Who puts our financial system at risk? A methodological approach to identify banks with potential significant negative effects on financial stability

Since the outbreak of the global financial crisis, a number of regulations have been issued to cope with the too-big-to-fail problem and its devastating effects on financial markets, government budgets and the broader economy in general. The aim of these regulations is to contain the risks stemming from large institutions which potentially jeopardize not only these institutions’ own existence but other institutions and segments of the economy as well. In particular, new legislation in macroprudential supervision and resolution that refers to systemically relevant institutions addresses the too-big-to-fail problem. Still, in practice, it is difficult for supervisory authorities to answer the question which institution may really compromise financial stability. The identification of systemically relevant banks is particularly important for banking systems (like the Austrian) with large numbers of banks, where even medium-sized banks might put stress on the entire financial system. Bringing together macroprudential regulations as well as recovery and resolution planning, this methodological paper aims to contribute to the literature and supervisory practice on the identification of systemically relevant banks. We develop a consistent and comprehensive framework that consists of more than 30 quantitative indicators reflecting four key stability criteria: financial market conditions, economic importance, direct contagion and indirect contagion. A particular challenge in this context is the setting of explicit thresholds for each of these indicators. To resolve this issue, we design a methodological approach to calibrating thresholds for different types of indicators: stress indicators, risk exposure indicators, system share indicators and network indicators. We identify thresholds based on quarterly panel data (from 1999 to 2016) for the Austrian banking sector. One basic assumption of our calibration is the idea of substitutability: If market activities of a failing bank can be absorbed promptly by other market participants, financial stability will not be at risk. As the substitution of bank activities also depends on the current phase of the economic cycle, we account for bust phases by developing stress scenarios.