Andreas Vester
Theories of Contagion
The Role of International Portfolio Flows
ISBN-10: 3-8324-9873-0
ISBN-13: 978-3-8324-9873-9
Druck Diplomica® GmbH, Hamburg, 2006
Zugl. Universität Duisburg-Essen, Standort Duisburg, Duisburg, Deutschland,
Diplomarbeit, 2006
Dieses Werk ist urheberrechtlich geschützt. Die dadurch begründeten Rechte,
insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von
Abbildungen und Tabellen, der Funksendung, der Mikroverfilmung oder der
Vervielfältigung auf anderen Wegen und der Speicherung in Datenverarbeitungsanlagen,
bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten. Eine Vervielfältigung
dieses Werkes oder von Teilen dieses Werkes ist auch im Einzelfall nur in den Grenzen
der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik
Deutschland in der jeweils geltenden Fassung zulässig. Sie ist grundsätzlich
vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des
Urheberrechtes.
Die Wiedergabe von Gebrauchsnamen, Handelsnamen, Warenbezeichnungen usw. in
diesem Werk berechtigt auch ohne besondere Kennzeichnung nicht zu der Annahme,
dass solche Namen im Sinne der Warenzeichen- und Markenschutz-Gesetzgebung als frei
zu betrachten wären und daher von jedermann benutzt werden dürften.
Die Informationen in diesem Werk wurden mit Sorgfalt erarbeitet. Dennoch können
Fehler nicht vollständig ausgeschlossen werden, und die Diplomarbeiten Agentur, die
Autoren oder Übersetzer übernehmen keine juristische Verantwortung oder irgendeine
Haftung für evtl. verbliebene fehlerhafte Angaben und deren Folgen.
© Diplomica GmbH
http://www.diplom.de, Hamburg 2006
Printed in Germany
Abstract
After the Mexican, Asian, and Russian crisis, the phenomenon of contagion
became increasingly important. Existing studies indicate that various explanations for
the transmission of crises exist. This paper gives an overview over theories that try to
explain contagion caused by portfolio flows of international investors. Theories such as
the occurrence of information cascades, the effects of international portfolio
diversification and optimization, the importance of information asymmetries, cross-
market re-balancing effects, risk aversion, and wealth effects are discussed in detail.
The analysis suggests that information asymmetries and changes in risk aversion hold
an important role in explaining contagious sellouts.
II
Table of contents
Page
INDEX OF FIGURES...IV
INDEX OF TABLES... V
INDEX OF SYMBOLS...VI
INTRODUCTION... 1
PART I: OVERVIEW ... 3
1
DEFINING CONTAGION ... 3
2
SPILLOVER EFFECTS AND FINANCIAL LINKAGES ... 4
2.1
B
ILATERAL AND THIRD PARTY TRADE
... 4
2.2
T
HE COMMON BANK LENDER
/
COMMON CREDITOR EFFECT
... 6
PART II: BASIC APPLICATIONS ... 8
3
HERD BEHAVIOR AND INFORMATION CASCADES ... 8
3.1
D
EFINING HERD BEHAVIOR AND INFORMATION CASCADES
... 8
3.2
A
BASIC MODEL OF INFORMATION CASCADES
... 9
4
PORTFOLIO DIVERSIFICATION AND OPTIMIZATION... 14
4.1
P
ORTFOLIO MANAGEMENT RULES
... 14
4.2
V
OLATILITY EVENTS
... 16
4.3
C
APITAL EVENTS
... 17
PART III: INFORMATION ASYMMETRIES ... 21
5
SIGNAL-EXTRACTION PROBLEMS AND INVESTOR PANICS ... 21
5.1
T
HE EMERGENCE OF INFORMATION ASYMMETRIES
... 21
5.2
T
HE MODELS
... 22
5.3
R
ATIONAL PANICS DUE TO INFORMATION ASYMMETRIES
... 24
6
CROSS-MARKET RE-BALANCING AND CROSS-HEDGING ... 25
6.1
T
HE SET UP OF THE GENERAL EQUILIBRIUM MODEL
... 25
6.2
T
HE CROSS
-
MARKET RE
-
BALANCING CHANNEL OF CONTAGION
... 27
6.3
O
N THE SEVERITY OF CONTAGION
... 29
III
PART IV: WEALTH EFFECTS AND RISK AVERSION ... 32
7
