How Profitable Is Capital Structure Arbitrage?

The article examines the risk and return of capital structure arbitrage, which exploits the mispricing between a company’s credit default swap (CDS) spread and equity price. The analysis uses the CreditGrades benchmark model, a convergence-type trading strategy, and 135,759 daily CDS spreads on 261 North American obligors. At the level of individual trades, substantial losses can occur as a result of the low correlation between the CDS spread and the equity price. An equally weighted portfolio of all trades, however, produced Sharpe ratios similar to those for other fixed-income arbitrage strategies and hedge fund industry benchmarks.Debt–equity trading, or capital structure arbitrage, has been increasingly embraced by hedge funds and bank proprietary trading desks in recent years. Some traders have even touted it as the “next big thing” or “the hottest strategy” in the arbitrage community. Fueling this enthusiasm, the popular press has given several vivid accounts of how traders were able to use simple debt–equity trades to produce fabulous returns.To understand capital structure arbitrage, I conducted a comprehensive study of the risk and return of the strategy by using 135,759 daily credit default swap (CDS) spreads on 261 North American obligors from 2001 through 2004. Following the common practice, I used the first 10 daily observations of a company’s CDS spread to calibrate an industry standard debt–equity model, the CreditGrades model. When the subsequent model spread deviated from the observed market spread, the hypothetical trader entered into a CDS position with a model-determined equity position taken as a hedge. Both positions were closed out when the model spread and the market spread converged or at the end of a prespecified holding period, whichever occurred first. I computed the daily returns to each trade by valuing the outstanding CDS position as a survival-contingent annuity, with the survival probabilities inferred from the CreditGrades model.The strategy entailed a high degree of risk at the level of individual trades. Not only did the trades rarely converge, they frequently suffered from large drawdowns of initial capital that, in practice, would almost surely have triggered withdrawal by hedge fund investors. Compared with other fixed-income arbitrage strategies, such as swap spread arbitrage, capital structure arbitrage requires a much higher initial capital to attain the same profile of profits and losses. Ultimately, these findings can be attributed to the relatively low correlation between daily changes in the CDS spread and the equity price, which averaged between –5 percent and –20 percent among the sample obligors I studied.To examine the risk and return from the perspective of a diversified investor, I aggregated the returns from the individual trades into an equally weighted monthly capital structure arbitrage index return. For the speculative-grade portfolio, the monthly index returns had a Sharpe ratio of 0.75, similar to the Sharpe ratios of hedge fund industry benchmarks. These returns were not correlated with common equity and bond market risk factors. Moreover, evidence suggests that the returns are consistent with “statistical arbitrage,” a notion of long-horizon profitability that is independent of any asset-pricing model.