Contagion
- Defining Contagion
- Causes and Channels of Contagion
- Examples of Crisis Contagion
- Policy Implications
In economics, contagion refers to the spread of a crisis from one country to others. Such a spread can show up in a variety of ways, such as its effects on interest rates, asset prices, exchange rates, capital flows, or the probability of a crisis. Economic usage of the word contagion draws on the analogy to the spread of contagious diseases. Thus the currency crisis in Thailand that spread tomuch of the rest of East Asia in 1997 was often referred to as the ‘‘Asian flu’’ or the ‘‘Asian contagion.’’ The use of the term contagion became particularly popularwith the crisis inMexico at the end of 1994,whichwas felt throughoutmost of Latin America and became known as the ‘‘tequila crisis.’’ Academic attention to contagion had been stimulated by the earlier crises in the European Monetary System in 1992 and 1993.
Defining Contagion
There is considerable debate among economists about how to define contagion. Some hold that it refers to any general transmissionof shocks from one market to another. Others argue that contagion exists only if there is a significant increase in cross-market relationships beyondwhat can be explained by the fundamental state of a country’s economy (‘‘fundamentals’’). Still others contend that contagion spreads only through certain channels. In an influential analysis, the economist Paul Masson (1999) defined contagion as a consequence of sudden shifts in market expectations and investors’ confidence. Such shifts are particularly relevant in the context of second-generation crisismodels inwhich a country’s fundamentals can be in an intermediate vulnerable zone between good and bad. In such situations, a crisis in one country can directly affect expectations of the situation in other countries, and hencemove an economy froma good equilibriumto a bad one and end in a crisis. Masson distinguishes contagion from other types of shocks. He calls a common external shock that affects a group of countries ‘‘monsoonal effects,’’ referring to the example of a monsoon season affecting several countries in a region. This type of effect,Masson argues, should not be considered contagion. Another example of amonsoonal effect is a shock to oil prices or supply that affects many countries at the same time. Still another is the effect on developing countries of major changes in the interest rates of large industrial countries. Both of the latter two examples helped contribute to the Latin American debt crises of the 1980s.
During noncrisis periods, contagion is usually related to the linkages of the economic fundamentals between one ormore countries, and is typically called interdependence. Possible factors leading to contagion include trade linkages, common external shocks, and political or economic policies. Contagion is more frequent during crisis periods than noncrisis periods, however, and is an important explanation of emerging market crises of the late 20th and early 21st centuries. Crisis contagion beyond what can be explained by fundamental links is often called ‘‘pure contagion.’’ ‘‘Unjustified’’ contagion refers to pure contagion resulting from panic and other forms of irrationalities and market imperfections. No commonly accepted name has yet been given to rational pure contagion that is transmitted through financial channels such as portfolio rebalancing.Acountry can be an ‘‘innocent victim’’ of fundamental-based contagion or of pure contagion. For example, the contagion effect from Argentina to its small neighbor Uruguay in 2001 was due primarily to economic interdependence, while the spread of the Asian crisis has often been attributed to panic contagion. The degree of a country’s innocence and the degree to which contagion is unjustified by fundamentals are often matters of considerable dispute.
Causes and Channels of Contagion
Contagion can be transmitted through real or financial linkages, or through other channels such as global investors’ behavior and market imperfection. Causes of contagion are often closely related to its transmission channels. Real linkages, such as trade, economic policies, shared fundamentals, and common shocks, all lead to increased interdependence among economic and financial markets. Contagion effects generally have a strong regional concentration (the Russian crisis in 1998 is a major exception). This has led many economists to stress the role of direct macroeconomic linkages, especially through international trade, as an important channel for contagion.
If contagion is taken to include the general interdependence among economic fundamentals, it can be measured using comovements of financial or economic variables. Pure contagion measurements typically test for cross-country relationships beyond fundamentals. Treatments of what are considered fundamentals also vary substantially. Typically, the broader the range of factors that analysts consider fundamentals, the less pure contagion they see.
