Banking Crises of the 1990s and 2000s in Developed Countries: How similar are they?
Thouraya Boudebbous
Abstract
Thouraya Boudebbous
Abstract
Abstract: The objective of this study is, on one hand, to determine whether the banking crises occurring in advanced economies in the 1990s and 2000s share the same roots, and on the other hand, whether aggregated accounting indicators are good predictors of crises in these economies. By means of the multivariate logit model, we have identified banking crises indicators for a set of 16 developed countries for the periods 1990-2006 and 2007-2012. Our results show the existence of certain similarities between the crises of the 1990s and 2000s, namely: a private credit boom and a deterioration of banks' balance sheets. In addition, we have tested the robustness of our results through the use of Bayesian averaging models. Our results have allowed us to confirm, in general, the robustness of the estimation results derived from the multivariate logit approach.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Abstract: The objective of this study is, on one hand, to determine whether the banking crises occurring in advanced economies in the 1990s and 2000s share the same roots, and on the other hand, whether aggregated accounting indicators are good predictors of crises in these economies. By means of the multivariate logit model, we have identified banking crises indicators for a set of 16 developed countries for the periods 1990-2006 and 2007-2012. Our results show the existence of certain similarities between the crises of the 1990s and 2000s, namely: a private credit boom and a deterioration of banks' balance sheets. In addition, we have tested the robustness of our results through the use of Bayesian averaging models. Our results have allowed us to confirm, in general, the robustness of the estimation results derived from the multivariate logit approach.
Key concepts: Business, Economics