## _pdp01

## Source details

**Canonical URL:** [_pdp01](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/pdp/2006/_pdp01.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/pdp/2006/_pdp01.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/pdp/2006/_pdp01.pdf.json)

---

### I. Introduction
- Paper: Prepared by Tatiana Didier, Paolo Mauro, and Sergio Schmukler; January 2006.
- JEL Classification Numbers: F02, F15, F36.
- Keywords: crises, international integration, international transmission of shocks.
- Core proposition:
  - While many emerging market crises of the 1990s were characterized by widespread contagion, more recent crises (notably Argentina 2001–02) were, for the most part, contained within national borders.
  - A prudent working assumption is that contagion has not vanished permanently.
  - Available data do not point to a disappearance of the main channels that transmit crises across countries.
  - Anticipation of the Argentine crisis by international investors may help explain the recent absence of contagion.
- Asset-market observations:
  - Emerging market bond spreads in the 1990s moved largely in near-unison; shocks in the 2000s appear more idiosyncratic with country-specific spread spikes (Argentina, Brazil, Nigeria, Venezuela).
  - The average six-month correlation coefficient (for 66 country pairs) was lower in 2000–2005 than in 1994–1999.
  - Six-month volatility of daily returns declined in the 2000s relative to the 1990s.
  - Caveats: correlation declines are not conclusive evidence of diminished contagion because correlations tend to be higher during turmoil and may reflect common factors; corrections require strong assumptions.

### II. Channels of Contagion — Overview
- Definition: “contagion” is a price movement in one market resulting from a shock in another market.
- Necessary (but not sufficient) condition for contagion channels: substantial international trade and financial integration.
- Empirical context:
  - Linkages among countries have deepened substantially over the past two decades.
  - International trade grew considerably faster than GDP for all groups of countries.
  - Several measures of financial integration increased.

### II.A The Trade Channel
- Mechanics:
  - Competitiveness effect: relative price changes affect the country’s ability to compete abroad after devaluation.
  - Income effect: crises reduce income measured in foreign currency, curbing import demand.
  - Effects can be amplified by credit constraints.
- Historical role:
  - Trade links implicated in transmission of several crises: 1992–93 ERM crisis, 1994 Mexican crisis, 1997 Asian crisis, 1999 Brazilian crisis.
  - The 1998 Russian crisis showed limited bilateral trade links with affected countries, implying trade alone cannot explain all contagion episodes.
- Empirical trade statistics (Imports plus exports, in percent of GDP):
  - Latin America and the Caribbean: 1985 = 28; 1990 = 32; 1995 = 38; 2000 = 42; 2003 = 45.
  - South Asia: 1985 = 18; 1990 = 21; 1995 = 28; 2000 = 32; 2003 = 34.
  - East Asia and Pacific: 1985 = 33; 1990 = 45; 1995 = 60; 2000 = 72; 2003 = 81.
  - World: 1985 = 38; 1990 = 38; 1995 = 42; 2000 = 50; 2003 = 52.
- Regional concentration example:
  - ASEAN+3 regional trade share rose from about 30 percent in the mid-1980s to about 45 percent in 2004.

### II.B The Financial Channel
- Mechanisms:
  - Direct financial linkages between residents of different countries in some cases.
  - More frequently, transmission via international investors (“common creditors”) who hold assets across multiple countries.
- Institutional mechanics:
  - Open-end mutual funds facing redemptions sell assets globally to raise cash.
  - Leveraged investors (banks, hedge funds) rebalance and sell assets in other countries when facing regulatory requirements, provisioning practices, or margin calls.
- Behavioral drivers:
  - Increased risk aversion can induce portfolio rebalancing away from risky assets broadly.
  - Information asymmetries can cause investors to generalize shocks to other countries with shared characteristics.
  - Information cascades and herding: less informed investors mimic others, reinforcing transmission.
  - Reputation and compensation structures (performance judged relative to peers) create incentives for managers to herd.
- Historical evidence:
  - Financial channel prominent in 1990s contagion episodes: ERM, Mexico 1994 (via U.S. investors), 1997 Asian crisis (via European and Japanese banks; mutual funds), 1998 Russian crisis (mutual funds and hedge funds; LTCM debacle).
- Contemporary exposure:
  - Exposure to emerging market countries by international banks, mutual funds, and hedge funds does not seem to have changed sufficiently to eliminate concerns about potential contagion.