BENCHMARK TRACKING AND RISK AVERSION ... 32
7.1
M
ODEL ASSUMPTIONS
... 32
7.2
D
EMAND
... 33
7.3
I
NELASTIC SUPPLY
... 34
7.4
E
LASTIC SUPPLY
... 36
8
PORTFOLIO LINKAGES ... 38
8.1
T
HE CHAIN STRUCTURE
... 39
8.2
C
ONTAGION
... 40
9
WEALTH EFFECTS IN A UNIQUE EQUILIBRIUM MODEL ... 42
9.1
T
HE MODEL
'
S ASSUMPTIONS
... 43
9.2
U
NIQUE EQUILIBRIUM IN THE MODEL
... 45
9.3
T
HE WEALTH EFFECT
,
CONTAGION
,
AND CORRELATION
... 49
9.4
R
EAL
-
WORLD APPLICATION AND EMPIRICAL EVIDENCE
... 52
SUMMARY & CONCLUSION ... 54
REFERENCES ...IX
APPENDIX
IV
Index of figures
Page
F
IGURE
1:
B
INARY DECISION TREE WITH THREE INVESTORS
... 13
F
IGURE
2:
P
RICE CHANGES DUE TO CONTAGION IN RESPONSE TO FACTOR LOADINGS
... 30
F
IGURE
3:
C
LOSED FOUR
-
LINK CHAIN
... 39
F
IGURE
4:
T
WO
-
LINK CHAINS
... 40
F
IGURE
5:
O
PEN CHAIN
... 41
F
IGURE
6:
T
HE ORDER OF EVENTS
... 45
F
IGURE
7:
T
HE CONCURRENCE OF LIMIT SIGNALS
... 48
F
IGURE
8:
T
RANSMISSION OF CRISES
... 51
F
IGURE
9:
G
RAPHICAL ILLUSTRATION OF INEQUALITY
(B.10) ... B4
V
Index of tables
Page
T
ABLE
1:
C
ONDITIONAL SIGNAL DISTRIBUTION PROBABILITIES
... 9
T
ABLE
2:
I
MPLICATIONS OF A CAPITAL EVENT TO A NON
-
LEVERAGED PORTFOLIO
... 18
T
ABLE
3:
I
MPLICATIONS OF A CAPITAL EVENT TO A LEVERAGED PORTFOLIO
... 18
T
ABLE
4:
I
MPLICATIONS OF A CAPITAL EVENT TO SINGLE ASSETS WITHIN A NON
-
LEVERAGED PORTFOLIO
. 19
T
ABLE
5:
I
MPLICATIONS OF A CAPITAL EVENT TO SINGLE ASSETS WITHIN A LEVERAGED PORTFOLIO
... 20
VI
Index of symbols
3 Herd behavior and information cascades
A
... Acceptance of an investment project
R
... Rejection of an investment project
V
... Liquidation value of an investment project
H
... Positive private signal (High) regarding an investment project
L
... Negative private signal (Low) regarding an investment project
1
V
= + ... Positive liquidation value, that is, a positive return on an
investment project
1
V
= - ... Negative liquidation value, that is, a negative return on an
investment project
p
... Probability of a private signal
( )
Prob
... Probability function
(
)
|
Prob A B ... Conditional probability
4 Portfolio diversification and optimization
,
1
t t
+ ... Time period (present and future)
,
1
i t
µ
+
... Mean of a conditional joint-normal distribution of returns, based on
period-t-information
2
,
1
i t
+
... Variance of a conditional joint-normal distribution of returns, based
on period-t-information
k
... Predefined rate of return on equity capital
,
1
i t
R
+
... Gross rate of return on equity capital of risky asset i from period t
to
1
t
+
R
... Predefined rate of return on equity capital
m
... Likelihood of returns falling short of a predefined rate of return on
equity capital
t
V
... Equity capital available to portfolio manager
n
... Z-value of a normal distribution
N
... Number of risky assets
r
... Gross interest rate (borrowing and lending rate)
t
W
... Desired holding of risky assets
L
... Lending
B
... Borrowing (leverage)
VII
5 Signal-extraction problems and investor panics
y
... Action taken by informed investors (can be either positive
( )
H
y
or
negative
( )
L
y
)
s
... Component of y relevant for all investors (return of an investment
project) (can be either positive
( )
H
s
or negative
( )
L
s
)
m
... Component of y relevant for informed investors only (can be
either positive
( )
H
m
or negative
( )
L
m
)
2
... Variance of s
2
... Variance of m
... Fraction of variances
... Set of possible outcomes
( )
Prob
... Probability function
6 Cross-market re-balancing and cross-hedging
N
... Number of risky assets
P
...
