On the Aggregation of Market and Credit Risks

This paper presents a new approach to aggregating market and credit risks in large complex financial firms, banks in particular. By identifying risk factors that are common to many business activities, dependencies between different risk types across various lines of business can be properly accounted for in the aggregate risk estimate. The risk factor aggregation model is illustrated using historical data on market and credit risk factors that are common to many business units, including interest rates, credit spreads, equity indices and implied volatilities. Economic capital estimates obtained using the model are compared with the economic capital data from several major banks. Applications to optimal risk diversification shows that, whilst the independent control of economic capital by business unit can be sub-optimal, the risk factor aggregation approach has the great advantage of allowing both risks and returns in different business activities to be modeled in the same framework. We show that this greatly facilitates the constrained optimization of a risk/return objective.