The procedures used in this analysis attempt to conjure up the spirit of performance future. In so doing, scant attention is paid to what did happen to manager returns. The analysis begins by using returns-based style analysis to identify the combination of passive indexes that replicate the manager's returns. The R2 in the table describes the degree of success. The R2 for the T. Rowe Price Growth Stock fund indicates that 97% of the variance in returns for this fund are explained by nine style indexes developed by Ron Surz at PPCA Inc., San Clemente, Calif. I use these indexes because they are the only ones that don't put some of the same stocks in both value and growth categories. I call this the manager's style benchmark.
The second step focuses on what could have happened rather than what did happen. To be precise, 200 years of annual returns that could have happened are generated based on 15 years of monthly returns that did happen with this manager's style benchmark. So, I am not looking at what the manager actually did but what the manager's style benchmark could have done. Furthermore, instead of looking at the average return (mean), as every other performance measure does, I calculate the upside potential relative to the downside risk for the style blend. This is called the upside-potential ratio, which for the T. Rowe Price Growth Stock fund is 0.94. This indicates a style with slightly more downside risk than upside potential (a score of 1 would mean they are exactly equal).
Nevertheless, the omega excess calculation indicates T. Rowe Price Growth was able to beat the style benchmark by 260 basis points after adjusting for downside risk. This is quite impressive given the high R2. Omega excess uses the difference between the actual return of the manager and the style blend for the past three years. Downside risk, however, is calculated from what could have happened and, in my opinion, captures the inherent risk.
As far as I know, the methodology used in this analysis is the only attempt to estimate future performance on a risk-adjusted basis. The results reported in Pensions & Investments for the past three years show a decided improvement over traditional performance measures in this regard. Of course, just because it seemed to do better at estimating future performance for the past three years does not mean it will continue to do so.
The funds in the table are ranked by the upside-potential ratio. No large growth funds have more upside potential relative to downside risk at this time, and most of them can't beat their style benchmark on a risk-adjusted basis.
Many large value funds indicate more upside potential than downside risk, but only a few have a positive omega excess. The small-cap funds seem particularly vulnerable at this time.
Frank Sortino is director of the Pension Research Institute, Menlo Park, Calif.