The Sharpe Ratio is a useful tool for calculating return against risk, but it's not perfect. The Sharpe Ratio is calculated by looking at historical returns, which can't tell you with certainty what will happen in the future.
 |
|
|||||||||
|
||||||||||
|
||||||||||
By
One thing that hasn't changed is the Sharpe Ratio. First presented by economist William Sharpe back in 1966, it remains one of the most popular tools used to calculate return against risk. It is relatively simple to compute, and easy to understand (high Sharpe ratios are good, low Sharpe ratios are bad).
However, the Sharpe Ratio is not perfect. For the rest of this article and over the next few weeks, we will point out how the Sharpe Ratio can come up short. None of this means to belittle the tremendous contributions of Mr. Sharpe. But, to use a tool effectively, you need to understand its limitations. And, unlike the interminable Florida vote coverage, I promise not to use the phrase "hanging chad" even once. (Doesn't "hanging chad" sound like a fraternity prank that they'd do at Electoral College?)
"Past performance is no guarantee of future results"
You've see this disclaimer for mutual funds, and we've said this before for our screens. At best, past returns might be an indicator of future performance, but they are certainly not a guarantee.
The Sharpe Ratio looks at both the average historical return and the historical volatility of those returns. Unfortunately, a small change in either of those factors can cause us to choose a strategy that underperforms in the future.
Here's an example. Below are the results we've used in earlier articles, showing the historical returns for a five-stock RS-26 strategy (annual hold, rebalanced every January), the S&P 500 Index, and one-year Treasury bills. The first table shows 1969-1985:
Year RS-26 S&P 500 U.S. T-bill
1969 -14.52% -11.36% 6.07%
1970 12.02% 3.94% 7.51%
1971 76.87% 14.30% 4.40%
1972 20.02% 18.99% 3.82%
1973 19.70% -14.69% 5.00%
1974 -14.79% -26.47% 5.58%
1975 18.58% 37.23% 7.01%
1976 34.40% 23.93% 5.44%
1977 15.14% - 7.16% 5.00%
1978 31.04% 6.57% 6.80%
1979 10.37% 18.61% 9.54%
1980 53.53% 32.50% 10.96%
1981 -22.01% - 4.92% 12.62%
1982 18.23% 21.55% 12.77%
1983 54.27% 22.56% 8.01%
1984 -25.65% 6.27% 9.07%
1985 60.37% 31.73% 8.33%
Avg. 20.44% 10.21% 7.53%
St. Dev. 29.50% 18.29% 2.74%
Sharpe 0.4290 0.1513
From 1969 through 1985, the RS-26 strategy listed here had a much higher Sharpe ratio than the S&P 500.
What does this mean? Remember the example of Nervous Ned? Ned did not want to invest in any strategy that was more volatile than the S&P 500 Index. Back in early 1986, Ned could have looked at this historical data and concluded that a blend of about 62.5% in an RS-26 strategy and 37.5% in T-bills would have the same volatility as the index, but with higher returns. Unfortunately for Ned, from 1986 through 1999 that blend would have been much more volatile than the index.
Year RS-26 S&P 500 U.S. T-bill
1986 53.01% 18.82% 7.21%
1987 13.11% 5.40% 5.46%
1988 -10.97% 15.99% 6.52%
1989 46.59% 31.56% 8.37%
1990 7.34% -2.97% 7.38%
1991 126.70% 30.51% 6.25%
1992 21.13% 7.45% 3.95%
1993 9.05% 10.09% 3.35%
1994 42.46% 1.33% 3.39%
1995 23.20% 37.28% 6.59%
1996 22.34% 22.69% 4.82%
1997 17.96% 33.60% 5.30%
1998 22.91% 30.73% 4.98%
1999 116.65% 21.10% 4.31%
Avg. 36.53% 18.83% 5.56%
St. Dev. 39.73% 12.99% 1.55%
Sharpe 0.7792 1.0528
As you can see, the S&P 500 Index had, by far, the higher Sharpe Ratio from 1986 through 1999. What a rude surprise for Ned.
Remember Crazy Cathy? She looked at that second table and concluded that investing in the S&P 500 on margin was a good idea. But, if the market now starts performing more like it did from 1969 through 1985, then her plan will be extremely risky. She'll be in big trouble with a market like we had in '73-'74, where the S&P 500 dropped 15% the first year, then 27% the second!
Historical returns should always be taken with a grain of salt (and that especially includes the historical returns of our strategies). And the Sharpe Ratio is based on historical returns, so it needs a bit of seasoning as well. Here at the Workshop, we expect that our strategies will continue to outperform the market, but we have no way of knowing what the returns will be in the future. Leveraging a strategy (i.e., running the strategy on margin) because it has a higher Sharpe Ratio might give you a higher theoretical return on paper, but we would never recommend that someone borrow large sums of money in an attempt to improve returns. Real life just doesn't follow the plan every time.
I don't want to knock the Sharpe Ratio too much. All of our strategies here at The Motley Fool are based on the belief that past results will be at least somewhat useful in predicting the future. Just don't take out a second mortgage on the house based on those historical returns.
Next week we'll talk a bit more about "risk." You might have noticed that we've been saying "risk" when we really mean "volatility." We'll cover that in detail next week, in a way that I hope you all find very interesting. I look forward to seeing you then -- same Fool time, same Fool channel!
RSS Headlines
Fool UK
