The Role of Credit Default Swaps and Other Alternative Betas in Hedge Fund Factor Analysis (Digest Summary)

Factor analysis helps investors analyze the systematic risks in hedge funds as well as highlight true alpha produced by the fund. Traditional factor analysis uses broad categories of factors to analyze risk. The author expands on this analysis by updating the research through June 2009, thus including the 2007–08 financial crisis. He also incorporates such alternative betas as credit default swaps, illiquidity, and market-timing factors. He finds that these additional factors help explain the risk as well as the true alpha produced by broad categories of hedge funds.

Many analysts conduct factor analysis on hedge funds to determine what broad attributes contribute to a hedge fund’s performance. By conducting this analysis, they can determine how much beta the hedge fund produces compared with the beta produced by simple exposure to a certain investment style. Previous studies have examined broad variables, such as large and small stocks, high-yield bonds, volatility, and Treasury yields. These studies may not have fully captured the true contribution of various attributes or styles because of the flexibility of hedge funds and their evolving styles over time. By not capturing the correct style of the fund, an analyst may attribute more alpha to the manager than is truly applicable.

The author expands the number of factors to consider, including such alternatives as commodity, currency, and volatility indices; illiquidity; market-timing measures; and the increasingly popular credit default swaps. He examines the data from January 1990 through June 2009 as well as from August 2001 through June 2009; these periods include the market crisis of 2007 and 2008, which may have altered many of the funds’ relationships with traditional and alternative factors.

To begin, the author tests the traditional factor model by running regressions against the S&P 500 Index, the EAFE index, emerging market stocks, the Barclays Capital Aggregate Bond Index, and a high-yield index. The model does a good job of determining hedge fund styles by producing an R² of 0.499 and an average alpha of 59.1 bps per month using data from January 1990 through June 2009. Only those factors that are significant are included in each regression. The author shows that risk factors between hedge funds vary widely among styles. He also shows that there is an increase in explanatory power and a decrease in alpha using the August 2001–June 2009 time period. He adds that since 2001, only 8 of the 19 hedge fund styles tested earned a statistically significant positive alpha. As a leverage factor is introduced to the regression, fixed-income hedge funds show an increase in explanatory power and a further decrease in alpha.

The author then adds alternatives to the model. He uses the S&P Goldman Sachs Commodity Index to represent commodity markets, the U.S. Dollar Index to represent currency markets, the difference between the S&P 500 and Russell 2000 to represent returns to small-capitalization securities, the difference between the Russell 1000 value and growth indices to represent the return to equity styles, the VIX to measure the volatility as a proxy for investor sentiment, and the difference between the 10-year and 1-year Treasury yields to measure the slope of the yield curve. These additional factors help to explain more of the hedge fund returns than the original model does.

Next, he adds credit default swaps (CDS) and finds that this factor helps explain 12 of the 19 hedge fund styles tested. Hedge fund returns tend to decline as CDS spreads widen. In fact, CDS spreads are more explanatory of hedge fund risks than rising equity market volatility as measured by the VIX.

When testing for an illiquidity factor with distressed and event-driven styles, the author finds that both had a high correlation with illiquidity risk. The styles that show high liquidity risk also show high autocorrelation. To capture the true risk of these styles, the monthly beta needs to be added together to include this autocorrelation risk faced by investors.

Finally, the author uses a squared-terms test to search for market-timing skill. A positive coefficient on squared factor returns shows superior market-timing skill, whereas a negative coefficient shows poor timing skills. He finds that at least six hedge fund styles have negative market-timing skills in high-yield bonds and timing is even less common within equity markets. Only macro and systematic, or trend-following, futures styles show positive market-timing skill.

Factor analysis helps investors analyze the systematic risks in hedge funds as well as highlight the true alpha produced by the fund. Traditional factor analysis uses broad categories of factors to analyze risk. The author expands on this analysis by incorporating alternative factors. He finds that these additional factors help explain the risk as well as true alpha produced by broad categories of hedge funds.