Abstract
One of the main challenges that banks face in quantifying 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 loss data of peer banks. In this method, we scale tail losses using total assets and a measure of risk management quality as scaling factors. Using supervisory operational loss data from large U.S. bank holding companies, we demonstrate that our method of incorporating scaled external data improves the stability of operational risk estimates. In addition, we show that our scaling method can be applied for stress testing operational losses to macroeconomic shocks by better depicting the relationship between losses and macroeconomic variables.
Similar content being viewed by others
Notes
Using publicly available information of 11 large bank holding companies (BHCs), Abdymomunov et al. (2019) report that the share of risk-weighted assets (RWA) for operational risk amounts to, on average, 27% of total RWA, relative to 8% and 65% for market and credit risks, respectively.
See the Wall Street Journal http://www.wsj.com/articles/home-depot-breach-bigger-than-targets-1411073571
In general, a VaR model simulates a loss distribution and risk is measured by a selected tail quantile. Jorion (2007) describes a VaR model in detail and its application in various risk areas. For example, Barber and Copper (1996), Singh (2004), Almeida and Vicente (2009) describe VaR models applied to market and interest rate risks. Embrechts et al. (2005), Embrechts and Puccetti (2006) describe applications of a VaR model to operational risk.
For a detailed description of the FR Y-14Q reporting form and instructions visit http://www.federalreserve.gov/apps/reportforms/.
The FR Y-9C form reports consolidated financial statements for holding companies. Details can be found at https://www.federalreserve.gov/apps/reportforms/
We set the modeling threshold at $100,000 for only two BHCs because of their higher data collection thresholds and their similar business focuses. However, we set our modeling threshold at $20,000 for all other BHCs to use more observations in our data.
SR 04-18 document of the Federal Reserve System describes the BHC rating system and is available at http://www.federalreserve.gov/boarddocs/srletters/2004/sr0418.htm.
We do not include index q in 𝜖i,k because the error term varies for each individual loss i rather than for a given quarter.
In our application of the scaling method, we always find a distribution that passes the goodness of fit tests. We do, however, have a very small number of such cases for individual bank’s data. The application of the bootstrap method to the individual bank’s data does not favor our scaling method in our analysis of the stability of risk estimates relative to the individual banks’ models.
We constrain the parameters of the Generalized Extreme Value and the Generalized Pareto distributions to have a finite first moment in order to avoid unstable and unrealistic estimates. This constraint is not binding in our scaling method. However, it happens to be binding for the individual banks. Thus, this constraint does not favor our scaling method in the stability analysis.
The macroeconomic variables are constructed using data obtained from the St. Louis FRED database.
References
Abdymomunov A, Blei S, Ergashev BA (2015) Integrating stress scenarios into risk quantification models. J Financ Serv Res 47:57–79
Abdymomunov A, Curti F, Mihov A (2019) U.S. Banking sector operational losses and the macroeconomic environment, Journal of Money, Credit, and Banking, forthcoming
Abdymomunov A, Mihov A (2017) Operational losses and risk management: evidence from U.S. Bank holding companies. J Financ Serv Res, pp 1–21. https://link.springer.com/article/10.1007/s10693-017-0284-3
Allen L, Bali T (2007) Cyclicality in catastrophic and operational risk measurements. J Bank Financ 31:1191–1235
Allen L, Bali T, Tang Y (2012) Does systemic risk in the financial sector predict future economic downturns? Rev Financ Stud 25:3000–3036
Almeida C, Vicente J (2009) Are interest rate options important for the assessment of interest rate risk? J Bank Financ 33:1376–1387
Barber J, Copper M (1996) Immunization using principal component analysis. J Portf Manag 23:99–105
Basel Committee on Banking Supervision (2006) International convergence of capital measurement and capital standards, Bank of International Settlements
Board of Governors of the Federal Reserve System (2014) Comprehensive capital analysis and review 2015: summary instructions and guidance, The Federal Reserve System
Board of Governors of the Federal Reserve System (2015) Federal reserve supervisory assessment of capital planning and positions for LISCC firms and large and complex firms, The Basel Coordination Committee of the Federal Reserve System
Board of Governors of the Federal Reserve System (2017) 2017 supervisory scenarios for annual stress tests required under the Dodd-Frank act stress testing rules and the capital plan rule, Board of Governors of the Federal Reserve System
Bollerslev T, Todorov V (2011) Tails, fears, and risk premia. J Financ 66:2165–2211
