CTA index performance

Let’s focus first on the Barclay index. Much of its gain (to $2,534) came in its first four years. The large positive outliers of 1987 and 1990 have not been approached since.

The Sharpe Ratio of this index underscores the appalling mediocrity of currency CTAs over time. The average monthly return has been 0.686 percent, with a standard deviation of 3.542 percent. Subtracting the threemonth T-bill average monthly return of 0.398 percent over this period and annualizing the results produces a Sharpe Ratio of 0.068. T-bills by definition have a Sharpe Ratio of 0.00. This isn’t much of a risk-adjusted gain over T-bills, is it?

The returns on the asset-weighted Stark currency CTA index are even more stuck in the Reagan administration. The combination of fiscal stimulus and tight money during those long-ago days of supplyside economics and a Federal Reserve fighting inflation aggressively created two massive dollar trends. The first, which began in 1981 and went through the first quarter of 1985, was a one-way affair: All any trader had to do was sell anything against the greenback. The 1982 annual return was 53.9 percent, 1983 was -6.2 percent, 1984 was 53.6 percent, and 1985 was 49.3 percent. The second trend, selling the dollar, was referenced above in the context of the Barclay index.

The longer history of the Stark index over some spectacular trading years is offset by a higher volatility involved for those returns. If we use the entire sample from 1981 forward, the average monthly returns are 0.964 percent with a standard deviation of 5.78 percent and a Sharpe ratio of 0.063. If we truncate the sample to match the Barclay index’s 1986 start, the average monthly return falls to 0.513 percent with a standard deviation of 3.982 percent and a Sharpe ratio of 0.023. If currency CTAs do not find this last number embarrassing, they should.