Contagion also can spread through financial connections. Global portfolio diversification allows large capital flows to enter emergingmarkets but also makes inflows quickly become outflows during a crisis. Such abrupt reversals of capital flows are often called ‘‘sudden stops.’’ In order to maintain the liquidity of their portfolios, institutional investors tend to sell their cross-border investments in one market when another market is in trouble. Portfolio diversification and rebalancing alone are often enough to explain contagion effects without invokingmarket imperfections. Similarly, common lender problems also lead banks to rebalance their assets portfolio and refuse to roll over debt to countries with risk characteristics similar to those countries in default. If several countries share the same type of investors, correlations among these countries’ financial assets increase, even among countries with few economic similarities. If investors trade only certain categories of assets, the probability of selling out that type of asset during a crisis also increases.
Financial market imperfections, such as asymmetric information, a situation in which one party hasmore information than another, and ill-designed incentive structures for fund managers, can also contribute to the spread of a crisis fromonemarket to another. When a market is less transparent and information is not readily available, investors aremore likely to herd. When investors are relatively uninformed theymay assume that if someone else takes an action, it is because that person has better information. Such behaviorwould increase the size ofmarket swings.
Fund managers’ compensation contracts, even if optimal at the firmlevel,may lead to inefficiencies at the macroeconomic level. Fund managers, for example, are usually evaluated on short time frames, which restrict them from taking advantage of mispricing. Moreover fund managers are typically evaluated against a benchmark, encouraging them to form a portfolio very close to that benchmark or to follow the actions of other fund managers. Such ‘‘separation of brain and resources’’ contributes to fund managers’ pursuing short-term profits and herding, which can cause cross-border contagion.
Owing to similarities among the countries, their geographic proximity, and limited information, investorsmay not always be able to differentiate among common shocks and country-specific shocks. As a result, theymightmake heuristic decisions. Herding and rational ignorance are such ‘‘rule of thumb’’ responses to market imperfections, and contagion is the consequence.
Investors’ sentiments panic, overoptimism, overreaction, and shifts in expectation can also lead to contagion.During a crisis, investorsmight panic and pull out funds in the entire crisis region regardless of each country’s fundamentals. They can be overly optimistic before a crisis. A crisis in one country might force them to reevaluate investments that they initially should have investigated more thoroughly. Thus a crisis in one country can serve as a wake-up call prompting a reevaluation of investments in other countries with shared characteristics.
Of course, if markets were already fully efficient there would be no need for such reevaluations. However, many economists and finance experts are now giving much greater attention to ways in which financial markets may be less than ideally efficient. Such analysis is called behavioral finance. It varies from emphasis on external constraints such as the high cost of information to possible psychological biases. Behavioral analysis is still at an early stage of development, but it promises a number of fruitful insights into the behavior of actual financialmarkets that fall between the extremes of ideally efficient markets andwildly irrational ones (see Willett 2000).
Examples of Crisis Contagion
Discussions of contagion often suggest that the spread of crises is especially associated with the effects of panic in financial markets. This panic contagion view has become particularlywidespread in light of the 1997 98 Asian crisis as the countries hit had strong macroeconomic fundamentals. Subsequent analyses, however, showed thatmany of the countries in crisis had microeconomic and financial weaknesses that suggested some justification for the spread.
Since many crisis countries in Asia had impressive economic growth records, balanced government budgets, and conservative monetary policies, some economists started to analyze the role market psychology and investors’ behavior played during the crisis. South Korea is a prime example. Despite its good macroeconomic fundamentals, Korea suffered devastating losses during the crisis largely because of financial andmicroeconomicweaknesses. Because of implicit guarantees against large depreciation of the exchange rate and the prospect of government bail out of failing institutions, the banking sector in South Korea had incentives to engage in excessively risky loans. As a result it became burdened with nonperforming loans while the corporate sector took on large dollar-denominated, short-term debt. Similarly, a large number of Indonesian and Thai corporations had been borrowing heavily in U.S. dollars. They and their overseas investors had believed that their countries’ exchange rates were safe from large depreciations. When this assumption proved false, therewas a rush to cover their positions, which in turn generated large capital outflows.
