B0885
Title: A statistical approach to limit the effects of pro-cyclicality
Authors: Marcel Brautigam - European Central Bank (Germany) [presenting]
Marie Kratz - ESSEC Business School, CREAR (France)
Michel Dacorogna - Prime Re Solutions (Switzerland)
Abstract: Pro-cyclicality, i.e. the tendency of risk measurements to overestimate future risk in times of crisis, while underestimating it in normal times, is a major problem faced by all financial institutions: insurance companies, banks, regulatory bodies. While various solutions based on a macroeconomic perspective (as is the majority of the academic literature) have been proposed to reduce pro-cyclicality in the context of Basel III, Solvency II, and EMIR, it is tackled differently by taking a statistical point of view and looking at the pro-cyclicality of risk measurements. Having quantified a pro-cyclical effect of sample estimators of quantile-based risk measures and identified some of its causes, the research gives means to fight pro-cyclicality on the basis of risk estimation. Here, a new method is suggested to correct pro-cyclicality observed in traditional risk measurements and use the same methodology as designed in a prior study to test its relevance in limiting pro-cyclicality in various markets.