Does Machine Learning Help us Predict Banking Crises?
Johannes Beutel, Sophia List, Gregor von Schweinitz
Journal of Financial Stability,
December
2019
Abstract
This paper compares the out-of-sample predictive performance of different early warning models for systemic banking crises using a sample of advanced economies covering the past 45 years. We compare a benchmark logit approach to several machine learning approaches recently proposed in the literature. We find that while machine learning methods often attain a very high in-sample fit, they are outperformed by the logit approach in recursive out-of-sample evaluations. This result is robust to the choice of performance metric, crisis definition, preference parameter, and sample length, as well as to using different sets of variables and data transformations. Thus, our paper suggests that further enhancements to machine learning early warning models are needed before they are able to offer a substantial value-added for predicting systemic banking crises. Conventional logit models appear to use the available information already fairly efficiently, and would for instance have been able to predict the 2007/2008 financial crisis out-of-sample for many countries. In line with economic intuition, these models identify credit expansions, asset price booms and external imbalances as key predictors of systemic banking crises.
Read article
26.06.2019 • 14/2019
Study: How financial crises lower life satisfaction and how to prevent this
Financial crises not only result in severe disruptions to the economic system, they also affect people’s life satisfaction. A new study by Martin Luther University Halle-Wittenberg (MLU) and the Halle Institute for Economic Research (IWH) shows that weaker members of society are more affected by increased uncertainty during crisis times, even if they may not be speculating on the stock market themselves. This could potentially also lower their propensity to consume, thereby intensifying the impact of a financial crisis. The study was recently published in “The B.E. Journal of Economic Analysis & Policy”.
Lena Tonzer
Read press release
An Evaluation of Early Warning Models for Systemic Banking Crises: Does Machine Learning Improve Predictions?
Johannes Beutel, Sophia List, Gregor von Schweinitz
Abstract
This paper compares the out-of-sample predictive performance of different early warning models for systemic banking crises using a sample of advanced economies covering the past 45 years. We compare a benchmark logit approach to several machine learning approaches recently proposed in the literature. We find that while machine learning methods often attain a very high in-sample fit, they are outperformed by the logit approach in recursive out-of-sample evaluations. This result is robust to the choice of performance measure, crisis definition, preference parameter, and sample length, as well as to using different sets of variables and data transformations. Thus, our paper suggests that further enhancements to machine learning early warning models are needed before they are able to offer a substantial value-added for predicting systemic banking crises. Conventional logit models appear to use the available information already fairly effciently, and would for instance have been able to predict the 2007/2008 financial crisis out-of-sample for many countries. In line with economic intuition, these models identify credit expansions, asset price booms and external imbalances as key predictors of systemic banking crises.
Read article
Crises and Rescues: Liquidity Transmission Through Global Banks
Michael Koetter, Claudia M. Buch, C. T. Koch
International Journal of Central Banking,
No. 4,
2018
Abstract
This paper shows that global banks transmit liquidity shocks via their network of foreign affiliates. We use the (unexpected) access of German banks' affiliates located in the United States to the Federal Reserve's Term Auction Facility. We condition on the parent banks' U.S. dollar funding needs in order to examine how affiliates located outside the United States adjusted their balance sheets when the U.S. affiliate of the same parent tapped into TAF liquidity. Our research has three main findings. First, affiliates tied to parents with higher U.S. dollar funding needs expanded their foreign assets during periods of active TAF borrowing. Second, the overall effects are driven by affiliates located in financial centers. Third, U.S.- dollar-denominated lending particularly increased in response to the TAF program.
Read article
15.06.2017 • 26/2017
Ailing banks increase leverage of ailing firms
Euro area countries such as Greece and Spain continue to struggle not only with their banks, but also with highly indebted domestic firms. Michael Koetter from the Halle Institute for Economic Research (IWH) and co-authors show the failure to resolve banks’ financial difficulties also prevents debt reduction of over-leveraged firms – and sometimes even contributes to increasing leverage of the weakest firms.
Michael Koetter
Read press release
24.04.2017 • 22/2017
Higher capital requirements: It’s the firms that end up suffering
61 European banks were scheduled to increase their capital cover by 2012 to provide a sufficient buffer for future crises. As the study by the research group chaired by Reint E. Gropp at the Halle Institute for Economic Research (IWH) – Member of the Leibniz Association shows, the banks did implement these requirements – not by raising their levels of equity, but by reducing their credit supply. This resulted in lower firm, investment, and sales growth for firms which obtained a larger share of their bank credit from these banks.
Reint E. Gropp
Read press release
On the Nonexclusivity of Loan Contracts: An Empirical Investigation
Hans Degryse, Vasso Ioannidou, Erik von Schedvin
Management Science,
No. 12,
2016
Abstract
We study how a bank's willingness to lend to a previously exclusive firm changes once the firm obtains a loan from another bank ("outside loan") and breaks an exclusive relationship. Using a difference-in-difference analysis and a setting where outside loans are observable, we document that an outside loan triggers a decrease in the initial bank's willingness to lend to the firm, i.e., outside loans are strategic substitutes. Consistent with concerns about coordination problems and higher indebtedness, we find that this reaction is more pronounced the larger the outside loan and it is muted if the initial bank's existing and future loans retain seniority and are protected with valuable collateral. Our results give a benevolent role to transparency enabling banks to mitigate adverse effects from outside loans. The resulting substitute behavior may also act as a stabilizing force in credit markets limiting positive comovements between lenders, decreasing the possibility of credit freezes and financial crises.
Read article
Friend or Foe? Crowdfunding Versus Credit when Banks are Stressed
Daniel Blaseg, Michael Koetter
IWH Discussion Papers,
No. 8,
2015
Abstract
Does bank instability push borrowers to use crowdfunding as a source of external finance? We identify stressed banks and link them to a unique, manually constructed sample of 157 new ventures seeking equity crowdfunding. The sample comprises projects from all German equity crowdfunding platforms since 2011, which we compare with 200 ventures that do not use crowdfunding. Crowdfunding is significantly more likely for new ventures that interact with stressed banks. Innovative funding is thus particularly relevant when conventional financiers are facing crises. But crowdfunded ventures are generally also more opaque and risky than new ventures that do not use crowdfunding.
Read article
Are Universal Banks Bad for Financial Stability? Germany During the World Financial Crisis
Diemo Dietrich, Uwe Vollmer
Quarterly Review of Economics and Finance,
No. 2,
2012
Abstract
This case study explores the contribution of universal banking to financial stability in Germany during the recent financial crisis. Germany is a prototype for universal banking and has suffered from a rather small number of banking crises in the past. We review the banking literature and analyze the major institutional and regulatory features of the German financial system to establish a nexus between universal banking and stability.
Read article