Calibrating the Magnitude of the Countercyclical Capital Buffer Using Market‐Based Stress Tests

Abstract

This paper proposes a novel methodology to calibrate the magnitude of the cap on the countercyclical capital buffer (CCyB) using market-based stress tests. The macroprudential authority in our paper aims to contain the possibility of a breach of a minimum capital ratio in the event of a severe system-wide shock within a certain permissible failure probability. To meet its objective during periods of challenging macro-financial conditions, the macroprudential authority requires banks to build up the CCyB during credit booms. We show how market-based stress tests can be used to estimate the necessary magnitude of the CCyB. We apply the methodology to major banks in six advanced economies. Our estimates suggest a magnitude of the cap on the CCyB in a range from 1.4 to 1.7 per cent of total assets, depending on the ability of the macro-prudential authority to forecast macrofinancial conditions.

Calibrating the countercyclical capital buffer using AUROCs

“…The Basel framework explicitly prescribes the use of the one side HP filter to extract the cyclical component on national credit aggregates, a cap of 2.5% for the capital add‐on and suggests an activation and termination thresholds of 2 and 10 percentage points, respectively, to map the credit gap into the buffer guide. While the existing literature has extensively studied the reliability of the indicator used to capture the credit cycle (e.g., Alessandri et al, 2022; Aikman et al, 2015; Edge & Meisenzahl, 2011) and the appropriateness of the 2.5% cap (see among others Aikman et al, 2019; and Van Oordt, 2018), only a few studies have called into question the min–max thresholds. The possibility to depart from the standard credit gap based on the one‐side HP filter and from the calibration suggested by the BCBS are provided for by the BCBS itself and, in the European context, by the guidelines of the European Systemic Risk Board.…”

“…These thresholds were calibrated to ensure a good signalling property of the one‐side HP filter based on the historical banking crises experienced by 25 countries (BCBS, 2010). While the existing literature has extensively studied the reliability of the indicator used to capture the credit cycle (Alessi & Detken, 2018; Drehmann & Yetman, 2018; Hamilton, 2018, among others) and the appropriateness of setting the maximum level of the CCyB rate to 2.5% (Aikman et al, 2019; Van Oordt, 2018; Faria‐e‐Castro, 2019, among others), only a few studies have called into question the levels of the min–max thresholds. Detken et al (2014) observe that in 18 European Union (EU) member states the gap exceeded the 10 percentage points upper bound outside crises, both 5 years before a crisis and when a crisis did not occur, while the minimum threshold of 2 percentage points seems broadly adequate.…”