Recent Advances in Credit Risk Management

In the last decade, the market for credit related products as well as techniques for credit risk management have undergone several changes. Financial crises and a high number of defaults during the late 1990s have stimulated not only public interest in credit risk management, but also their awareness of its importance in today's investment environment. Also the market for credit derivatives has exhibited impressive growth rates. Active trading of credit derivatives only started in the mid 1990s, but since then has become one of the most dynamic financial markets. The dynamic expansion of the market requires new techniques and advances in credit derivative and especially dependence modelling among drivers for credit risk. Finally, the upcoming new capital accord (Basel II) encourages banks to base their capital requirement for credit risk on internal or external rating systems [4]. This regulatory body under the Bank of International Settlements (BIS) becoming effective in 2007 aims to strengthen risk management systems of international financial institutions. As a result, the majority of international operating banks sets focus on an internal-rating based approach to determine capital requirements for their loan or bond portfolios. Another consequence is that due to new regulatory requirements there is an increasing demand by holders of securitisable assets to sell or to transfer risks of their assets. Recent research suggests that while a variety of advances have been made, there are still several fallacies both in banks' internal credit risk management systems and industry wide used solutions. As [15] point out, the use of the normal distribution for modelling the returns of assets or risk factors is not adequate since they generally exhibit heavy tails, excess kurtosis and skewness. All these features cannot be captured by the normal distribution. Also the notion of correlation as the only measure of dependence between risk factors or asset returns has recently been examined in empirical studies, for example [7]. Using the wrong dependence structure may lead to severe underestimation of the risk for a credit portfolio. The concept of copulas [13] allowing for more diversity in the dependence structure between defaults as well as the drivers of credit risk could be a cure to these deficiencies.