of the mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
30
Table 10
Regression Results by Crisis: Manager Only

 Q j ,t . Q j ,t . 1
.

Q j ,t


 )Q
j
Independent Variables
j ,t

j
. Q j ,t . 1 ∗Pj ,t 

 > β , χ 1 R j ,t , χ 2 R j ,t . 1 , χ 3 RLA,t , χ 4 RLM ,t , χ 5 RUS ,t , φ j ,t
Q j , t Pj , t


Mexican Crisis
Asian Crisis
0.0058***
6.85
0.0040*
1.94
0.0012
0.95
0.0078
1.32
-0.0005
-0.16
0.0002
0.07
0.0103*
1.92
0.0078
1.15
0.0080*
1.88
-0.0025
-0.434
-0.0004
-0.09
0.0047
1.12
T-statistic
-0.0087
-0.82
-0.0218***
-4.22
-0.220***
-2.84
Constant
T-statistic
0.2959
1.07
0.9580*
1.93
0.309***
6.037
L0M (χ1)
T-statistic
L1M (χ2)
T-statistic
L0C (χ3)
T-statistic
Local Index Return (χ4)
T-statistic
US Return (χ5)
Observations
Adjusted R-squared
297
0.11
910
0.02
Russian Crisis
430
0.03
These results are "Manager Only" in that they control for investor redemption effects as in equation (2). T-statistics are corrected for
heteroskedasticity across funds. The Mexican Crisis portion of the sample includes December 1994-June 1995. The Asian Crisis
portion of the sample includes July 1997-March 1998. The Russian Crisis portion of the sample includes August 1998-October 1998.
For robustness, results in each cell are based only on observations within three standard deviations of the mean.
*
Statistically Significant at the 10-percent level
** Statistically Significant at the 5-percent level
*** Statistically Significant at the 1-percent level
31
Table 11: Variance Ratio Test of Stock Returns
Horizon
COUNTRY
3-months
Argentina
12-months
24-months
36-months
48-months
60-months
1.02
0.88
0.70
0.62
0.60
0.60
(0.84)
(0.59)
(0.35)
(0.35)
(0.39)
(0.45)
Brazil
1.01
0.99
0.82
0.71
0.75
0.83
(0.93)
(0.95)
(0.57)
(0.46)
(0.59)
(0.75)
Chile
1.34
1.94
2.50
2.88
3.16
2.97
(0.00)
(0.00)
(0.00)
(0.00)
(0.00)
(0.00)
1.43
2.22
2.40
2.63
2.76
2.81
(0.00)
(0.00)
(0.00)
(0.00)
(0.00)
(0.00)
1.31
1.50
1.61
1.74
1.85
1.84
(0.00)
(0.02)
(0.06)
(0.06)
(0.07)
(0.11)
1.07
0.80
0.64
0.70
0.82
0.58
(0.45)
(0.37)
(0.27)
(0.45)
(0.69)
(0.42)
1.15
1.59
1.53
1.10
0.97
0.88
(0.09)
(0.01)
(0.10)
(0.80)
(0.95)
(0.82)
Colombia
Mexico
Peru
Venezuela
USA
0.96
0.91
0.83
0.90
0.91
0.94
(0.64)
(0.69)
(0.60)
(0.81)
(0.84)
(0.91)
Note: P-values are shown in parentheses
32
References
Asness, C., J. Liew, and R. Stevens, 1997, “Parallels Between the Cross-Sectional
Predictability of Stock and Country Returns,” Journal of Portfolio Management, 3:
79-87.
Badrinath, S. and S. Wahal, 1998, “Momentum Trading by Institutions,” typescript,
Rutgers University, April.
Bank for International Settlements, 1998, 68th Annual Report, Basle, Switzerland.
Barth, M., and Z. Xin, 1999, “Foreign Equity Flows and the Asian Financial Crisis,”
typescript, World Bank, March.
Bodurtha, James, Dong-Soon Kim, and Charles Lee, 1995, "Closed-End Country Funds
and U.S. Market Sentiment," Review of Financial Studies, 8, 879-918.
Bohn, H., and L. Tesar, 1996, “US Equity Investment in Foreign Markets: Portfolio
Rebalancing or Return Chasing?” American Economic Review, 86: 77-81.
Brennan, M. and H. Cao, “International Portfolio Investment,” Journal of Finance, 52,
Dec. 1997.
Brown, K., W. Harlow, and L. Starks, 1996, "Of Tournaments and Temptations: An
Analysis of Managerial Incentives in the Mutual Fund Industry," Journal of
Finance, 51, 85-110.
Brown, S., W. Goetzmann, and J. Park, 1998, “Hedge Funds and the Asian Currency
Crisis of 1997,” NBER Working Paper 6427, February.
Calvo, G., 1996, “Varieties of Capital Market Crises,” typescript, University of Maryland.
Calvo, G., 1999, “Contagion in Emerging Markets: When Wall Street Is a Carrier,”
University of Maryland working paper.
Calvo, G., L. Leiderman, and C. Reinhart, 1994, International Monetary Fund Staff
