Journal of Credit Risk

First International Conference on Credit Analysis and Risk Management

Austin Murphy from Oakland University

On July 21-23, 2011 the First International Conference on Credit Analysis and Risk Managementwas held at Oakland University. The conference, which included experts from both academia and industry, provided a forum for expanding our knowledge of all aspects of investment in debts, but there was an especially strong focus on the individual credit.

As was stated in the introductory comments at the conference, credit analysis originally relied solely on the subjective judgment of the lender. By the turn of the millennium, however, the mathematical and statistical evaluation of credit had expanded to the point where some felt that subjective judgment was no longer required. That view is still held by some, although others believe that a qualitative analysis of credit is essential to avoid another credit crisis like the one that occurred in 2008.

Many of the conference sessions were devoted to estimating the probability of default, the payout in default or both. Most of the papers presented relied on statistical or mathematical models, but there was extensive discussion of incorporating qualitative factors into the evaluation of credit. Some of the excellent research that was presented is published in this issue of The Journal of Credit Risk.

The first paper in the issue, "Approximating default probabilities with soft information" by Dror Parnes, provides a quantitative model and simulated empirical results of how qualitative information may improve credit decisions, especially for borrowers with higher risk of default. This model is based on an assumption that the qualitative judgments are biased toward easier credit, as is typically the case with lending officers being compensated based on the volume of loan originations. All the qualitative information that is obtained through interactions with borrowing applicants might be more effectively used if lending agents had incentives to make their subjective credit ratings less biased in favor of granting loans. Integrating their subjective judgments with statistical models of credit risk might then further enhance accuracy in estimating default risk across the entire credit spectrum. This subject represents a very important area for future theoretical and empirical work.

The second paper, "Pricing corporate loans under the risk-neutral measure" by Terry Benzschawel, Julio DaGraca and Cheng-Yen (Mike) Lee, provides a very useful analysis of loans and revolving credit lines that accounts for both prepayment and default risk. The model is calibrated to firms' bond prices and five-year credit default swap (CDS) spreads as well as adjusting for funding costs and liquidity premiums. By incorporating market prices into the valuation of illiquid debt instruments held by lending institutions, the model imputes fair value for the loans and credit lines they hold. In addition, this theory may enable improved estimates of the comovement and/or portfolio risk of such assets/obligations. These estimates could lead to vital newresearch into the relationship between the systematic risk of investments in debts, credit risk spreads and liquidity premiums.

The third paper, "The impact of counterparty risk on credit default swap pricing dynamics" by Stefan Morkoetter, Johanna Pleus and Simone Westerfeld, shows theoretically and empirically how the market pricing of CDSs is affected by the riskiness of the party providing protection against default. In particular, when a guarantor of any debt through a CDS has a greater chance of defaulting on the obligation to provide protection, the resulting increase in counterparty risk causes buyers of the CDS protection to pay a lower amount for that protection. The impact is lower when the insured debts have a higher credit quality as there is a reduced likelihood of a CDS protection seller actually being called upon to make insurance payments in such cases.

The fourth paper, "New risk analysis tools with accounting changes: adjusted Z-score" by Seong Cho, Liang Fu andYinYu, develops a modification of the Altman Z-score that adjusts for distortions caused by changes in accounting methods that are widespread and that can distort the relevant variables incorporated into that statistical model. Empirical tests indicate that the adjustments lead to significant changes in the Z-score, and the modified Z-score is also empirically shown to result in improved accuracy in bankruptcy prediction.

The fifth paper, "Credit loss and systematic loss given default" by Jon Frye and Michael Jacobs Jr., develops and tests a model for loss given default that incorporates the negative correlation between default frequency and recovery values and that uses only parameters that are already in hand. Testing their model against more complicated approaches, they find that no other approach improves on their estimates for total portfolio credit loss. Their theoretical framework could potentially enhance appraisals of the systematic risk of debts, thereby possibly leading to improved estimates of required credit spreads that should include yield premium compensation for undiversifiable portfolio risk.

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: