Since the Asian crisis in 1997, the term contagion has been used to refer to the spread of financial turmoil across countries. Among economists, there is little agreement on what exactly the term contagion entails. The exam of the literature provides three prevailing definitions¹:
1. Fundamentals-based contagion (interdependence): the contagion is the transmission of global or local shocks across countries through fundamentals (spillover effects). According to this definition, contagion could arise also during stable periods (Calvo and Reinhart (1996), Pristker (2000));
2. Excess of co-movements: the contagion is the transmission of global or local shocks across countries through mechanisms that do not include fundamentals. This type of contagion is considered to be caused by “irrational” phenomena, such as financial panic, herd behaviour, increase in risk aversion or a loss of confidence (Claessens, Dornbusch, Park (2001), Jeanne and Masson (1998));
3. Shift contagion: the contagion is a significant change in cross-market linkages after a shock in an individual country (or group of countries). (Forbes and Rigobon (1998)).
The transmission mechanisms underlying the three definitions could be represented by the following model, partly taken from Forbes and Rigobon (1999), Pristker (2001) and Pericoli and Sbracia (2001):

where:
is the stock price in country i at time t,
is the vector of stock prices in countries different from I ,
is a common aggregate shock linked to fundamentals, and
is an idiosyncratic and independent shock.
On the basis of this equation, it is possible to show how transmission mechanisms work according to the three different definitions of contagion:
the first transmission mechanism of aggregated or specific shocks is measured by and , and the direct effect of these shocks on each country i is embodied respectively by and .
The second transmission mechanism is measured by the correlation of idiosyncratic shocks of different countries , it is interpreted as contagion because there is excess of co-movements that cannot be explained by fundamentals;
the third transmission mechanism is measured by a shift in cross-market linkages and therefore is embodied in changes in both parameters and (i.e. structural break).
The first definition, as noticed by Claessens, Dornbusch and Park (2001), should not be properly considered as contagion. Therefore, it reflects the interdependence between countries that exists in each state of the world. According to the interdependence definition, global or local shocks are transmitted internationally by financial or real channels².
The second definition, as noticed in Pritsker (2001)³, presents two orders of problems: the first is that the finding of contagion can always be questioned on the basis that the correct set of fundamentals was not controlled for (i.e. omitted variables problem). The second is related to the possibility that contagion occurs through the channel that Kodres and Pritsker (2000) refer to as "cross market hedging".
In the cross market hedging models, some operators receive information (information shock) about country-specific components. After the shock, the informed operators will optimally alter their portfolio for the country where the shock occurred. But they will also hedge the change in their macroeconomic risk exposures by rebalancing in other countries. The rebalancing transmits the idiosyncratic shock across markets, generating correlation in short-run stock returns.
The third definition "not only clarifies that contagion arises from a shift in cross market linkages, but it also avoids taking a stance on how this shift occurs". As explained in Forbes and Rigobon (1999), the adoption of the shift contagion definition provides three types of advantages.
Firstly, the test for shift contagion is a test of the effectiveness of international diversification in reducing the portfolio risk during a crisis. Indeed, if shift contagion occurs after a negative shock, it would undermine much of the rationale for international diversification.
Secondly, the definition is useful in evaluating the role and potential effectiveness of international institutions and bailout funds. If shift contagion occurs the transmission mechanisms of the financial crisis do not involve the macroeconomic fundamentals. Thus there will not be any endogenous adjustment process and the intervention of the international organisation is necessary. On the contrary, if we are not in the presence of shift contagion the intervention of the international organisation could be an obstacle to the endogenous adjustment process of the economies.
Thirdly, tests based on this definition provide a useful method to classify theories as those that entail a change in propagation mechanisms after a shock (crisis-contingent theories) versus those which represent a continuation of existing mechanisms (non-crisis-contingent theories).
------------------------------------------------------
¹Pericoli and Sbracia (2001) include five different definitions, the two missing in this paragraph are i) "contagion is a significant increase in the probability of a crisis in one country conditional on a crisis occurring in another country" and ii) "contagion occurs when volatility spills over from the crisis country to the financial markets of other countries".
²The others two categories occur when the transmission of a crisis cannot be linked to observed changes in fundamentals and they result solely from the behavior of investors or other financial agents.
³"I prefer a broad definition because the economics profession will probably never reach agreement on the appropriate set of fundamentals which are needed to make a narrow definition operational"

Bibliography
CALVO  S. and  REINHART C., (1996), Capital flows to Latin America: is there evidence of contagion effects?, World Bank Policy Research W.P. 1619
CLAESSENSl S., DORNBUSH R. and  PARK Y. C. (2001), "Contagion: Why Crises spread and how this can be stopped", in International financial contagion ed. by S. Claessens e K.J. Forbes, Klwer Academic Publishers, Boston.
FORBES K. and  RIGOBON  R., (1999), "No contagion only interdependence: Measuring stock market co-movement", NBER WP, n.7267, July.
FORBES K. and  RIGOBON  R., (2001), Measuring contagion: Conceptual and empirical issues, in International financial contagion ed. by S.Claessens and K.J. Forbes, Klwer Academic Publishers, Boston.
KODRES  L. E. and  PRITSKER M., (1998), A rational expectations model of financial contagion, Board of Governors of the Federal Reserve System, Finance and Economics D.P. 48
MASSON  P., (1998), "Contagion: Monsoonal effects, Spillovers, and jumps between multiple equilibria", IMF WP 98/142, Washington.
MASSON  P., (1999), Contagion: macroeconomic models with multiple equilibria, Journal of International Money and Finance 18, 587-602
PERICOLI  M. and  SBRACIA M., (2001), A primer on financial contagion. Temi di discussione Banca d’Italia, n. 407, June.
PRITSKER  M., (2000), The channels of financial contagion, in International financial contagion, edited by Claessens S. and Forbes K., Kluwer Academic Publishers, Boston/Dodrecht/London.

Editor: Roberta DE SANTIS
© 2009 ASSONEBB