### International investor exposure: banks, mutual funds, and hedge funds
- International bank claims of residents of the largest advanced economies:
  - Peaked prior to the East Asian crisis, declined in its aftermath, and recently returned to pre-crisis levels.
  - Regional shifts: international bank claims on Eastern European countries have recently increased; claims on Latin American countries have declined; claims on East Asian countries are now close to pre-crisis levels.
- Mutual funds:
  - Assets issued by Asian economies under management by mutual funds grew by more than US$62 billion (316 percent) between March 1999 and February 2005.
  - Mutual fund holdings increased by US$3.5 billion (18 percent) for Latin American assets between March 1999 and February 2005.
  - Mutual fund holdings increased by US$17.8 billion (407 percent) for Eastern European assets between March 1999 and February 2005.
  - Data for 2005 reflect asset holdings at end-February (middle-panel data).
- Hedge funds:
  - Assets under management doubled between 2000 and 2004.
  - Hedge fund activity in emerging market countries grew considerably since early 2004 (IMF, 2005).
  - Hedge funds have high turnover; trading activity by hedge funds has been a dominant factor in trading activity of emerging country bond markets.
  - Among hedge funds, funds dedicated to emerging market countries are capturing a rising share of trading activity.

### Empirical evidence on mutual fund behavior around crises
- Observed mutual fund holdings (end-of-month series for approximately 700 equity mutual funds) show:
  - Sharp declines in holdings after crises for Hong Kong SAR, Indonesia, and Korea following Thailand's July 1997 devaluation.
  - Argentina stands out: holdings fell from a peak of US$3.98 billion in April 1999 to US$0.25 billion in November 2001—about a month before the devaluation of the peso, bank closures, and overall financial collapse.
- Valuation correction:
  - Uses the overall stock market index (MSCI or S&P/IFC equity index) as an approximation for mutual fund portfolios.
  - Once adjusted for market performance, Argentina is the only case where mutual fund reductions survive the correction: mutual funds began considerably reducing exposure more than two years before the crisis.
  - From April 1999 to November 2001, mutual funds sold (net) almost US$3.5 billion, equivalent to 87 percent of their initial asset holdings.

### Table 2 highlights — International mutual fund net buying/selling before crises (in Millions of U.S. Dollars)
- Table structure: Crisis economy | Month of Peak | Month of Crisis | Since Peak | Six Months Before | Three Months Before
- Key entries:
  - Argentina: Month of Peak 04/99; Month of Crisis 12/01; Since Peak -3,456; Six Months Before -259; Three Months Before -63
  - Hong Kong SAR: Month of Peak 05/97; Month of Crisis 07/97; Since Peak 142; Six Months Before -1,134; Three Months Before 58
  - Indonesia: Month of Peak 02/97; Month of Crisis 07/97; Since Peak 1,702; Six Months Before 48; Three Months Before -79
  - Malaysia: Month of Peak 02/97; Month of Crisis 07/97; Since Peak -865; Six Months Before -671; Three Months Before -1,068
  - Russia: Month of Peak 09/97; Month of Crisis 08/98; Since Peak 2,944; Six Months Before 432; Three Months Before 218
  - Korea: Month of Peak 08/96; Month of Crisis 07/97; Since Peak 6,548; Six Months Before 716; Three Months Before 76
  - Thailand: Month of Peak 01/96; Month of Crisis 07/97; Since Peak 3,828; Six Months Before 229; Three Months Before 6

