9511252 Eichengreen Recent turbulence in foreign exchange markets, especially within the European Monetary System (ERM), has raised new questions about the causes of speculative attacks on pegged exchange rates. It is surprising in this light that we do not possess a body of evidence that establishes stylized facts about the behavior of macroeconomic variables around the time of attacks. The goal of this project is to develop stylized empirical facts about the behavior of critical economic and political variables around the time of speculative attacks on pegged currencies. It will do so by developing a data set suitable for studying the causes and consequences of speculative crises and subjecting it to univariate and multivariate statistical analysis. The project plans to address questions such as the following: Are there differences in the behavior of key macroeconomic variable sin periods prior to speculative attacks on pegged exchange rates compared to other periods? Are there "early warning signals" of pending currency crises? Does the behavior of these key variables change in the aftermath of speculative attacks? Are political outcomes affected by speculative attacks? Do answers to these questions different in the different times and places in which exchange rates are pegged? Do they differ for ERM and non- ERM currencies, in particular? Or are speculative attacks all alike?