The Pernicious Effects of Contaminated Data in Risk Management

Mardi | 2010-03-16


Banks hold capital to guard against unexpected surge in losses and long freezes infinancial markets. The minimum level of capital is set by banking regulators as a function ofthe banks’ own estimates of their risk exposures. As a result, a great challenge for both banksand regulators is to validate internal risk models. We show that a large fraction of US andinternational banks uses contaminated data when testing their models. In particular, mostbanks validate their market risk model using profit-and-loss (P/L) data that include fees andcommissions and intraday trading revenues. This practice is inconsistent with the definition ofthe employed market risk measure. Using both bank data and simulations, we find that datacontamination has dramatic implications for model validation and can lead to the acceptanceof misspecified risk models. Our estimation reveals that the use of contaminated data reduces(market-risk induced) regulatory capital by around 17%.