When statistics fail : extreme events in financial institutions

![CDATA[Extreme value theory (EVT) is regularly put forward by academics, practitioners and banking regulators as a methodology for measuring the likelihood of operational risk losses that have a very low probability of occurrence, but which have the potential for catastrophic outcomes in terms of financial losses. Given the potential for extreme events to threaten the financial viability of a banking institution, these groups argue that it makes sense to allocate capital against the likelihood of extreme events, and EVT forms the basis for such a capital allocation methodology. This paper challenges this proposition, pointing to recent large losses in banking institutions that either maimed or destroyed the institutions in question. In all of these cases organizational risk culture was at the centre of losses, and more specifically, the incentives inherent in remuneration schemes. It is argued that EVT is inadequate when it comes to identifying adverse cultural or incentive issues in banking institutions.]]