1
N
× price vector
v
... Random vector of liquidation values
... Expected liquidation value v
u
... Residual
I
X
... Investment decision of informed investors
UI
X
... Investment decision of uninformed investors
f
...
1
M
× vector of common factors
B
... N
M
×
matrix of factor loadings
...
1
N
× vector of country-specific factors
7 Benchmark tracking and risk aversion
i
D
... Dividends of country i
H
D
... Dividends of the country an investor is optimistic about
L
D
... Dividends of the country an investor is pessimistic about
i
... Coefficient of absolute risk aversion of investor i
i
U
... Investor i's utility
t
i
W
... Investor i's wealth in period t
... Degree to which investors care about relative returns as opposed to
absolute returns
2
... Volatility of dividends
VIII
c
P
... Country c's asset price
c
P
... Country c's fixed asset price
,
i c
X
... Investor i's share of country c
c
K
... Fixed number of country c's shares
,
i c
b
... Investor i's share of county c with respect to her investment
portfolio (
(
)
,
,1
,2
/
i c
i
i
X
X
X
+
)
9 Wealth effects in a unique equilibrium model
( )
R
... Gross return of an investment project
i
... Fundamentals of country i
i
n
... Proportion of agents who prematurely withdraw their investments
in country i
j
i
... Error term or noise signal for country i observed by investor j
j
i
... Posterior belief of investor j over fundamentals in country i
,
... Positive and negative outcome of fundamentals
j
i
w
... Wealth of investor j resulting from her investment in country i
u
... Utility
*
2,r
... Threshold for investors who ran in Country One
*
2,nr
... Threshold for investors who did not run in Country One
*
... Cut-off point regarding a particular level of fundamentals
~
,
... Limit signals
(
)
2,
2,
r
nr
w
w
... Country Two wealth if an investor run (does not run) in Country
One
2
Ew
... Expected return in Country Two
Introduction
1
Introduction
In recent years academics and policy makers have become more and more
interested in the phenomenon of contagion, a concept involving the transmission of a
financial crisis from one country to one or more other countries. During the 1990s world
capital markets witnessed a number of financial crises. In 1992 the Exchange Rate
Mechanism (ERM) crisis hit the European continent. Several countries in Latin
America have been rocked during the 1994-95 Tequila crisis, and the Asian Flu spread
through East Asian countries in 1997-98 with dramatic social implications. Later in
1998 the famous hedge fund Long Term Capital Management (LTCM) had to file for
bankruptcy and the Russian debt failure shocked international capital markets and
increased volatility on a global scale. The crisis spread to as far as Brazil in early 1999
and developed markets have become victims as well.
The question asked by academics and policy makers is how countries should
behave in order to avoid contagion. To answer this question it is necessary to
understand the different channels of contagion in greater detail and how a crisis can be
transmitted from one country to another. The objective of this paper is to highlight those
channels and to present a number of models and theories of contagion, which have
recently been developed by academics.
