Working Paper: NBER ID: w4898
Authors: Barry Eichengreen; Andrew K. Rose; Charles Wyplosz
Abstract: This paper presents an empirical analysis of speculative attacks on pegged exchange rates in 22 countries between 1967 and 1992. We define speculative attacks or crises as large movements in exchange rates, interest rates, and international reserves. We develop stylized facts concerning the univariate behavior of a variety of macroeconomic variables, comparing crises with periods of tranquility. For ERM observations we cannot reject the null hypothesis that there are few significant differences in the behavior of key macroeconomic variables between crises and non-crisis periods. This null can be decisively rejected for non-ERM observations, however. Precisely the opposite pattern is evident in the behavior of actual realignments and changes in exchange rate regimes. We attempt to tie these findings to the theoretical literature on balance of payments crises.
Keywords: speculative attacks; pegged exchange rates; European Monetary System; macroeconomic variables
JEL Codes: F31; F33
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| deteriorating macroeconomic conditions (E66) | increased likelihood of speculative attacks (F31) |
| macroeconomic variables behave differently during speculative attacks (E19) | behavior of reserves and possibly interest rates (E43) |
| speculative attacks (D84) | behavior of macroeconomic variables (E70) |
| behavior of macroeconomic variables (E70) | speculative crises (G01) |
| speculative crises (G01) | higher inflation and interest rates during realignments (E31) |
| first-generation speculative attack models (D84) | expected causal mechanisms (C22) |
| lack of significant differences in macroeconomic variables (E19) | expected causal mechanisms do not hold (C22) |
| macroeconomic conditions (E66) | speculative attacks (D84) |