Chavez-Demoulin V, Embrechts P, Hofert M (2016) An extreme value approach for modeling operational risk losses depending on covariates. J Risk Insur 83:735–776
Chernobai A, Joriony P, Yu F (2011) The determinants of operational risk in U.S. financial institutions. J Financ Quant Anal 46:1683–1725
Chernobai A, Rachev S, Fabozzi F (2014) Composite goodness-of-fit tests for left-truncated loss samples. Handbook of Financial Econometrics and Statistics, pp 575–596
Cope E, Carrivick L (2013) Effects of the financial crisis on banking operational losses. Journal of Operational Risk 8:3–29
Cope E, Labbi A (2008) Operational loss scaling by exposure indicators: evidence from the ORX database. Journal of Operational Risk 3:25–45
Cope E, Piche M, Walter J (2012) Macroenviromental determinants of operational loss severity. J Bank Financ 36:1362–1380
Cope EW, Mignola G, Antonini G, Ugoccioni R (2009) Challenges in measuring operational risk from loss data. Journal of Operational Risk 4:3–27
Cummins JD, Lewis CM, Wei R (2006) The market value impact of operational loss events for US banks and insurers. J Bank Financ 30:2605–2634
Dahen H, Dionne G (2010) Scaling models for the severity and frequency of external operational loss data. J Bank Financ 34:1484–1496
de Fontnouvelle P, DeJesus-Rueff V, Jordan J, Rosengren E (2006) Capital and risk: new evidence on implications of large operational losses. J Money Credit Bank 38:1819–1846
de Jongh R, De Wet T, Raubenheimer H, Venter JH (2015) Combining scenario and historical data in the loss distribution approach: a new procedure that incorporates measures of agreement between scenarios and historical data. Journal of Operational Risk, 10:1–31
Ellul A, Yerramilli V (2013) Stronger risk controls, lower risk: evidence from U.S. Bank holding companies. J Financ 68:1757–1803
Embrechts P, Hoing A, Puccetti G (2005) Worst VaR scenarios. Insurance Math Econom 37:115–134
Embrechts P, Puccetti G (2006) Aggregating risk capital, with an application to operational risk. Geneva Risk Insur Rev 31:71–90
Embrechts P, Samorodnitsky G (2002) Ruin theory revisited: stochastic models for operational risk, Cornell University Operations Research and Industrial Engineering
Ergashev BA (2012) A theoretical framework for incorporating scenarios into operational risk modeling. J Financ Serv Res 41:145–161
Final Rule (2007) Federal register, vol 72, no 235 rules and regulations. Department of Treasury, Federal Reserve System, and Federal Deposit Insurance Corporation
Ganegoda A, Evans J (2012) A scaling model for severity of operational losses using generalized additive models for location scale and shape (GAMLSS). Annals of Actuarial Science 7:61–100
Hess C (2011) The impact of the financial crisis on operational risk in the financial services industry: empirical evidence. Journal of Operational Risk 6:23–35
Hominal P, Deheuvels P (1979) Estimation non paramétrique de la densité compte-tenu d’informations sur le support. Revue de statistique appliquée 27:47–68
Jorion P (2007) Value at risk: the new benchmark for managing financial risk, 3rd. McGraw-Hill, New York
Kelly B, Jiang H (2014) Tail risk and asset prices. Rev Financ Stud 27:2841–2871
Klugman S, Panjer H, Willmot G (1998) Loss models: from data to decisions. Wiley, New York
Knaup M, Wagner W (2012) Forward-looking tail risk exposures at U.S. Bank holding companies. J Financ Serv Res 42:35–54
Lilliefors H (1967) On the Kolmogorov-Smirnov test for normality with mean and variance unknown. J Am Stat Assoc 62:399–402
Na H, van den Berg J, Miranda L, Leipoldt M (2006) An econometric model to scale operational losses. Journal of Operational Risk 1:11–31
Newey WK, West KD (1987) A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica 55:703–708
Schuster E (1985) Incorporating support constraints into nonparametric estimators of densities. Communications in Statistics-Theory and Methods 14:1123–1136
Shih J, Samad-Khan A, Medapa P (2000) Is the size of an operational loss related to firm size? Operational Risk, 2:21–22
Singh M (2004) Value at risk using principal components analysis. J Portf Manag 24:101–112
Wang T, Hsu C (2013) Board composition and operational risk events of financial institutions. J Bank Financ 37:2042–2051
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The views expressed in this paper are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of Richmond or the Federal Reserve System. We thank the Editor and the Referee, as well as Adam Ashcraft, Alexander Cavallo, Jens H. E. Christensen, participants at the 2015 Federal Reserve Stress Testing Research Conference, the 2015 ABA Risk Modeling Forum, the 18th OpRisk North America conference, and the 2017 IBEFA Summer Meeting for their thoughtful comments. All remaining errors are our own.
Rights and permissions
About this article
Cite this article
Abdymomunov, A., Curti, F. Quantifying and Stress Testing Operational Risk with Peer Banks’ Data. J Financ Serv Res 57, 287–313 (2020). https://doi.org/10.1007/s10693-019-00320-w
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10693-019-00320-w