Furthermore, while many initial analyses treated the whole set of crises as a single event, the crisis developed over a period of months and had several distinct phases. The first phase, between July and October 1997, started with the run on theThai baht. Its subsequent depreciation spread mainly to Indonesia, Malaysia, and the Philippines, all of which were forced to allow large depreciations of their currencies. The financial markets of a much broader group of countries felt ripples, but did not experience strong speculative attacks or capital flight. In the second phase, which started in October 1998, the devaluation of the Taiwan dollar generated fears that the Hong Kong dollar would follow. The Hong Kong stockmarket lost 40 percent of its value, which led to the devastating devaluation of the South Korea won in less than twomonths.While Korea felt some immediate repercussions from the Thai crisis, they were relatively mild. The strong speculative attacks on the won did not occur untilmonths later. Simple panic contagion would not explain such a protracted spread.
The Asian financial crisis drove down raw material prices. This had severe consequences for Russia, which was highly dependent on exports of raw materials. Even though interest rates soared to 150 percent to attract investors to buy government bonds, bymid-1998Russia was in need of help from the InternationalMonetary Fund (IMF) to maintain its exchange rate. However, the Russian government failed to implement a realistic budget and necessary legislation to meet IMF requirements. Fearing the IMFmight pull the plug on its loans, global investors continued to flee Russia, and Russia soon after defaulted on its debt, sending shock waves through financial markets across the globe.
The financial market expected a substantial devaluation of the ruble, but most observers believed that Russia was too important geopolitically to be allowed to default. Thus markets were stunned. Combined with the near collapse of the hedge fund Long-Term Capital Management in the United States, the Russian default set off a shift to extreme risk aversion among investors. Risk premiums rose substantially, and emerging market countries found it difficult or impossible to find investors for new international credit instruments for many months.
Policy Implications
The best ways to prevent contagion and to deal with it when it occurs vary depending on its causes and channels. A number of policies that are good in their own right, such as the promotion of sound economic fundamentals and the development of better policies in both the public and private sector for assessing and managing risk, also help reduce problems frominflation.The role played by contagionduring the crises of the 1990s prompted many proposals for greater national and international control over international financial flows.
Confusion has resulted from the tendency of the popular press to use the term contagion to describe any effects of a crisis in one country on currency or financial markets of other countries. The crisis in Iceland in 2006 was felt in currency markets as far away as Eastern Europe and Africa, but these fairly mild ripple effects from Iceland were of a quite differentmagnitude than the devastating crises that hit a number of Asian countries after Thailand’s currency crisis.The global repercussions of the Russian default in 1998 fell between these extremes.
The available evidence clearly suggests that major shocks often do cause indiscriminant contagion in financial markets, but with the exception of the Russian default this tends to last for only a brief period,measured in hours, days, or weeks.Mediumterm responses from the currency and financial markets tend to be much more differentiated, although the determinants of these more focused medium- term responses can be quite complex, including both trade and financial linkages and a broad range of fundamentals including financial and political as well as economic considerations. Since information is often quite imperfect, perception of the fundamentals can shift without any change in the actual fundamentals themselves. Thus a crisis in one country can act as a wake-up call that generates reevaluations of conditions in other countries. The Thai crisis provides a vivid example.
In general, the fallout on other countries and markets from the crises occurring since the Russian default have beenmuchmilder, and as a consequence support for major reforms in the international financial architecture has fallen off.Many economists, however, still believe that there is a strong case for strengthening the capabilities of the IMF to act as an international lender of last resort to deal with liquidity crises, especially those that would have the potential to generate considerable contagion. See also asymmetric information; banking crisis; capital flight; capital flows to developing countries; currency crisis; financial crisis; hedge funds; hot money and sudden stops; international financial architecture; International Monetary Fund (IMF); International Monetary Fund conditionality; International Monetary Fund surveillance; Latin American debt crisis; lender of last resort; sovereign risk; spillovers