Modelling Operational Risk

The Basel Committee on Banking Supervision ("the Committee") released a consultative document that included a regulatory capital charge for operational risk. The complexity of the object "operational risk" led from the time of the document's release to vigorous and recurring discussions. We show that for a production unit of a bank with well-defined workflow processes where a comprehensive self-assessment based on six risk factors has been carried out, operational risk can be unambiguously defined and modelled. Using techniques from extreme value theory, we calculate risk measures for independent and dependent risk factors, re-spectively. The results of this modelling exercise are relevant for the implementation of a risk management framework: Frequency dependence among the risk factors only slightly changes the independency results, severity dependence on the contrary changes the independency results significantly, the risk factor "fraud" dominates all other factors and finally, only 10 percent of all processes have a 98 percent contribution to the resulting VaR. Since the definition and maintenance of processes is very costly, this last results is of major practical relevance. Performing a sensitivity analysis, it turns out that the key 10% of relevant processes is rather robust under this stress testing.