Journal of Risk Model Validation

Daniel Rösch

University of Hannover

Harald Scheule

University of Melbourne

In response to the current capital shortage in the industry, The Journal of Risk Model Validation is presenting you with a special issue on stress testing. The financial crisis has led to a general criticism of Basel II. We hope to show that various methods exist within current or possibly future regulatory environments to cope with present and future financial stresses. Thus, the issue covers an interesting mix of theoretical, conceptual and applied approaches on stress testing. The fundamental question that we would like to address and hope to answer is how economic downturns and other stresses may be incorporated into modern risk measurement frameworks before rather than after the occurrence of a financial crisis.

Banks have to formally stress-test their assessment of capital adequacy in line with Basel II. These stress measures must be compared against the measure of expected exposure and considered by the bank as part of its internal capital adequacy assessment process. In recent years, banks have developed internal models, which are currently under scrutiny by the respective regulators for approval. This special issue is organized to coincide with the launch of a new Risk Book entitled Stress Testing for Financial Institutions - Applications, Regulations and Techniques, written by leaders in the stress-testing discipline. The book has a similar objective and addresses a broad range of areas, from general statistical frameworks to retail and corporate credit portfolio risks to the provision of economic and regulatory capital. As such, the book provides a valuable guidance for regulators and practitioners.

The present special issue contains a strong collection of five papers. With the first paper "Downturn LGDs for Hong Kong mortgage loan portfolios", we address one major problem of economic downturns - the correlation between default and loss given default (LGD) rates. This correlation leads to the requirement to base a bank's capital on the "Downturn" loss rate given default which is also known as Downturn LGD. This article evaluates alternative concepts for the modeling of Downturn LGDs and presents a practical application of Hong Kong mortgage loan portfolios.

Burd shows in the second paper "Breaking correlation breakdowns: nonparametric estimation of downturn correlations and their application in credit risk models" howdownturn correlations between asset value returns can be incorporated into a Basel II type asymptotic single risk factor model. He presents an approach for bootstrapping time-varying correlations from empirical data and analyzes the impact of model and calibration error. hHöse and Huschens show in their paper "Worst-case asset, default and survival time correlations" how credit risk model parameters can be stressed by assuming the most adverse scenario and provide a practical implementation of their concept. Hamerle and Plank analyze the issue of "Stress-testing CDOs". Collateralized debt obligations are more sensitive to the economy and hence to stress-testing than single-name products. They apply a dynamic factor model and analyze the difference between portfolios of bonds and portfolios of structured finance tranches. Finally, Sibbertsen et al present a framework for "Measuring model risk". The contribution critically reviews the calibration of financial risk models by providing a thorough overview on the topic and discussion of potential directions for future research.

We hope that you will enjoy reading this special issue and would like to encourage you to share any thoughts and experiences with our quickly evolving community.