Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Bachelor's Thesis |
| Scope | Discipline-based scholarship |
| Title | Determinants of Currency Crashes |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 27 |
| Date | 2017 |
| Zusammenfassung | The aim of this thesis is to compare the reliability of well-known early warning indicators regarding a currency crash between two periods. I investigated 125 countries between 1991 and 2015 and took the starting of the global financial crises 2008 as separation point between the two periods. Should it be possible to develop a reliable early warning model for the long term, at least the trend of the warning indicators around crashes should be the same in two differing periods, and, hence, in disregard of economic and global circumstances. As these circumstances changed much during and after the 2008 crisis in comparison to the previous period it offers itself to conduct a study on this topic. Another reason for choosing 2008 as the separation point is the fact that between 2003 and 2007 there was low crash activity. Hence, the existing literature on early warning indicators models concentrate mostly on the pre-2002 period. To evaluate the data I first conducted an event study to describe the behaviour of the variables around crashes. Second, I ran a probit regression to look at the marginal contribution of every single variable conditional on the others and so to determine the statistical significance of each indicator. The behaviour of most variables is similar in both periods, but differing in the size of the movement, and the time at which the variable dissolves from the no-crisis observations and becomes worse. Only the worsening in growth of real GDP and total reserves indicate a currency crash early enough in both periods to act on this early warning signal. Noteworthy is the behaviour of the no-crash observations in the 2008 period. It exhibits the same course as the crash observations – lagged by one year. In both periods statistical significant in indicating a currency crash are only total reserves and regional contagion. To develop a reliable early warning model based on only two variables seems to be quite ambitious. Hence, it seems as if the risk of a currency crash needs to be accepted. |
| Export |
BibTeX
|