### Interpretation: anticipation, contagion channels, and implications
- Anticipation matters:
  - Many investors were taken by surprise by crises of the 1990s, whereas more recent crises—especially Argentina—were anticipated well in advance, as reflected in asset price movements and mutual fund behavior.
  - Anticipated crises may allow investors to re-balance portfolios in an orderly way, avoiding overreaction in asset prices.
  - Unanticipated shocks can generate large effects through rapid unwinding and liquidation, amplifying contagion through financial channels.
- Channels of contagion:
  - Financial channel: international investors’ holdings and trading practices can transmit shocks across countries when positions are unwound.
  - Trade channel: crises transmitted via trade links should not depend on whether crises are anticipated; evidence here points to financial links (and anticipation) as key to recent low contagion incidence.
- Overall assessment:
  - Exposure of international institutional investors to emerging market economies has increased in recent years.
  - Because exposure has not diminished, a prudent working assumption is that contagion might re-emerge, particularly if present conditions of abundant global liquidity are reversed or a large unanticipated adverse shock occurs in a country whose assets are held by international investors.
  - The low incidence of contagion in the 2000s may reflect widespread anticipation of episodes (notably Argentina) rather than a permanent disappearance of contagion risk.

### Conclusion and research needs
- Main conclusions:
  - It is premature to conclude contagion has vanished.
  - Factors underlying trade and financial channels remain at least as strong as during the 1990s; international integration has deepened.
  - Anticipation likely contributed to the near-absence of contagion in recent episodes.
- Recommendation for further work:
  - Additional research and evidence are needed to determine whether the low incidence of contagion in recent years reflects a permanent change versus a temporary pattern driven by anticipation and market conditions.
- Interim practical stance:
  - Maintain a prudent working assumption that contagion could re-emerge in the medium term.

*Source: IMF Policy Discussion Paper PDP/06/1, “Vanishing Contagion?” by Tatiana Didier, Paolo Mauro, and Sergio Schmukler (January 2006).*

### Section 1

### Vanishing Contagion?

### I. Introduction
- Paper: Prepared by Tatiana Didier, Paolo Mauro, and Sergio Schmukler; January 2006.
- JEL Classification Numbers: F02, F15, F36.
- Keywords: crises, international integration, international transmission of shocks.
- Core proposition:
  - While many emerging market crises of the 1990s were characterized by widespread contagion, more recent crises (notably Argentina 2001–02) were, for the most part, contained within national borders.
  - A prudent working assumption is that contagion has not vanished permanently.
  - Available data do not point to a disappearance of the main channels that transmit crises across countries.
  - Anticipation of the Argentine crisis by international investors may help explain the recent absence of contagion.

- Observations from asset-market data:
  - Emerging market bond spreads in the 1990s moved largely in near-unison; shocks in the 2000s appear more idiosyncratic with country-specific spread spikes (Argentina, Brazil, Nigeria, Venezuela).
  - The average six-month correlation coefficient (for 66 country pairs) was lower in 2000–2005 than in 1994–1999.
  - Six-month volatility of daily returns declined in the 2000s relative to the 1990s.
  - Caveats: correlation declines are not conclusive evidence of diminished contagion because correlations tend to be higher during turmoil and may reflect common factors; corrections require strong assumptions.

### II. Channels of Contagion — Overview
- Broad definition used: “contagion” is a price movement in one market resulting from a shock in another market.
- Necessary (but not sufficient) condition for contagion channels: substantial international trade and financial integration.
- Empirical context: linkages among countries have deepened substantially over the past two decades; international trade grew considerably faster than GDP for all groups of countries; several measures of financial integration increased.

### II.A The Trade Channel
- Mechanics:
  - Competitiveness effect: relative price changes affect the country’s ability to compete abroad after devaluation.
  - Income effect: crises reduce income measured in foreign currency, curbing import demand.
  - Effects can be amplified by credit constraints.
- Historical role:
  - Trade links implicated in transmission of several crises: 1992–93 ERM crisis, 1994 Mexican crisis, 1997 Asian crisis, 1999 Brazilian crisis (citations: Eichengreen and Rose, 1999; Glick and Rose, 1999; Forbes, 2001 and 2004).
  - However, the 1998 Russian crisis showed limited bilateral trade links with affected countries, implying trade alone cannot explain all contagion episodes.
- Empirical trade statistics (Imports plus exports, in percent of GDP):
  - Latin America and the Caribbean: 1985 = 28; 1990 = 32; 1995 = 38; 2000 = 42; 2003 = 45.
  - South Asia: 1985 = 18; 1990 = 21; 1995 = 28; 2000 = 32; 2003 = 34.
  - East Asia and Pacific: 1985 = 33; 1990 = 45; 1995 = 60; 2000 = 72; 2003 = 81.
  - World: 1985 = 38; 1990 = 38; 1995 = 42; 2000 = 50; 2003 = 52.
- Regional concentration example:
  - ASEAN+3 regional trade share rose from about 30 percent in the mid-1980s to about 45 percent in 2004.

