Presidential Address: Globalization Hazard and Delayed Reform in Emerging Markets

Capital inflows to emerging market economies rose to unprecedented heights in the first part of the 1990s and then collapsed very rapidly in the second. Such volatility could partly be explained by financial vulnerability in the emerging markets themselves, but the global nature of the phenomenon raises the suspicion that the world financial market is wrought with systemic problems that are largely independent of the individual countries affected. This paper puts forward the conjecture that phenomena such as contagion could stem from the way the capital market operates (for example, crises generated by margin calls). These systemic phenomena require systemic instruments. Unfortunately, few are available. The International Monetary Fund (IMF) operates more like a fire department than like a central bank. Liquidity is sprayed where fire is found, not on the system as a whole in the manner of a central bank faced with a liquidity crisis. ¹

The combination of domestic financial vulnerability and the lack of a worldwide safety net gives rise to what I call globalization hazard, that is, risk generated by the sudden large expansion of credit to emerging market economies in the first half of the 1990s, probably as a result of imperfect information and underdeveloped financial institutions. Several recent financial crises were low-probability events that were uninsured and perhaps uninsurable in the private sector, and they called for ex post government intervention. Government intervention, however, represents a major roadblock in the presence of delayed reform, a condition in which the government delays the implementation of socially desirable reform and wealth redistribution. Delayed reform may thus exacerbate the impact of low-probability events and possibly help to coordinate expectations on "bad" equilibria, contributing to the severity of globalization hazard.

The policy implications of the globalization hazard view are diametrically opposed to those of the moral hazard view recently popularized by Meltzer. ² This makes the present discussion greatly relevant for the design of a new financial architecture, an issue of enormous urgency and importance.

Moral Hazard versus Globalization Hazard

A salient characteristic of currency crises after the 1994 Mexican crisis (the so-called tequila crisis) is their frequent recurrence. The tequila crisis was followed by massive crises in Asia (1997), Russia (1998), Ecuador (2000), and Turkey (2001), as well as the protracted crisis in Argentina (2000-02). With the exception of Argentina, these crises have been relatively short-lived, especially compared to the debt crisis in the 1980s. However, they followed each other domino fashion. Why?

A leading explanation is moral hazard. According to this point of view, large and timely bailout packages, orchestrated by the IMF, allowed fixed-income investors to exit the market following the occurrence of each crisis without suffering major capital losses, even though the rate of return on these assets far exceeded those of safe assets like U.S. Treasury bills. The expectation that future crises would be resolved in the same manner emboldened fixed-income investors to take high risks in other emerging market economies, thereby increasing the probability of a crisis. Plausible as it sounds, however, this view has slim empirical support. In the first place, as shown in figure 1, net private capital flows to emerging markets started to subside after 1995, a trend that is even sharper for portfolio flows (see figure 2). ³ Second, after the tequila crisis the composition of capital flows shifted in favor of foreign direct investment (FDI) (see figure 3). Investors in those and related assets (stocks) suffered major losses during crises and thus cannot easily be claimed to have greatly benefited from bailout packages.

Questioning the moral hazard view is not tantamount to saying that policymakers and investors will not take advantage of generous giveaways, but the existence of distortion-driven behavior does not prove that distortions are seriously costly. The moral hazard view claims that bailouts by the Group of Seven (G-7) countries are a major cause of both the succession of...