In general, there are several strands of theories in the literature that try to explain
the transmission of crises. During the mid and late 1990s fundamental-based contagion
and spillovers became popular among researchers and policy makers. Furthermore,
financial linkages have been known to contribute to contagion. In contrast, in recent
years, portfolio flows of international investors moved into the focus of academics.
The advocates of fundamental-based contagion and spillovers argue that trade
linkages between countries are responsible for contagion. For instance, a devaluation of
a country's currency may lead to a negative change in fundamentals of its trading
partners. On the other hand, contagion due to financial linkages is mainly explained by
the fact that countries share the same banks and therefore have common creditors. A
crisis in one country then leads to a deteriorating balance sheet of those common
creditors. This in turn may force banks to withdraw money out of other countries in
order to avoid further losses, a fact that leads to contagious sellouts.
The role of international portfolio flows, which is at the core of this paper, has
become increasingly important in recent years, mainly due to the substantial progress in
globalization. International financial markets became increasingly integrated and
significant changes in international portfolio flows can easily lead to contagious sellouts
and the transmission of crises on a global scale. Compared to other explanations of
contagion, these theories abstract from the fact that countries must have fundamental
linkages of any kind or that they engage in bilateral or third-party trade.
The paper is divided into four parts. Part I provides an overview. The idea of
contagion will be defined in section one and the transmission of crises based on
fundamentals and spillovers will be introduced in section two. Part II through IV builds
the heart of this paper. These parts offer a detailed analysis of various theories of
Introduction
2
contagion based on international portfolio flows, which have been discussed recently in
the literature.
Part II is comprised of some basic applications. The analysis starts in section three
with an introduction to herd behavior and, more specifically, to information cascades.
Even though information cascades are not directly a theory of contagion with respect to
international portfolio flows, this concept can be viewed as relevant as it may explain
how herds occur, which in turn can lead to changes in international portfolios. An
application of basic principles of portfolio diversification will be at the heart of section
four. It will be demonstrated how shocks to an asset's volatility or to equity capital
available to an investor will influence international portfolio diversification.
Part III deals with information asymmetries and contains two different models. The
analysis starts with an introduction to information asymmetries and their influence on
investment decisions in section five. There will be an illustration of a rational investor
panic that is not justifiable by deteriorating fundamentals but is simply due to a
particular type of herding. In section six, the second model then introduces the method
of cross-hedging, which is also known as cross-market re-balancing. This theory is
based on classic asset pricing models and illustrates, using examples, how contagion
can occur due to portfolio re-balancing processes.
Part IV finally discusses the concepts of risk aversion and wealth effects. In section
seven investors' risk aversion and the compensation schemes of fund managers will be
used to demonstrate the concepts of benchmark tracking and momentum trading. This
behavior pattern can lead to portfolio re-allocations and may further step up the
transmission of financial crises. A basic concept of portfolio linkages is used in section
eight to introduce the effects of a change in investors' wealth and in their aversion to
risk. This concept will be part of a model that will also be discussed in section nine. It
will explain how investors are influenced by the decision-making process of other
investors. This influence may then lead to a financial crisis where every investor
rationally pulls out of a country. The model avoids the problem of multiple equilibria as
it is known from traditional second-generation models of balance-of-payments crises.
Instead, the model will always provide a unique equilibrium.
The paper ends with some summarizing and concluding remarks. The previously
discussed theories will be analyzed as a whole. Various proofs are included in appendix
A through D in order to keep the paper as readable as possible.
1 Defining contagion
3
PART I: OVERVIEW
1 Defining contagion
Although a number of papers regarding contagion exist, there is no consensus
about a single definition of this particular term. Broadly speaking, contagion can be
seen as the transmission of financial crises among emerging markets economies
(EMEs). Masson (1998, pp. 4-5) delivered a definition of contagion that is commonly
accepted among various academics. He distinguishes between three types of crises
transmission:
1. Monsoonal effects, defined as major economic shifts in industrial countries that
trigger crises in EMEs.