Papers.
Calvo, G., and E. Mendoza, 1998, “Rational Herd Behavior and the Globalization of
Securities Markets, University of Maryland working paper.
Campbell, J., A. Lo, and C. MacKinlay, 1997, The Econometrics of Financial Markets,
Princeton University Press, Princeton, New Jersey.
Carhart, Mark, 1997, "On Persistence in Mutual Fund Performance," Journal of Finance,
52, 57-82.
Chan, L., N. Jegadeesh, and J. Lakonishok, 1996, "Momentum Strategies," Journal of
Finance, 51, 1681-1714.
Choe, H., B. Kho, and R. Stulz, 1999, “Do Foreign Investors Destabilize Stock Markets?
The Korean Experience in 1997,” typescript, Ohio State University, January.
Corsetti, G., P. Pesenti, and N. Roubini, 1998, “What Caused the Asian Currency and
Financial Crisis?” typescript, New York University, March.
Corsetti, G., P. Pesenti, N. Roubini, and C. Tille, 1998, “Structural Links and Contagion
Effects in the Asian Crisis: A Welfare Based Approach,” New York University
working paper.
Coval, J. and F. Sturzenegger, 1998, “Can we build a Story to Justify Tobin Taxes?”
typescript, University of Michigan, Business School.
34
De Bondt, W., and R. Thaler, 1985, “Does the Stock Market Overreact?” Journal of
Finance, 40: 793-805.
De Long, J. B., A. Shleifer, L. Summers, and R. Waldmann, 1990, “Positive Feedback
Investment Strategies and Destabilizing Rational Speculation,” Journal of Finance,
45, 379-396.
Edelen, R., and J. Warner, 1999, “Why are Mutual Fund Flow and Market Returns
Related? Evidence from High-Frequency Data,” typescript, University of Rochester.
Eichengreen, B., and D. Mathieson, 1998, “Hedge Funds and Financial Market
Dynamics,” Occasional Paper No. 166.
Eichengreen, B, A. Rose, and C. Wyplosz, 1996, “Contagious Currency Crises,” NBER
working paper No. 5681.
Elton, Edwin, Martin Gruber, and Christopher Blake, 1996, "Survivorship Bias and
Mutual Fund Performance," Review of Financial Studies, 9, 1097-1120.
Forbes, K., and R. Rigobon, 1998, “No Contagion, Only Interdependence: Measuring
Stock Market Co-movements,” typescript, MIT Sloan School, November.
Frankel, J. and S. Schmukler, 1996, “Country Fund Discounts and the Mexican Crisis of
December 1994: Did Local Residents Turn Pessimistic Before International
Investors?” Open Economies Review, Vol. 7.
Frankel, J. and S. Schmukler, 1998a, “Country Funds and Asymmetric Information,”
World Bank Working Paper #1493.
Frankel, J., and S. Schmukler, 1998b, “Crisis, Contagion, and Country Funds,” in R.
Glick, ed., Managing Capital Flows and Exchange Rates (Cambridge University
Press).
Froot, K., D. Scharfstein, J., and J. Stein, 1992, “Herd on the Street: Informational
Inefficiencies in a Market with Short-Term Speculation,” Journal of Finance, 47,
1461-1484.
Froot, K., P. O’Connell, and M. Seasholes, 1998, “The Portfolio Flows of International
Investors, I,” typescript, April.
Fung, W. and D. Hsieh, 1997, “Empirical Characteristics of Dynamic Trading Strategies:
The Case of Hedge Funds,” Review of Financial Studies, 10, 2.
Glick, R., and A. Rose, 1998, “Contagion and Trade: Why are Currency Crises
Regional?” NBER Working Paper 6806, November.
Grinblatt, M. and S. Titman, 1989, “Mutual Fund Performance: An Analysis of Quarterly
Portfolio Holdings,” Journal of Business, 62, 3.
Grinblatt, M. and S. Titman, 1993, “Performance Measurement without Benchmarks: An
Examination of Mutual fund Returns,” Journal of Business, 66, 1.
Grinblatt, M. and S. Titman, 1994, “A Study of Monthly Mutual Fund Returns and
Performance Evaluation Techniques,” Journal of Financial and Quantitative
Analysis, 29, 3.
Grinblatt, M., S. Titman, and R. Wermers, 1995, “Momentum Investment Strategies,
Portfolio Performance, and Herding: A Study of Mutual Fund Behavior,” American
Economic Review, 85, 1088-1105.
