20 Sep

Aldo Nassigh, “Default and Credit Migration Risk in Trading Portfolios”

Tuesday September 20 2011
16:00
Scuola Normale Superiore
Aula Bianchi

Monica Billio
Unicredit, Milano

Abstract
Default and credit migration risk was treated as negligible for a long time in banks’ trading portfolios, characterized by an investment horizon of few days. This was consistent with the ‘Constant Level of Risk’ assumption according to which, in case of deterioration of the creditworthiness of the obligor, exposures with high credit quality would have been replaced with the goal of moving the asset allocation back to the original risk profile. If perfect market liquidity and continuous Brownian motion for asset prices are granted, losses induced by the frequent rebalancing of the portfolio can indeed be neglected. The rise and blow up of the Credit Trading bubble (also named Sub-Prime, Lehman and Sovereign crises) showed the shortcomings of such approach. In 2004, the Basel Committee on Banking Supervision asked banks to set aside capital for credit risk in trading portfolios, in response to the rising credit exposures and the improvements in risk management best practice observed in the banking system. Such capital add-on (named ‘Incremental Risk Charge’) will enter into force in December 2011. The proper evaluation of default and credit migration risk under the constant level of risk assumption translates into the call for modeling portfolio credit risk in the framework of short-term, multi-step simulations. Aim of the seminar is to give an update on recent developments regarding modeling the Incremental Risk Charge and to raise some critical and unresolved issues as: the difficulty in adapting to this problem the mainstream treatment of portfolio credit risk by continuous-time Markov Chains applied to the rating migration process; the lack of an unambiguous approach to the estimation of asset correlations, leading to large discrepancies in the capital level required by the various models developed so far.

Download Flyer Download Slides