2. Spillovers, interpreted as interdependence of macroeconomic fundamentals
among EMEs. That is, a crisis in one particular EME may influence the
fundamentals of another EME.
3. Pure contagion, defined as the transmission of a crisis for reasons unexplained
by macroeconomic fundamentals. One can easily think of shifts in investors'
sentiment, investors' risk aversion or changes in international portfolio flows.
Masson (1998, p. 5) applies the term contagion only to the third category. Kumar &
Persaud (2001, p. 6) interpret contagion as risk-based, where contagion takes place
because of changes in investors' risk appetite and is therefore independent of
macroeconomic fundamentals or spillovers.
A set of more detailed definitions of contagion has been introduced by Pericoli &
Sbracia (2001). Those definitions are usually associated with certain types of empirical
studies. The following five distinct definitions are mentioned:
1. "Contagion is a significant increase in the probability of a crisis in one country
conditional on a crisis occurring in another country" (p. 9).
2. "Contagion occurs when volatility spills over from the crisis country to the
financial markets of other countries" (p. 9).
3. "Contagion is a significant increase in co-movements of prices and quantities
across markets, conditional on a crisis occurring in one market or group of
markets" (p. 10).
4. "(Shift-)contagion occurs when the transmission channel is different after a
shock in one market" (p. 10).
5. "Contagion occurs when co-movements cannot be explained by fundamentals"
(p. 10).
According to Pericoli & Sbracia (2001) the first definition is used in order to
determine contagion in conjunction with exchange rate crises because such crises
usually involve several countries. Note that this definition does not imply any particular
channel of contagion. This means that also monsoonal effects and spillovers, as defined
by Masson (1998), would nonetheless be labeled as contagion. The core of the second
2 Spillover effects and financial linkages
4
definition is that crises are usually associated with an increase in volatility. As a rise in
volatility can be interpreted as a rise in uncertainty, this definition would account for
pure contagion. The third definition takes into consideration the fact that co-movements
across markets might be due to normal interdependencies. Contagion is therefore
defined as excessive co-movements compared to normal or standard co-movements
during tranquil periods. The fourth definition is somewhat comparable. Differences in
the transmission channel can also be measured in terms of excessive co-movements.
The last definition of contagion is again compatible with Masson's (1998) definition of
pure contagion. It implies changes in the behavior of market participants, among other
things.
An additional way to define contagion is to make a distinction between two
categories of contagion, namely fundamental-based contagion and contagion based on
investors' behavior (Claessens et al., 2000, p. 4). The first category of contagion
includes common shocks, trade links and competitive devaluations as well as financial
links. This set of causes can be best associated with monsoonal effects and spillovers.
Investors' behavior, on the other hand, is again a form of pure contagion as it does not
involve any macroeconomic fundamentals but takes into consideration liquidity and
incentive problems, information asymmetries and coordination problems, multiple
equilibria and changes in the rule of the game (explanations to the first category are
given in section two, while the role of international portfolio flows, described in section
three through nine, is part of the second category).
Finally, Calvo & Mendoza (2000, p. 81) define the term contagion as follows: "...
contagion is reflected in portfolio re-allocations that are not directly related to the
'fundamentals' determining the risk and return properties of asset returns." This
definition also treats the term contagion as pure contagion comparable to Masson
(1998), Kumar & Persaud (2001) and Pericoli & Sbracia (2001). As portfolio aspects
are at the core of this definition, it will be used throughout the rest of this paper and
especially in section three through nine.
2 Spillover effects and financial linkages
In the mid to late 1990s, spillover effects and financial linkages became popular
among academics, and various studies have been conducted in order to examine the
importance of these concepts. This section provides a brief overview of spillover
effects, which are based on direct or indirect trade linkages. Moreover, contagion due to
financial linkages is introduced. Both subsections are supplemented with various
empirical studies that illustrate the real-world application of these theories.