International Monetary Fund, 1995, Capitals Market Report, August.
Jegadeesh, N., 1990, “Evidence of Predictable Behavior of Security Returns,” Journal of
Finance, 45: 881-898.
35
Jegadeesh, N. and S. Titman, 1993, “Returns to Buying Winners and Selling Losers:
Implications for Stock Market Efficiency,” Journal of Finance, 48, 1.
Kaminsky, G., 1998, “Currency and Banking Crises: The Early Warnings of Distress,”
International Finance Discussion Paper No. 629, Board of Governors of the Federal
Reserve System.
Kaminsky, G. and C. Reinhart, 1999, “On Crises, Contagion, and Confusion,” George
Washington University Working Paper, forthcoming Journal of International
Economics.
Kaminsky, G. and S. Schmukler, 1999, “What Triggers Market Jitters? A Chronicle of
the Asian Crisis,” Journal of International Money and Finance, Vol. 18, No. 4.
Kaufman, H., 1994, “Structural Changes in the Financial Markets: Economic and Policy
Significance,” Economic Review, Federal Reserve Bank of Kansas City, Second
Quarter.
Kim, W., and S. Wei, 1999, “Foreign Portfolio Investors Before and During a Crisis,”
NBER Working Paper No. 6968, February.
Klibanoff, Peter, Owen Lamont, and Thierry Wizman, 1998, "Investor Reaction to Salient
News in Closed-End Country Funds," Journal of Finance, 53, 673-699.
Kodres, L., and M. Pritsker, 1997, “Directionally-Similar Position Taking and Herding by
Large Futures Market Participants,” typescript, International Monetary Fund and
Board of Governors of the Federal Reserve System.
Kodres, L., and M. Pritsker, 1999, "A Rational Expectations Model of Financial
Contagion," typescript, International Monetary Fund and Board of Governors of the
Federal Reserve System, May.
Kraus, A and H. Stoll, 1972, “Parallel Trading by Institutional Investors,” Journal of
Financial and Quantitative Analysis,” 7, 5.
Lakonishok, J., A. Shleifer, R. Thaler, and R. Vishny, 1991, “Window Dressing by
Pension Fund Managers,” American Economic Review (Papers and Proceedings),
81, 2.
Lakonishok, J., A. Shleifer, R. Thaler, and R. Vishny, 1992, “The Impact of Institutional
Trading on Stock Prices,” Journal of Financial Economics, 32, 1.
Marcis, R., S. West, and V. Leonard-Chambers, 1995, “Mutual Fund Shareholder
Response to Market Disruptions,” Perspective, Investment Company Institute, Vol.
1, No. 1.
Poterba, J. and L. Summers, 1988, “Mean Reversion in Stock Returns: Evidence and
Implications,” Journal of Financial Economics, 22, 27-60.
Rea, J., 1996, “U.S. Emerging Market Funds: Hot Money or Stable Source of Investment
Capital?” Perspective, Investment Company Institute, Vol. 2, No. 6.
Rigobon, R., 1998, “Informational Speculative Attacks: Good News is No News,”
typescript, MIT, January.
Rouwenhorst, Geert, 1998, "International Momentum Strategies," Journal of Finance, 53,
267-284.
Scharfstein, D. and J. Stein, 1990, “Herd Behavior and Investment,” American Economic
Review, 80, 3.
36
Tesar, L., and I. Werner, 1994, “International Equity Transactions and US Portfolio
Choice,” in J. Frankel (ed.), The Internationalization of Equity Markets, University
of Chicago Press, 185-220.
Valdes, Rodrigo, 1998, “Emerging Markets Contagion: Evidence and Theory,” Banco
Central de Chile, typescript, May.
Warther, V., 1995, “Aggregate Mutual Fund Flows and Security Returns,” Journal of
Financial Economics, 39: 209-235.
Wermers, R., 1999, “Mutual Fund Herding and the Impact on Stock Prices,” Journal of
Finance, 54: 581-622.
World Bank, 1997, Private Capital Flows to Developing Countries, World Bank Policy
Research Report.
World Bank, 1998, East Asia: The Road to Recovery, Washington, DC.
World Bank, 1998, Global Development Finance, Washington, DC.
World Bank, 1998, Global Economic Perspectives, Washington, DC.
37