### II.B The Financial Channel
- Mechanisms:
  - Direct financial linkages between residents of different countries in some cases.
  - More frequently, transmission via international investors (“common creditors”) who hold assets across multiple countries.
  - Institutional mechanics:
    - Open-end mutual funds facing redemptions sell assets globally to raise cash.
    - Leveraged investors (banks, hedge funds) rebalance and sell assets in other countries when facing regulatory requirements, provisioning practices, or margin calls.
  - Behavioral drivers:
    - Increased risk aversion can induce portfolio rebalancing away from risky assets broadly.
    - Information asymmetries can cause investors to generalize shocks to other countries with shared characteristics.
    - Information cascades and herding: less informed investors mimic others, reinforcing transmission.
    - Reputation and compensation structures (performance judged relative to peers) create incentives for managers to herd.
- Historical evidence:
  - Financial channel prominent in 1990s contagion episodes: ERM, Mexico 1994 (via U.S. investors), 1997 Asian crisis (via European and Japanese banks; mutual funds), 1998 Russian crisis (mutual funds and hedge funds; LTCM debacle).
- Contemporary exposure:
  - Exposure to emerging market countries by international banks, mutual funds, and hedge funds does not seem to have changed sufficiently to eliminate concerns about potential contagion.

*Source: IMF Policy Discussion Paper PDP/06/1, “Vanishing Contagion?” by Tatiana Didier, Paolo Mauro, and Sergio Schmukler (January 2006).*

### Section 2

### _pdp01 - Section 2

### International investor exposure: banks, mutual funds, and hedge funds
- International bank claims of residents of the largest advanced economies:
  - Peaked prior to the East Asian crisis, declined in its aftermath, and recently returned to pre-crisis levels.
  - Regional shifts: international bank claims on Eastern European countries have recently increased; claims on Latin American countries have declined; claims on East Asian countries are now close to pre-crisis levels.
- Mutual funds:
  - Assets issued by Asian economies under management by mutual funds grew by more than US$62 billion (316 percent) between March 1999 and February 2005.
  - Mutual fund holdings increased by US$3.5 billion (18 percent) for Latin American assets between March 1999 and February 2005.
  - Mutual fund holdings increased by US$17.8 billion (407 percent) for Eastern European assets between March 1999 and February 2005.
  - Data for 2005 reflect asset holdings at end-February (middle-panel data).
- Hedge funds:
  - Assets under management doubled between 2000 and 2004.
  - Hedge fund activity in emerging market countries grew considerably since early 2004 (IMF, 2005).
  - Hedge funds have high turnover; trading activity by hedge funds has been a dominant factor in trading activity of emerging country bond markets.
  - Among hedge funds, funds dedicated to emerging market countries are capturing a rising share of trading activity.

### Empirical evidence on mutual fund behavior around crises
- Observed mutual fund holdings (end-of-month series for approximately 700 equity mutual funds) show:
  - Sharp declines in holdings after crises for Hong Kong SAR, Indonesia, and Korea following Thailand's July 1997 devaluation.
  - Argentina stands out: holdings fell from a peak of US$3.98 billion in April 1999 to US$0.25 billion in November 2001—about a month before the devaluation of the peso, bank closures, and overall financial collapse.
- Correcting for price (valuation) effects:
  - Valuation correction uses the overall stock market index (MSCI or S&P/IFC equity index) as an approximation for mutual fund portfolios.
  - Once adjusted for market performance, Argentina is the only case where mutual fund reductions survive the correction: mutual funds began considerably reducing exposure more than two years before the crisis.
  - From April 1999 to November 2001, mutual funds sold (net) almost US$3.5 billion (Table 2), equivalent to 87 percent of their initial asset holdings (Figure 5).
  - For other crises (Thailand, Russia, Korea, Indonesia), mutual funds did not appear to reduce exposure in advance once price effects are accounted for; Malaysia is a partial exception with exposure falling considerably less and only a few months before the crisis.