2.1 Bilateral and third party trade
Trade linkages have been quite popular in recent literature on contagion as
comprehensive data is available for a long period of time. Additionally, the quantitative
2 Spillover effects and financial linkages
5
implications of those linkages are comparable easy to measure during times of crises.
Many authors highlighted the effects that both bilateral trade links and third party trade
could have on the transmission of crises.
1
Regarding bilateral trade, the transmission of a crisis is based on two effects, the
so-called competitive devaluation and the income effect. The former can be best
described through an example. If Country One, for instance, gets hit by a currency crisis
and experiences a sharp devaluation of its currency, relative prices of goods and
services are affected. If Country One has bilateral trade links with Country Two,
Country Two will experience deteriorated price competitiveness. Export prices of
Country Two (or equally import prices of Country One) will increase and therefore
exports of Country Two (or imports of Country One) will decrease. On the other hand,
the devaluation in Country One leads to increased imports initiated by Country Two.
This in turn leads to a deteriorating current account and can be the cause for a balance-
of-payments crisis in Country Two. Hence, there arises an incentive for Country Two to
devalue its currency, too. As Frenkel & Fendel (2004, pp. 136-137) point out, this
model is somewhat in contrast to reality as central banks usually attempt to keep an
exchange rate peg for as long as possible.
The income effect states that if a country experiences a devaluation of its currency
in conjunction with a serious downturn of its economy, then the demand for goods and
services from other countries will decrease. Thus, trading partners of that particular
country will face lower exports. As a result, these trading partners experience a
macroeconomic shock transmitted from the country that initially faced the currency
crisis (the ground-zero country).
In addition to bilateral trade, third party trade also plays a significant role in
explaining contagion. This channel implies that a crisis can be transmitted from one
country to another despite the fact that the two countries do not possess any direct trade
linkages. If Country One faces a currency crises combined with a sharp devaluation, it
becomes more price-competitive. This in turn may reduce demand from third party
countries for products from Country Two. Note that this is only true if Country Two's
exports overlap with those of Country One, so that both countries compete with the
same products in international markets. The reduced demand for Country Two products
then leads to an economic downturn in Country Two.
Empirical research on crises transmitted via trade suggests that this channel might
be significantly relevant. In addition, the fact that recent financial and currency crises
only spread regionally, rather than globally, can be explained with trade linkages as
most international trade of EMEs is conducted with other EMEs in geographic
proximity (Glick & Rose, 1998, p. 17). Other studies, such as Kaminsky & Reinhart
(2000, pp. 162-163 and 166), came to the conclusion that trade linkages do not
necessarily do a good job in explaining crises contagion. These studies see financial
linkages, among others, as more explanatory channels of contagion. For instance, the
contagion of the Russian crises to Brazil does not seem to fit into classical explanations
1
See Caramazza et al. (2000), Frenkel & Fendel (2004), Glick & Rose (1998), Hall & Taylor (2002),
Hashimoto & Ito (2005), Hernández & Valdés (2001), as well as Kaminsky & Reinhart (2000),
among others.
2 Spillover effects and financial linkages
6
of contagion via trade. This is particularly true because the spread of the Russian crises
crossed international borders and has not been regional at all. Possible explanations can
be seen in changes in international portfolio flows and will be discussed in detail in later
sections.
2.2 The common bank lender / common creditor effect
Another popular explanation for contagion is the so-called common bank lender or
common creditor effect. If different EMEs are dependent on the same creditor banks,
contagion can occur despite the non-existence of any direct financial linkages between
these particular EMEs. Consider a commercial bank that suffers losses in its loan
portfolio because a borrowing EME experiences a currency crisis combined with an
economic downturn and thus fails to repay its loans. Such a loss impacts the bank's
balance sheet and it may be necessary to re-balance its loan portfolio. This could either
involve that the bank calls in its loans from the crisis EME and therefore dries up credit
lines (this would exacerbate the original crisis), or the creditor bank may additionally
call in loans in other EMEs, if it thinks that these economies may also suffer a crisis in
the future. Even if the creditor bank does not expect the crisis to spread, the bank could
be forced to lend less (if at all) to other EMEs because of banking regulations,
provisions, or margin calls. The bank may also be forced to re-balance its loan portfolio
in order to adjust to its lower level of wealth or it can be subject to Value-at-Risk
regulations.
