Journal of Risk

Stan Uryasev

University of Florida

Is there light at the end of the current financial crisis? The recent performance of major financial players and the evolution of the gross domestic products of the two largest economies of continental Europe seems to point that way. However, other indicators, mostly related to employment and credit markets, temper somewhat this optimistic outlook. It is clear that the regulatory landscape will change and will affect risk perception and management. However, irrespective of risk sources, perception or regulations affecting its assessment, our community will get opportunities aplenty to address related issues. This issue contains articles that address in particular interest rate and market risks in investments, plus issues of adverse selection in an insurance context.

Though interest rates have reached historical lows in the past few months, their variability has a direct impact particularly on fixed income vehicles such as liability-driven investments. Previous research has shown that capturing better the dynamics of the yield curve does not necessarily lead to better hedging. In his article, N. Carcano focuses on hedging based on Principal Component Analysis (PCA) to address the impact of model error. His empirical comparison between two- and threecomponent models leads to the conclusion that the former results in a substantially smaller hedging error, as suggested by theory. His study also shows that controlling exposure to model errors leads to a substantial reduction in transaction fees.

Another prevalent form of market risk is through the variability of equity prices. In their paper, Doran and Fodor address the issue of including options in portfolios holding long positions in the underlying asset. In particular, they are interested in the relation between the idiosyncratic risk of an option and that of its underlying asset. They find significant negative abnormal returns associated with purchasing put options, in contrast to the effect of writing such contracts. An investor can also augment portfolio returns by taking relatively large covered call positions. However, their study suggests that there is little evidence of increased portfolio value resulting from consistently purchasing any option. Additionally, they find that the premiums from writing put options are not related to any specific firm characteristic, suggesting a pervasive risk premium. Asset pricing tests that include market option factors are unable to explain the returns to portfolios which include firm specific options. Tests on delta-hedged portfolios confirm gains to put options are related to idiosyncratic volatility risk and not market risk volatility risk. This is indicative of an idiosyncratic risk premium for options that is distinct from idiosyncratic risks in stocks. Barrier options are particularly useful for targeted market risk strategies. In their paper, Engelmann et al evaluate empirically dynamic hedging strategies for such options under different stickiness assumptions on the dynamics of the implied volatility surface. Specifically they compare sticky-strike, sticky-moneyness and local volatility-implied (model-consistent) hedges for barrier options with a maturity of one and two years. They find that sticky-strike performs best. Their study also indicates that the hedging strategy is much more important than the stickiness assumption.

The recent crisis has also highlighted issues inherent to risk insurance, namely moral hazard and adverse selection. An insurer is thus aware of the higher proportion of customers seeking coverage who are likely to file for their claims. In the fourth paper, Shahidi follows the competitive equilibrium route to determine the coverage strategies of high- and low-risk insurance seekers, accounting for adverse selection. In a non-expected utility set-up and under first-order stochastic dominance, the results indicate that high-risk candidates seek full coverage, in contrast to their lowrisk counterparts who optimize a non-convex problem.