Warrington College of Business Administration, University of Florida
Credit risk, deregulation, nonrational behavior and the unreasonable assumption of normality in models have been front and center of the debate concerning the 2007-8 financial crisis. The present issue of The Journal of Risk consists of four papers that each touch on one of these enduring topics.
The first, by Burkhard Raunig and Martin Scheicher, addresses credit default swaps (CDSs), which are considered to be essential to the price discovery process in the corporate debt market. This paper investigates the risk of holding CDSs in a trading book, comparing the value-at-risk (VaR) of CDS positions with the VaR associated with investing in the respective firms’equities. The authors use a sample of CDS–stock price pairs for eighty-six actively traded firms between March 2003 and October 2006 – the period which saw explosive growth in these derivatives. Their empirical findings confirm that, when viewed in isolation, theVaR for a stock is typically far larger than the VaR for a position in the same firm’s CDS. However, the ratio between equity VaR and CDS VaR shrinks considerably for firms with higher credit risk and declines with increasing holding period. Both risk measures are also significantly positively correlated. Panel regressions suggest that these empirical results are consistent with qualitative predictions of the structural modeling approach to default risk.
The growth in derivatives such as CDSs is generally attributed to financial deregulation in advanced economies. While much has been written about purely financial derivatives (those based on interest or exchange rates, say), an increasing amount of attention has come to be focused on the area of deregulated energy markets. As for purely financial derivatives, correct pricing and hedging are essential in such an environment. A characteristic feature of energy prices is their partial predictability due to seasonality and broad weather phenomena such as El Niño. However, daily spot prices for electricity are known to experience significant jumps and to be positively skewed. In this context, the second paper included here, by Hipòlit Torró, addresses the issue of balancing the hedging period of the spot position (one day) with the futures settlement period of the underlying (one month). Using weekly futures from the Nord Pool electricity market to hedge weekly spot price risk, this study shows a reduction of between 60% and 80% in hedging risk.
The derivatives mentioned so far typically involve sophisticated counterparties that attempt to follow rational or formal processes in their decision making. On the other hand, a significant number of contracts with embedded American-style exercise options are held by individuals who are not necessarily financially trained. This is particularly true for variable-annuity contracts, which have been experiencing significant growth in North America, due in part to the demand for retirement savings from baby boomers. The third paper in this issue, by Dmitriy Levchenkov, Thomas F. Coleman and Yuying Li, considers the problem of hedging American options under irrational exercising and shows that such strategies can significantly affect hedging performance and costs.
While derivatives are viewed by many as risk-management tools, the assessment of portfolio VaRs could be considered as another tool. The fourth and last paper in this issue, by José Alfredo Jiménez and Viswanathan Arunachalam, considers the explicit incorporation of skewness and kurtosis into VaR calculations. They propose the use of Tukey’s classical g and h transformations applied to the normal distribution to capture these distributional features. They show that their approach leads to more realistic VaR and conditional VaR (tail expectation) estimates when compared with the normal and Cornish–Fisher expansion techniques.
In closing, I have the pleasure of announcing that Alper Atamturk has joined the editorial board of The Journal of Risk. Alper is Chancellor’s Professor in the Department of Industrial Engineering and Operations Research at the University of California, Berkeley, and is currently leading the risk optimization research group at Bloomberg. Welcome Alper!
Using Tukey's g and h family of distributions to calculate value-at-risk and conditional value-at-risk