Credit risk management for Indian banks

Abstract

Sub-prime crisis is no longer a buzzword for an elite few in high finance. It is pretty much a household name with even common man in India. Though not many understand the details of sub-prime crisis, everyone is now aware that this a much larger monster which has brought dozens of banks down and now is threatening the whole global economy. This is one event in economic and finance history which might sit right up there at the top with the 1929 great depression and will be talked about for decades to come. While this crisis can be blamed on various factors like lack of government insight, complex derivatives, interest rate, regulatory failure etc, the role of credit risk measurement systems cannot be ignored. This clearly shows that in spite of advanced studies and various credit risk models, world is nowhere close to estimating and managing credit risk effectively. Basel-II accord is just another step in managing credit risk better. This academic study starts with an introduction to Basel-II Framework, Credit Risk Management and then goes on to do a literature review of the available models for credit risk management. There are primarily two types of models: Structural and Reduced Form. Both these models are described in detail in this paper. While structural models like Merton s, KMV appear to be well researched and more widely used, reduced form models have their own takers as well. This paper highlights the difference between structural models and reduced form model for the benefit of the reader before going into details of a reduced form model. After a quick literature survey of various reduced form models, this paper attempts to demonstrate use of Jarrow-Lando-Turnbull model for Indian market. This paper demonstrates building a Jarrow-Lando-Turnbull reduced form model using the available market data. Transition matrices from CRISIL and Moody s were used along with credit spread data from FIMDA to construct the model. When applied to the tradable bonds, this model gives as output the price of the bond that appropriately factors credit risk, and hence can be used to manage the treasury portfolio better. This also gives you a probability of default though pricing of a bond is the primary output from this model. Since Indian bond market is still not well developed, this paper could not demonstrate the application of this model on Indian corporate bonds. Jarrow-Lando-Turnbull model has its own limitations, due to which it cannot be used as is for credit risk management in banks. These limitations are highlighted as well in this paper. This paper is not intended to provide to a readymade solution to the banking industry to manage credit risk better, but rather is an attempt to explore the relatively new and less applied reduced form models and test its suitability in Indian context. As Indian bond market is still in early stages this model cannot be applied for lack of tradable bond data. However as Indian bond markets mature over the years, reduced form models like Jarrow-Lando-Turnbull with some adjustments can be used to model credit risk better.