Theoretically, one can combine U.S. equity indexes with foreign and fixed-income indexes when performing style analysis.
And theoretically it doesn't matter if the manager actually holds X% in U.S. large-cap stocks as long as the portfolio behaves as if he did.
However, we found the PRI allocation model (Pensions & Investments, May 3) was filling the U.S. large growth allocation with an international fund manager. Why? It turns out European stocks are highly positively correlated with U.S. large growth. The generally accepted practice would result in an overallocation to European stocks and an underweighting in U.S. large growth stocks. This would cause a conflict between investment policy and actual allocations.
To correct this type of error we decided to change the PRI style analyzer to make three passes: one for U.S. equities; one for foreign equities; and one for fixed income. This, plus a few other changes — such as eliminating funds with high minimums or funds that were closed to new accounts — resulted in a substantial difference between the funds we rank as best of breed in this issue as opposed to the May 3 rankings.
The PRI preferred list in Table 1 was created by eliminating all funds with less than $300 million, performing a style analysis and eliminating the bottom half by upside-potential ratio. Then the top three funds in each category were chosen on the basis of their ability to beat their style blend on a risk-adjusted basis.
This is the kind of procedure fiduciaries should use to select a list of funds for their defined contribution plans. They should at least attempt to find the best funds instead of focusing on the lowest cost.
Notice none of these funds are "pure" anything They are all a blend of styles. That means you can not choose any one fund to fill an allocation to, say, 25% large-cap value, without overweighting something else. Yet, that is what most, if not all, in the industry are doing.
