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
Too Connected to Fail? Inferring Network Ties from Price Co-movements
Jakob Bosma, Michael Koetter, Michael Wedow
Journal of Business and Economic Statistics,
No. 1,
2019
Abstract
We use extreme value theory methods to infer conventionally unobservable connections between financial institutions from joint extreme movements in credit default swap spreads and equity returns. Estimated pairwise co-crash probabilities identify significant connections among up to 186 financial institutions prior to the crisis of 2007/2008. Financial institutions that were very central prior to the crisis were more likely to be bailed out during the crisis or receive the status of systemically important institutions. This result remains intact also after controlling for indicators of too-big-to-fail concerns, systemic, systematic, and idiosyncratic risks. Both credit default swap (CDS)-based and equity-based connections are significant predictors of bailouts. Supplementary materials for this article are available online.
Read article
Avoiding the Fall into the Loop: Isolating the Transmission of Bank-to-Sovereign Distress in the Euro Area and its Drivers
Hannes Böhm, Stefan Eichler
Abstract
We isolate the direct bank-to-sovereign distress channel within the eurozone’s sovereign-bank-loop by exploiting the global, non-eurozone related variation in stock prices. We instrument banking sector stock returns in the eurozone with exposure-weighted stock market returns from non-eurozone countries and take further precautions to remove any eurozone crisis-related variation. We find that the transmission of instrumented bank distress, while economically relevant, is significantly smaller than the corresponding coefficient in the unadjusted OLS framework, confirming concerns on reverse causality and omitted variables in previous studies. Furthermore, we show that the spillover of bank distress is significantly stronger for countries with poorer macroeconomic performances, weaker financial sectors and financial regulation and during times of elevated political uncertainty.
Read article
Upturn Loses Momentum – World Economic Climate Grows Harsher
Roland Döhrn, Oliver Holtemöller, Stefan Kooths, Claus Michelsen, Timo Wollmershäuser
Wirtschaftsdienst,
No. 10,
2018
Abstract
Der Aufschwung in Deutschland hält bereits seit mehr als fünf Jahren an. Er hat allerdings an Fahrt eingebüβt. Dies hat sowohl nachfrageseitige als auch produktionsseitige Gründe. Zum einen hat sich das Auslandsgeschäft im Einklang mit der Verlangsamung der Konjunktur in den wichtigsten deutschen Absatzmärkten abgeschwächt. Zum anderen sehen sich Unternehmen zunehmend Engpässen gegenüber, vor allem bei Arbeitskräften und beim Bezug von Vorleistungsgütern. Überlagert wird dies durch Probleme in der Automobilindustrie im Zusammenhang mit der Einführung des neuen Prüfverfahrens WLTP. Aufgrund der hohen gesamtwirtschaftlichen Bedeutung der Branche hinterlassen diese sichtbare Spuren beim Zuwachs des Bruttoinlandsprodukts. Die Umstellungsprobleme dürften aber im Winterhalbjahr überwunden werden. Hinzu kommen Impulse aufgrund von finanzpolitischen Maβnahmen, die zum Jahresbeginn 2019 in Kraft treten.
Read article
SMEs and Access to Bank Credit: Evidence on the Regional Propagation of the Financial Crisis in the UK
Hans Degryse, Kent Matthews, Tianshu Zhao
Journal of Financial Stability,
2018
Abstract
We study the sensitivity of banks’ credit supply to small and medium size enterprises (SMEs) in the UK with respect to the banks’ financial condition before and during the financial crisis. Employing unique data on the geographical location of all bank branches in the UK, we connect firms’ access to bank credit to the financial condition (i.e., bank health and the use of core deposits) of all bank branches in the vicinity of the firm for the period 2004–2011. Before the crisis, banks’ local financial conditions did not influence credit availability irrespective of the functional distance (i.e., the distance between bank branch and bank headquarters). However, during the crisis, we find that SMEs with banks within their vicinity that have stronger financial conditions faced greater credit availability when the functional distance is close. Our results point to a “flight to headquarters” effect during the financial crisis.
Read article
Kommentar: Mit bester Absicht in die Krise
Reint E. Gropp
Wirtschaft im Wandel,
No. 4,
2018
Abstract
Zehn Jahre nach der Lehman-Pleite werden die Finanzmärkte besser kontrolliert denn je. Das kann böse Folgen haben.
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
On DSGE Models
Lawrence J. Christiano, Martin S. Eichenbaum, Mathias Trabandt
Journal of Economic Perspectives,
No. 3,
2018
Abstract
The outcome of any important macroeconomic policy change is the net effect of forces operating on different parts of the economy. A central challenge facing policymakers is how to assess the relative strength of those forces. Economists have a range of tools that can be used to make such assessments. Dynamic stochastic general equilibrium (DSGE) models are the leading tool for making such assessments in an open and transparent manner. We review the state of mainstream DSGE models before the financial crisis and the Great Recession. We then describe how DSGE models are estimated and evaluated. We address the question of why DSGE modelers—like most other economists and policymakers—failed to predict the financial crisis and the Great Recession, and how DSGE modelers responded to the financial crisis and its aftermath. We discuss how current DSGE models are actually used by policymakers. We then provide a brief response to some criticisms of DSGE models, with special emphasis on criticism by Joseph Stiglitz, and offer some concluding remarks.
Read article
