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
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
26.04.2022 • 10/2022
Regional effects of a recession in Germany triggered by an import stop for Russian gas
A halt in Russian gas deliveries would lead to a recession in the German economy. Not all regions would be equally affected: The Halle Institute for Economic Research (IWH) expects a significantly stronger slump in economic output in regions where the manufacturing sector has a large weight than elsewhere.
Oliver Holtemöller
Read press release
East Germany Three Decades After the Wall Came Down: What has Been Achieved and What Should Economic Policy Do?
Reint E. Gropp, Gerhard Heimpold
Wirtschaftsdienst,
No. 7,
2019
Abstract
The persistent difference in productivity between East and West Germany not only results from the relative absence of large firms based in the East as many believe. Companies of all sizes exhibit an East-West productivity gap. The gap is larger in urban regions. Scarcity of skilled labour has emerged as the new barrier to business development. In order to boost productivity, economic policy should avoid additional subsidies that are conditional on creating jobs. Additionally, the potential of East German urban areas should be better explored. Mitigating the shortage in qualified workers requires in-migration of skilled labour from abroad, supported by an open mindset and environment.
Read article
05.05.2022 • 11/2022
IWH-Insolvenztrend: Zahl der Insolvenzen stabilisiert sich im April
Die Zahl der Insolvenzen von Personen- und Kapitalgesellschaften ist im April nicht weiter angestiegen, zeigt die aktuelle Analyse des Leibniz-Instituts für Wirtschaftsforschung Halle (IWH). Für die nächsten beiden Monate ist nicht mit stark veränderten Insolvenzzahlen zu rechnen.
Steffen Müller
Read press release
Does Working at a Start-up Pay Off?
Daniel Fackler, Lisa Hölscher, Claus Schnabel, Antje Weyh
Small Business Economics,
No. 4,
2022
Abstract
Using representative linked employer-employee data for Germany, this paper analyzes short- and long-run differences in labor market performance of workers joining start-ups instead of incumbent firms. Applying entropy balancing and following individuals over ten years, we find huge and long-lasting drawbacks from entering a start-up in terms of wages, yearly income, and (un)employment. These disadvantages hold for all groups of workers and types of start-ups analyzed. Although our analysis of different subsequent career paths highlights important heterogeneities, it does not reveal any strategy through which workers joining start-ups can catch up with the income of similar workers entering incumbent firms.
Read article
The Macroeconomics of Testing and Quarantining
Martin S. Eichenbaum, Sergio Rebelo, Mathias Trabandt
Journal of Economic Dynamics and Control,
May
2022
Abstract
We develop a SIR-based macroeconomic model to study the impact of testing/ quarantining and social distancing/mask use on health and economic outcomes. These policies can dramatically reduce the costs of an epidemic. Absent testing/quarantining, the main effect of social distancing and mask use on health outcomes is to delay, rather than reduce, epidemic-related deaths. Social distancing and mask use reduce the severity of the epidemic-related recession but prolong its duration. There is an important synergy between social distancing and mask use and testing/quarantining. Social distancing and mask use buy time for testing and quarantining to come to the rescue. The benefits of testing/quarantining are even larger when people can get reinfected, either because the virus mutates or immunity is temporary.
Read article
Explaining Regional Disparities in Housing Prices across German Districts
Lars Brausewetter, Stephan L. Thomsen, Johannes Trunzer
IZA Institute of Labor Economics,
March
2022
Abstract
Over the last decade, German housing prices have increased unprecedentedly. Drawing on quality-adjusted housing price data at the district level, we document large and increasing regional disparities: growth rates were higher in 1) the largest seven cities, 2) districts located in the south, and 3) districts with higher initial price levels. Indications of price bubbles are concentrated in the largest cities and in the purchasing market. Prices seem to be driven by the demand side: increasing population density, higher shares of academically educated employees and increasing purchasing power explain our findings, while supply remained relatively constrained in the short term.
Read article