Athens University Economics and Business (AUEB)
Early warning systems for banking crises: evidence for the euro-zone banking sector
Abstract
dc:descriptionIn the last decade, there has been a dramatic increase in the literature on banking crises, with researchers focusing on developing mechanisms, such as Early Warning Systems, for predicting systemic or individual banking distress. The aim of the present study is to propose an Early Warning System for banking crises in the Eurozone. For this purpose, data from the Macroprudential Database of the European Central Bank (ECB) are used. The first part of the study reviews the landmarks in the literature associated with Early Warning Models for banking crises, with a focus on the most important surveys and methods. Additionally, the frame for the empirical analysis is set. In the second part, we proceed to an application of the most widespread and popular method used in the literature of Early Warning Systems, the binary multivariate logistic regression. The initial implementation takes place in ten Eurozone banking systems for which data were fully available, for the period from 1999 to 2016, in the ECB Macroprudential Database. Traditional macroeconomic parameters, banking indicators as well as, for first time, two composite risk indicators provided by the MPDB, an investment sentiment indicator (Eurostat) and an economic freedom indicator (Heritage Institute) are employed as explanatory variables. A binary dependent variable is used for the banking crisis and takes the value "1" during the "early warning periods" that we aim to predict and define from one to four years before the crisis onset. A series of robustness and out of sample tests on our model concludes this part. In the third part, we apply a more complex method of logit, the multinomial logit model, on our initial data sample, employing the same explanatory variables, but taking into account separately the crisis observations; a period that is characterized and analyzed as "crisis duration period" and lasts until tranquil periods return. Most of the robustness tests of the previous part are repeated here, as well as the model specification that we apply to the multivariate binary logit model in the second part of the study. Overall, the results indicate that our binomial EWM with three years early warning period, excluding the observations following the onset of crises, performs clearly better and is the superior Early Warning logit model of this study for the successful prediction of a systemic banking crisis -with the characteristics of 2008 Euro-zone banking crisis three years before the explosion. Additionally, we prove that the risk indicators provided by the new Macroprudential Database of the European Central Bank are a very useful tool for forecasting systemic banking crises. Subsequently, the establishment of the MPDB (in 2015), was an important development towards even more efficient central bank supervision in Eurozone.
Degree
thesis:*- Grantor dc:publisher
- Athens University Economics and Business (AUEB)
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Filippopoulou, Chryssanthi
- Φιλιπποπούλου, Χρυσάνθη
Subjects
dc:subject × 14- Συστήματα έγκαιρης προειδοποίησης
- Ευρωπαϊκή Κεντρική Τράπεζα
- Συστημικός κίνδυνος
- Μοντέλα λογιστικής συνάρτησης
- Μακροπροληπτική βάση δεδομένων
- Early warning systems
- European Central Bank
- Systemic risk
- Logit models
- Macroprudential database
- Κοινωνικές Επιστήμες
- Οικονομικά και Επιχειρήσεις
- Social Sciences
- Economics and Business
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/48467