The papers in this issue of The Journal of Risk reflect the breadth of risk management topics.
This issue contains a technical paper on credit risk, a paper on the estimation of economic capital, a paper on the vulnerability of the public sector in an emerging country, and finally a paper on portfolio market risk measures.
The first paper by Haaf, Reiß and Schoenmakers, “Numerically Stable Computation of CreditRisk+”, discusses numerical computation issues for CreditRisk+, a major portfolio credit risk system. CreditRisk+ requires a convolution to construct the probability density function of credit losses and originally proposed a so-called “Panjer” recursion method for the convolution. This method, however, is numerically unstable due to the accumulation of roundoff errors. The present paper presents an alternative recursion method similar to that proposed by Giese in Risk Magazine (2003), and proves numerical stability of the new method.
The second paper by Kupiec, “Estimating Economic Capital Allocation for Market and Credit Risk”, argues that a VAR measure computed from the initial value to the quantile on the target date does not necessarily translate into the correct initial amount of equity capital to maintain the desired confidence level. In particular, the author shows that omitting the cost of debt creates a bias in the amount of initial economic capital. The paper provides corrections for the economic capital that account for these effects. The issue is not too important for market risk given the short horizon. This is not the case, however, for credit risk given typical one-year or longer horizons.
The next paper, “Assessing Fiscal Vulnerability Under Uncertainty”, by Barnhill and Kopits deals with an unusual but nevertheless important application of value-at-risk, namely the assessment of vulnerability of the public sector’s position. The general consensus is that the Asian crisis was made worse by the fragility and lack of transparency of financial systems. Dornbusch (1998), when discussing new directions for the international financial system, wrote that “The appropriate conceptual framework is value at risk – a model-driven estimate of the maximum risk for a particular balance sheet situation over a specified horizon. There are genuine issues of modeling, but there is no issue whatsoever in recognizing that this approach is the right one.” Here, the authors present an application to Ecuador. The paper analyzes various sources of risk for Ecuador’s public sector and reports that the volatilities of sovereign spreads and oil prices are major risk factors for that country. This is the first step toward better management of risk.
Finally, in “An Empirical Investigation of the Rank Correlation between Different Risk Measures”, Pfingsten, Wagner and Wolferink perform an empirical comparison of various risk measures for two actual bank trading portfolios. The authors had access to the daily distribution of profits and losses for a “government bond options book” and an “interest rate derivatives book”. These data allow comparisons of the standard deviation, valueat- risk (VAR), expected tail loss (ETL), and lower partial moments for realistic portfolios with skewed distributions. Comparing the ranking of risk measures across trading days, the authors find that correlations across risk measures are very high, suggesting little practical difference across measures.
The mission of The Journal of Risk is to further our understanding of risk management. Contributions to the journal are welcome from academics, practitioners and regulators in the field. With this in mind, authors are encouraged to submit full-length papers.