Three essays on capital flows, banking weakness, and real exchange rates in East Asia and Latin America

The dissertation consists of an introduction, three essays, and conclusions. It focuses on the experiences of East Asian and Latin American economies (Thailand, Indonesia, Philippines, Malaysia, Argentina, Brazil, Mexico, Chile).; The first essay addresses real exchange rate movements in the long run for East Asia and for Latin America. We use Johansen multivariate cointegration tests and study the causes of such possible cointegration and its implications. The results show that real effective exchange rates across East Asian countries are tied by only one long run cointegration vector while they are not cointegrated at all across Latin American countries. We argue that what connects real effective exchange rates in East Asia are trade policies directed at export promotion.; The second essay explores the following empirical and theoretical issues. First, we empirically investigate the effect of moral hazard on overlending in East Asia and Latin America. The results demonstrate that the surge of capital inflows and the high level of loanable capacity that has been associated with largely deregulated financial systems played an important role in generating overlending. Moreover, the tequila effect is positive and relatively high in East Asian countries indicating that the crisis led to an increase in overlending in order to bail out the unsound firms. Second, we present a simple theoretical model that explains the rationale behind the monetary authority's reluctance to take prompt corrective actions in troubled financial intermediaries, which greatly increases the costs of resolution. The model also explores the relationship between monetary authority behavior, overbanked system, and the conditions under which this system enters into the crisis zone in both developed and developing countries.; The third essay uses structural vector autoregressive models to address the following questions. What are the sources of overlending and real exchange rate appreciation in both crisis and non-crisis East Asian and Latin American countries? How does monetary policy react to shocks caused by a surge of capital inflows, an overlending, or a real exchange rate appreciation. The results demonstrate the following. First, variance decompositions for lending and the relevant impulse response functions show that the banking systems in East Asian countries seem to be more vulnerable than those in Latin American countries to a sudden depreciation in real effective exchange rate. This usually occurs when the private sector undertakes substantial unhedged loans. Mexico and Chile experienced no major changes in domestic policies in the period under consideration (like the currency board regime in Argentina and the Real Plan in Brazil), and therefore to some extent look like East Asian countries in terms of this vulnerability. Second, the overlending does not play a notable role in explaining the movements in the real exchange rate in Philippines, Malaysia, Mexico, and Chile (mostly non-crisis countries). In contrast it plays a significant role in Thailand, Indonesia, Brazil, and Argentina. Third, monetary policy reactions in non-crisis countries are similar in both regions. Monetary policy reacts countercyclically to a real exchange rate appreciation shock in Malaysia, and Chile. It reacts to capital inflows or lending in almost all the crisis countries.