2
The common bank lender effect is strongest when an EME is predominantly
dependent on a single creditor and if its loans are of short maturity. If the creditor
refuses to rollover existing loans, the borrowing country would face a huge capital
outflow in a short period of time. The large dependence on a single creditor makes it
hard for the country to raise additional funds. If different EMEs face the same credit
restrictions, the crisis will spread from the ground-zero country (the country that
initially suffers the financial crisis) to other EMEs through common creditors due to
their unwillingness or incapability, respectively, to provide additional funds.
In recent literature on contagion the bank lender effect has repeatedly been
examined empirically. Many studies conclude that this effect is significant in order to
explain crises contagion. These studies also take a stand for the regional spreading of
financial crises, rather than contagion on a global scale. It is commonly understood that
trade linkages mostly imply financial linkages in the form of a common creditor. For
instance, Van Rijckeghem & Weder (2001) constructed a measure of competition for
bank funds. The authors came to the conclusion that the bank lender effect helps to
explain contagion during the Mexican, Asian, and Russian financial crises. For instance,
they point out that 29% of Japanese capital had been exposed to Thailand during the
Asian currency crisis. This is a large share and as Thailand began to struggle, Japanese
banks not only called in loans from Thailand but all around Asia. This fact helped to
spread the crisis.
2
An extensive review of Value-at-Risk regulations can be found in Jorion (2001).
2 Spillover effects and financial linkages
7
This is also confirmed by Kaminsky & Reinhart (2001, p. 9). Prior to the Asian
crisis capital inflows to Indonesia, Malaysia, the Philippines, South Korea and Thailand
(Emerging Asia) surged as a result of dramatic increases in Japanese bank lending. On
the other hand, during the Asian crisis there have been massive capital outflows from
Emerging Asia. According to the authors (p. 27), between June 1997 and June 1998
Japanese banks pulled out roughly USD 23 billions or about 23.6% of their total
lending, respectively. Although European and American banks retrenched about USD
8.5 billions (10%) and USD 7.2 billions (30.2%), respectively, Japanese banks played
the major role in this liquidity crunch. About two-thirds of Japanese loans have been
granted to Emerging Asia, while Thailand accounted for more than 25% of loans (p.
29). This means that Japanese banks have been quite dependent on economic
developments of the country that first experienced a currency crisis during the Asian
Flu. Kaminsky and Reinhart (2001, p. 10) are of the opinion that the huge capital
outflow, initiated by Japanese banks calling in loans and followed by European and
U.S. banks, respectively, helped spread the crisis across Emerging Asia.
In an earlier study, Kaminsky & Reinhart (2000, p. 155) also highlighted that the
common creditor effect is not a new crisis transmission channel. They point out that
U.S. banks played a major role in the debt crisis in the early 1980s, comparable to
Japanese banks during the Asian crisis. In addition, to further test the bank lending
channel the authors formed clusters of countries that are either dependent on U.S. banks
or Japanese banks. As a result, if there are crises in several countries within a cluster,
the conditional probability of a crisis is higher than the unconditional probability. Banks
are not willing or are unable (for reasons explained earlier in this subsection) to provide
new or rollover existing loans to countries that are of high risk and might face a
financial crises as well.
