Sudden Stops and Phoenix Miracles in Emerging Markets

A decade has passed since the salvos from Mexico’s Tequila Crisis of 1994–1995 echoed around the financial world. Since then, many more crises have taken place in emerging market economies (EMs). Furthermore, crises have tended to bunch together, bringing to the forefront the systemic nature of these events. True, every new crisis has its own idiosyncratic features, but useful policy lessons must be derived from robust, empirical regularities. This is the research strategy we have pursued in the last few years. We will report on two types of regularities that strike us as highly robust across EM crises: (a) Sudden Stops (of capital inflows) and (b) Phoenix Miracles. A Sudden Stop is a sharp fall in capital inflows relative to their past trajectory. Sudden Stops are not a common feature in developed economies and display a large degree of temporal bunching, suggesting that global capital market turmoil acts as a coordinating factor external to EMs. As shown in Section I, however, balance-sheet effects—namely, the interaction of large changes in the real exchange rate during Sudden Stops and Liability Dollarization (i.e., foreign-exchange-denominated debts)—are key in influencing the likelihood of a Sudden Stop. Thus, even though the initial shock is, in principle, exogenous to the economy, whether or not it materializes into a Sudden Stop depends on domestic financial vulnerabilities. On the other hand, a Phoenix Miracle is defined as a case in which output recovers relatively quickly from a sharp collapse with virtually no recovery in credit or capital inflows, and a very weak recovery in investment— hence the reference to the mythical bird “rising from the ashes.” The existence of phoenix-like recoveries suggests that financial frictions play a key role in pushing economies to the abyss from which, in some way or another, they can crawl back to safe ground by means less than apparent to the conventional observer looking for standard “fundamentals” and, thus, may appear miraculous. Interestingly, the Great Depression of the 1930s shares some of the key features of Phoenix Miracles in EMs, but shows salient differences as well that suggest nominal labor market rigidities are not crucial in explaining output collapse in EMs. Understanding these regularities could, and we believe does, shed light on policies aimed at preventing crises and attenuating their effects.