Estimativa do risco de concentração individual baseada em modelos de simulação de perdas em operações de crédito

Abstract

Concentration risk is one of the main risks for banks, particularly in emerging economies. A concentrated credit portfolio can amplify losses and lead to crises. In the past, several banking crises have arisen from an increased concentration of risk. Under the Basel Agreement, Pillar 1 capital requirements for credit risk do not include concentration risk. In Pillar 2, regulators demand the banks to estimate concentration buffers on their own. Poor estimations of concentration risk can lead to insufficient capital levels even when the capital ratios are comfortable. The purpose of this article is to verify the influence of the name concentration risk on the economic capital (EC) calculation models, as well as to propose a methodology that allows estimating the portion of EC to cover such risk. The results show that the higher the portfolio concentration, measured by the Herfindahl- Hirschman index, the greater the EC estimation and the greater the additional capital required to cover the name concentration risk. The theoretical model adopted is of Merton-based type, combined with application of Monte Carlo simulation techniques.