Quantifying and Stress Testing Operational Risk with Peer Banks' Data

52 Pages Posted: 25 Jun 2015 Last revised: 19 Apr 2019

See all articles by Azamat Abdymomunov

Azamat Abdymomunov

Federal Reserve Banks - Federal Reserve Bank of Richmond

Filippo Curti

Federal Reserve Banks - Federal Reserve Bank of Richmond

Date Written: January 2018

Abstract

One of the main challenges that banks face in modeling operational risk is the instability of risk estimates caused by heavy-tailed and insufficient loss data. To address these issues, we propose a loss scaling method to combine a bank’s internal loss data with external loss data of other banks. Using supervisory operational loss data from large U.S. bank holding companies, we find that the severity of tail losses is related to bank size, while smaller losses are not. Based on this finding we propose scaling tail losses using total assets as a scaling factor. We demonstrate that our method of incorporating scaled external data improves the robustness of operational risk estimates. In addition, our scaling method helps depict the relationship between tail operational losses and macroeconomic variables. We demonstrate an application of the method to stress testing operational risk to severe macroeconomic shocks.

Keywords: Operational risk, Banking Capital, Stress testing, Loss scaling

JEL Classification: C22, C23, G21

Suggested Citation

Abdymomunov, Azamat and Curti, Filippo, Quantifying and Stress Testing Operational Risk with Peer Banks' Data (January 2018). Available at SSRN: https://ssrn.com/abstract=2622175 or http://dx.doi.org/10.2139/ssrn.2622175

Azamat Abdymomunov (Contact Author)

Federal Reserve Banks - Federal Reserve Bank of Richmond ( email )

P.O. Box 27622
Richmond, VA 23261
United States

Filippo Curti

Federal Reserve Banks - Federal Reserve Bank of Richmond ( email )

P.O. Box 27622
Richmond, VA 23261
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
205
Abstract Views
689
rank
149,749
PlumX Metrics