Figures never lie, but they are subject to interpretation and more often, misinterpretation.
The PRI Preferred List shows the top funds in each of nine style categories: large cap, midcap, small cap, value, core and growth. We began with approximately 20,000 mutual funds and filtered them down to these 22 funds. (The procedure for doing this was explained in Pensions & Investments, Nov. 1.)
Here's how they're open to misinterpretation.
Suppose a 401(k) participant had been shown this list at the end of the third quarter of 2004. The participant probably would have said, "They were projecting all these high returns in the teens, but look, they almost all earned negative returns in the last quarter. I'm not going to buy these losers. I'll keep my money in T-bills, where it's safe." Well, all the funds on the list were up substantially in the fourth quarter. Furthermore, those safe T-bills guarantee the participant will never earn enough money on her investment to retire. How's that for safety? What's worse, the participant probably won't realize it until a year or two after retirement. The participant would have been much better off paying a professional 150 basis points than forgoing last quarter's results. There is a great deal of research to support the notion that 401(k) participants should not manage their own portfolio. This also applies to the proposal for individual Social Security accounts.
Does this demonstrate the predictive power of the PRI methodology? No. The whole market was up in the fourth quarter. How did these funds do relative to the market? The Standard & Poor's 500 index was up 9.2% for the quarter and 10.9% for the year. Does the fact that 81% of the funds on PRI's preferred list outperformed the S&P 500 for the year provide validation for our methodology? No. It is not reasonable to compare a small-cap growth fund like the Merrill Lynch Small Cap Growth fund with a basically large-cap index like the S&P 500. The real market has more than 6,000 stocks.
There are a few who say, "One should compare each fund with a peer group (e.g., large value, small growth etc.)." Barclays Global Investors now has an iShare index for each of the nine style categories in the PRI Preferred table. This would be an improvement over comparisons to the S&P 500. The problem is, the Merrill Lynch Small Cap Growth Fund is 70% small-cap core and only 16% small-cap growth. Only iShares currently has a small-cap core index you can buy. But, even if you use the iShare index, you still will get a misleading answer because it is missing 30% of the source of returns for this fund.
So what is an investor to do? First, recognize that all managers are a blend of passive indexes; purity is non-existent. Then use William F. Sharpe's return-based style analysis procedure to identify each manager's style blend. Create a comparative benchmark for each fund's style blend by using the BGI iShares. And then see if the fund beat its style blend on a risk-adjusted basis. We call this the Omega excess return.
Frank Sortino is director of the Pension Research Institute, San Francisco.
