In this issue of The Journal of Credit Risk we present three research papers and one technical report.
The first paper in the issue is "Sovereign credit ratings and the new European Union member states" by Nick Wilson, Pavol Ochotnicky and Marek Kacer. In this study the authors focus on the determinants of sovereign credit rating for the new entrants to the European Union since 2004, plus Turkey and Croatia in the time period between their first rating and the end of 2012. The results of the study confirm that per capita gross domestic product and gross debt are highly significant and are the most important determinants of sovereign credit rating throughout all models, agencies and specifications. The authors model three time series of sovereign ratings: one for each of the three major credit rating agencies. They find that the political stances (or changes therein) and geographic locations of countries' trading partners are important factors in the determination of ratings during more turbulent periods. The authors use three model specifications for each agency to estimate the parameters: the linear regression model with individual random effects, the ordered probit model and the ordered probit model with random effects.
Our second paper is "Analysis of credit portfolio risk using hierarchical multifactor models" by Pak-Wing Fok, Xiuling Yan and Guangming Yao. The paper generalizes Vasicek's asymptotic single-risk factor solution to a special case of a factor model where all companies share the same global factor while companies within a sector share the same sector factor. The authors derive closed-form solutions for the value-at-risk under the following assumptions: that the number of companies in each sector (n) is large, that the number of sectors in the portfolio (N) is large, and that the exposures scale as the inverse of nN.
The issue's third paper is "Redesigning ratings: assessing the discriminatory power of credit scores under censoring" by Holger Kraft, Gerald Kroisandt and Marlene Müller. The authors derive upper and lower bounds for the accuracy ratio and the Kolmogorov-Smirnov statistic for the full sample when only censored data is available. Such an application is potentially useful if a censored credit data set based on a bank's existing selection criterion derived from the credit score Sold is used to develop a new selection criterion Snew.
A technical report describes a particular practical technique and enumerates situations in which it works well and others in which it does not. Such reports provide extremely useful information to practitioners in terms of saved time and minimizing duplication of effort. The contents of technical reports complement rigorous conceptual and model developments presented in the research papers and provide a lot of value to practitioners.
Our technical report in this issue is "A parametric approach to counterparty and credit risk" by Giuseppe Orlando and Maximilian Härtel. In this paper the authors present the results of a business solution on how to measure credit and counterparty risk, with the main focus being over-the-counter derivatives. The authors' approach strikes a middle ground between the overly simplistic regulatory approach and the more formalistic and rigorous - but complex and impractical - mathematical approaches. The report describes a more useful, and effective, approach for managing business risk. In fact, the report reads almost like a "recipe" for the implementation of a counterparty credit risk process that could be accomplished by most trained financial engineers.
JPMorgan Chase, New York