Date of Award

Fall 2022

Degree Type

Open Access Dissertation

Degree Name

Economics, PhD


School of Social Science, Politics, and Evaluation

Advisor/Supervisor/Committee Chair

Thomas Willett

Dissertation or Thesis Committee Member

Levan Efremidze

Dissertation or Thesis Committee Member

John Rutledge

Terms of Use & License Information

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Rights Information

© 2022 Istvan Zambori


Currency Crises, Economic Factors, Exchange Rate Crises, Floating Exchange Rate, Survival Analysis, Time Duration Analysis

Subject Categories



Currency crises can wreak havoc on countries economically, financially, and politically. This study has been conducted in the spirit of attempting to find a general notion of why such crises occur. The aim is to describe, based on empirical evidence, a general time dependent evolution of currency crisis probabilities. Furthermore, the goal is to uncover general and systemic variables driving currency crises. The research utilizes survival analysis. Durations of tranquility are used to determine probabilities of crises. In turn, economic and financial variables are used to test possible linkages to these crises. In the study, durations of tranquil periods are defined by the absence of a downward pressure on the domestic currency. Pressure periods are defined by an ‘Exchange Market Pressure’ indicator. The end of a tranquil period is a crisis and probabilities of these are determined through the model. The study revealed that the evolution of crisis probabilities over time are generally upward sloping. In the model these are shown by the hazard functions increasing. This implies that the more time a country in a tranquil period the higher the chances that it undergoes a crisis in its currency market. The conclusion holds for countries that survived to a particular time thus, these probabilities are conditional. This suggests that time spent in a tranquil period does not lower the probability of having a currency crisis. In the context of survival analysis explanatory variables are hypothesized to influence currency crises. Nine variables are included in the study: current account, direct investment, portfolio investment, IMF credit as percent of quota, real GDP growth, real effective exchange rate, unemployment, the LIBOR and the VIX. Of the variables included two have shown promise in robustly affecting currency crises probabilities. These were unemployment, and real GDP growth. Results suggest that a higher level of unemployment is expected to lower currency crises probabilities while increased real GDP is shown to increase them. In the last section of the study country specific hazard graphs are drawn to compare the evolution of their respective crisis probabilities to the ‘average’ country. The comparison may aid investors in gauging the relative hazard of depreciation across countries and relative potential loss across currencies.



Included in

Economics Commons