In another study, Caramazza et al. (2000) developed some indicators of
vulnerability and found that the common creditor effect is highly significant in
explaining contagion. They compared different variables of crises and non-crises
countries. In terms of the common creditor effect they found that "the common creditor
holds a 10 percentage point higher share of the external bank liabilities of the crises
countries than of the non-crises countries, whereas the average crisis country holds a 5
percentage point higher share of the external loan portfolio of the common creditor than
the average non-crisis country" (Caramazza et al., 2000, p. 16). Regression analysis
results indicate that the probability of a crisis is higher, if a country has financial
linkages with the major creditor of the country that first experiences a crisis (Caramazza
et al., 2000, p. 35).
To sum up, empirical research indicates that the bank lender / common creditor
effect may play an important role in explaining the contagion of currency and financial
crises across boarders. One of the main explanations can be seen in the necessity of re-
balancing a bank's credit portfolio for various reasons.
3 Herd behavior and information cascades
8
PART II: BASIC APPLICATIONS
3 Herd behavior and information cascades
During recent financial crises it became obvious that market participants fled
affected countries in a relatively short period of time. Market commentators blamed
fund managers and other investors for acting as herds. Most of the time this behavior
has been called irrational and many argue that herd behavior destabilizes markets and
adds to uncertainty and increased volatility, respectively. This section provides an
introduction to herd behavior and delivers a basic explanation of information cascades.
Although information cascades do not directly contribute to the explanation of
contagion due to international portfolio flows, the subject is nevertheless highly relevant
in order to understand rational herd behavior and the flight to safer havens by many
market participants during financial crises. To put it differently, the occurrence of
information cascades can be the reason for international investors to re-allocate their
investment positions. Information cascades therefore indirectly add to the major subject
of this paper.
The remainder of this section is structured as follows: there will be a definition of
herd behavior and, more specifically, of information cascades. Subsequently, a basic
model of information cascades is presented while the section ends with some
concluding remarks.
3.1 Defining herd behavior and information cascades
In economics, the term 'herd behavior' is used when an individual's acts are
conditional on the action of other individuals (see for this and in the following:
Bikhchandani & Sharma, 2001, p. 280). An investor is said to herd if she decides to
make an investment once she knows that other investors are also going to invest.
Alternatively, an investor herds if she is not making an investment because other
investors are not investing, too.
Information cascades are an integral part of herd behavior. In general, an
"information cascade occurs when it is optimal for an individual, having observed the
actions of those ahead of him, to follow the behavior of the preceding individual
without regard to his own information" (Bikhchandani et al., 1992, p. 994).
The first models of information cascades were developed independently in the
early 1990s by Banerjee (1992), Bikhchandani et al. (1992), and Welch (1992). More
recent papers by Bikhchandani & Sharma (2001), Hirshleifer & Teoh (2003) and
Freiberg (2004) provide a comprehensive overview of information cascades and herd
behavior in financial markets. Those papers build the theoretical framework for the
model of information cascades discussed below.
Ende der Leseprobe aus 85 Seiten
- nach oben
Details
- Arbeit zitieren
- , 2006, Theories of Contagion, Hamburg, Bedey Media GmbH, https://www.diplom.de/document/224990
Ähnliche Arbeiten
Deutschlands größter Sharing Community
Jetzt Zusammenfassungen, Skripte und Klausuren kostenlos downloaden!
Uniturm.de
Ihre Vorteile als Autor
Geld verdienen Für jedes verkaufte Exemplar erhalten Sie Autorenhonorar - bis zu 40%.Kostenlose Buchveröffentlichung
Als eBook-Studie im Original und / oder auf Wunsch zusätzlich als hochwertiges Fachbuch in unserem renommierten Buchverlag.
Renommee als Fachbuchautor
Präsentieren Sie sich als Fachfrau oder Fachmann in Ihrem Fachgebiet und machen Sie sich bekannt.
Persönliche Betreuung
Unser Lektorat wird Sie persönlich betreuen und ist auch telefonisch unter 040-655992-14 erreichbar.
Weltweit im Buchhandel
Ihr Buch ist weltweit im Buchhandel und Online-Buchhandel wie z.B. amazon erhältlich.
Jetzt Autor werden!
Kommentare