Editor: Marcelo Cruz
Published: 30 Sep 2013
September 2013 marks the fifth anniversary of the collapse of Lehman Brothers - a collapse that triggered the greatest financial crisis of our generation. Much has been said about real estate bubbles, credit issues and liquidity crunches, but did operational risk events play any role in this crisis? Likewise, did the financial crisis have any impact on the field of operational risk? In the past few years a number of papers have been published in The Journal of Operational Risk that have tried to analyze these relationships...
Papers in this issue
by Patrick McConnell
by Eric W. Cope and Luke Carrivick
by Lukáš Štěpánek, Roman Urban and Rudolf Urban
by Dominique Guégan and Bertrand K. Hassani
Welcome to the third issue of the eighth volume of The Journal of Operational Risk.
Regarding the role that operational risk played in the crisis, it is widely known that the greed of mortgage brokers, coupled with aggressive business objectives, pushed them to grant mortgages to clients with low credit ratings and a lack of documentation: mortgage documents were therefore falsified so that credit was not rejected. These stories were extensively reported by the press and were even made into Hollywood films. As these mortgages were packaged into mortgage-backed securities, sold to investors worldwide and kept on banks' balance sheets, when the price of these securities collapsed the financial industry crashed. These events can clearly be assigned to operational risk.
As one would expect, given the stress and disruption that the financial crisis has brought into the lives of a great many people around the world, a number of papers have been published looking at the impact of the crisis on operational risk. For example, in our Spring 2011 issue we presented a study from Christian Hess titled "The impact of the financial crisis on operational risk in the financial services industry: empirical evidence". The author used data from an external loss data vendor to assess the impact of the financial crisis on operational risk and broke down the data by business line, focusing particularly on two business lines: trading and sales, and retail brokerage. The paper computed a 157% increase in the operational value-at-risk for the trading and sales business line and a 52% increase for the retail brokerage business line. These increases were due to the addition of large losses associated with the financial crisis and because of corresponding large losses from some (internationally operating) investment banks that marketed and distributed these securities. In this issue we present another paper that tackles the subject. The financial crisis has lasted five years and, despite the fact that it appears that economic indicators are starting to show positive signs, the financial industry will never be same as it was before the crisis.
I would like to ask potential authors to continue to submit to The Journal of Operational Risk regarding the state of operational risk research, and I would again like to emphasize that the journal is not solely for academic authors. Please note that we do publish papers that do not have a quantitative focus, with one example of this in the current issue. We at The Journal of Operational Risk would be happy to receive more submissions containing practical, current views of relevant matters as well as papers focusing on the technical aspects of operational risk.
There are two very interesting technical papers in this issue. One reports a study into applying techniques for coping with a lack of data in a fast-growing market, while the other takes a novel nonparametric approach to assessing risk.
In the first paper, "Effects of the financial crisis on banking operational losses", Eric W. Cope and Luke Carrivick discuss a very interesting and timely subject. They argue that although the financial crisis that started in 2007-8 was largely triggered by credit and market risk events, it also had a substantial effect on operational risk. The authors identify these effects using approximately ten years of data from several dozen international banks reporting losses to a leading operational loss data consortium. They have found that the impact of the crisis was concentrated in a few lines of business, loss categories and types of banks, in terms of both loss frequency and severity. However, the effects usually appear to be temporary anomalies in an otherwise steady decline across all loss categories in the numbers of losses per unit of income, including extreme losses. In addition, the industry-wide nature of the effects of the crisis are less substantial than has been reported elsewhere in the literature.
In our second paper, "Using a time series approach to correct serial correlation in operational risk capital calculation", Dominique Guégan and Bertrand K. Hassani propose a solution to the issues that arise when autocorrelations are detected between losses. Their approach suggests working with the losses organized as a time series. A number of models are tested in these time series, like autoregression, autoregressive fractionally integrated and Gegenbauer processes, and then a distribution is fitted to residuals. Finally, a Monte Carlo simulation enables construction of the loss distribution function, and the pertaining risk measures are evaluated. In order to show the impact of internal models retained by financial institutions on the capital charges, the paper draws a parallel between the static traditional approach and an appropriate dynamical modeling. If, by implementing the traditional loss distribution approach, no particular distribution proves that it fits the data - as soon as the goodness-of-fit tests reject them - keeping the loss distribution approach corresponds to an arbitrary choice. This paper suggests an alternative and robust approach. For instance, for the two data sets explored in this paper, with the introduced time series strategies, the independence assumption is relaxed and the autocorrelations embedded within the losses are captured. The construction of the related loss distribution function enablescomputation of the capital charge and therefore allows compliance with the regulations, simultaneously taking into account the large losses with adequate distributions on the residuals, and the correlations between the losses with the time series processes.
In this section of the journal we publish papers that report day-to-day experiences in operational risk management. In this issue we have two papers in this section. The first provides an almost forensic dissection of the manipulation of the LIBOR rate that went on during the financial crisis of 2008, while the second shows a risk-ranking methodology for various scenarios and gives an application of the methodology using insurance company data.
In the first forum paper, "Systemic operational risk: the LIBOR manipulation scandal", Patrick McConnell argues that the manipulation of LIBOR rates was not a localized event. Unscrupulous traders and managers in some of the largest banks around the world deliberately and systematically manipulated borrowing rates. McConnell argues that this deplorable event was not the work of isolated "rogue traders" but part of business-as-usual in the international money markets. The paper describes the LIBOR scandal and argues that it is an example of systemic operational risk - in particular, people risk. The paper first describes the LIBOR setting process. It then goes on to describe the explosive growth in the use of interest rate swaps over the past twenty-five years, and the process of resetting rates on interest rate swaps, which is what ultimately led to the unethical manipulation of the underlying LIBOR rates. The paper then looks at official inquiries into manipulation of LIBOR at three banks - Barclays, UBS and Royal Bank of Scotland - to identify examples of operational risk. Transcripts of conversations unearthed by these investigations detail the rampant illicit activities that were apparently a normal part of doing business, as traders, LIBOR submitters and brokers colluded to manipulate LIBOR for their own interests. Finally, the paper makes some suggestions as to how the management of systemic operational risks may be addressed by banks and regulators.
In the second paper in the forum section, "A new operational risk assessment technique: the CASTL method", Lukáš Štěpánek, Roman Urban and Rudolf Urban provide an overview of a brand new operational risk quantification method that they have named CASTL. This method offers a practical classification and ranking of risk scenarios. The authors offer an application of the method using information from an insurance company.
Search the archive
Subscribe to gain full access to The Journal of Operational Risk and its archive.
Call for papers
Submit your work and we can offer you:
Please contact [email protected] for more information.