### Table 2 highlights — International mutual fund net buying/selling before crises (in Millions of U.S. Dollars)
- Table structure: Crisis economy | Month of Peak | Month of Crisis | Since Peak | Six Months Before | Three Months Before
- Key entries (exact figures from Table 2):
  - Argentina: Month of Peak 04/99; Month of Crisis 12/01; Since Peak -3,456; Six Months Before -259; Three Months Before -63
  - Hong Kong SAR: Month of Peak 05/97; Month of Crisis 07/97; Since Peak 142; Six Months Before -1,134; Three Months Before 58
  - Indonesia: Month of Peak 02/97; Month of Crisis 07/97; Since Peak 1,702; Six Months Before 48; Three Months Before -79
  - Malaysia: Month of Peak 02/97; Month of Crisis 07/97; Since Peak -865; Six Months Before -671; Three Months Before -1,068
  - Russia: Month of Peak 09/97; Month of Crisis 08/98; Since Peak 2,944; Six Months Before 432; Three Months Before 218
  - Korea: Month of Peak 08/96; Month of Crisis 07/97; Since Peak 6,548; Six Months Before 716; Three Months Before 76
  - Thailand: Month of Peak 01/96; Month of Crisis 07/97; Since Peak 3,828; Six Months Before 229; Three Months Before 6

### Interpretation: anticipation, contagion channels, and implications
- Anticipation matters:
  - Many investors were taken by surprise by crises of the 1990s, whereas more recent crises—especially Argentina—were anticipated well in advance, as reflected in asset price movements and mutual fund behavior.
  - Anticipated crises may allow investors to re-balance portfolios in an orderly way, avoiding overreaction in asset prices.
  - Unanticipated shocks can generate large effects through rapid unwinding and liquidation, amplifying contagion through financial channels.
- Channels of contagion:
  - Financial channel: international investors’ holdings and trading practices can transmit shocks across countries when positions are unwound.
  - Trade channel: crises transmitted via trade links should not depend on whether crises are anticipated; evidence here points to financial links (and anticipation) as key to recent low contagion incidence.
- Overall assessment:
  - Exposure of international institutional investors to emerging market economies has increased in recent years.
  - Because exposure has not diminished, a prudent working assumption is that contagion might re-emerge, particularly if present conditions of abundant global liquidity are reversed or a large unanticipated adverse shock occurs in a country whose assets are held by international investors.
  - The low incidence of contagion in the 2000s may reflect widespread anticipation of episodes (notably Argentina) rather than a permanent disappearance of contagion risk.

### Conclusion and research needs
- The paper argues it is premature to conclude contagion has vanished:
  - Factors underlying trade and financial channels remain at least as strong as during the 1990s; international integration has deepened.
  - Anticipation likely contributed to the near-absence of contagion in recent episodes.
- Recommendation for further work:
  - Additional research and evidence are needed to determine whether the low incidence of contagion in recent years reflects a permanent change versus a temporary pattern driven by anticipation and market conditions.
- Interim practical stance:
  - Maintain a prudent working assumption that contagion could re-emerge in the medium term.

*Source: _pdp01 - Section 2 (PDF chapter/section).*

### Section 3

### _pdp01 - Section 3

### Content summary
- Section 3 consists of the bibliography/references cited in the chapter.
- The section lists academic articles, IMF reports, working papers, and other sources related to financial contagion, currency crises, emerging-market crises, and related topics.
- The section itself does not present substantive findings, projections, analysis, or policy recommendations; it provides source citations supporting the chapter's analysis.

*Source: _pdp01 - Section 3*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/pdp/2006/_pdp01.pdf_