The WSJ reports that the average stock fund has 172 holdings. What is the point of having that many holdings? Diversification? The table below reveals that there is very little incremental reduction in annual standard deviation once you get past about 20 holdings.
(Click to Enlarge)
Source: M. Statman, “How Many Stocks Make a Diversified Portfolio?” Journal of Financial and Quantitative Analysis 22 (September 1987), pp. 353-64.
The real reason mutual funds own so many stocks was revealed in an academic study conducted by researchers from Yale. The real goal of most mutual fund managers is to reduce tracking error (volatility of portfolio return around a benchmark index.) Many fund managers have realized the challenges associated with deviating from the benchmark and have chosen to increase the number of holdings so that they will never be too much worse than the benchmark. Of course, they will never be too much better either. With the impact of fees, such a closet-indexing approach is very unlikely to add any value over time. However, that doesn’t keep the manager from telling a great story and attracting investors based on their perceived investment prowess. The active-share study completed by K. J. Martijn Cremers and Antii Petajisto examined the proportion of stock holdings in a mutual fund’s composition that was different from the composition found in its benchmark. The greater the difference between the asset composition of the fund and its benchmark, the greater the active share. According to active-share study, there was a positive correlation between a fund’s active-share value and the fund’s performance against its benchmark. For example, a mutual fund with an active-share percentage of 75% indicates that 75% of its assets differed from its index, while the remaining 25% mirrored the index.
The study found that funds with a higher active-share value would tend to be more consistent in generating high returns against their benchmark indexes, which implies that more actively managed funds have more skilled managers. However, higher active share necessarily means higher tracking error. Since the 1980s there has been a steady rise in closet-indexing.
Investors need to understand the real reason that most mutual funds have so many holdings. After all, an active manager can only add value relative to the index by deviating from it. If an investor’s goal is to beat the benchmark over time, buying a mutual fund with over 100 holdings (a closet indexer) is not likely the way to go. To beat the benchmark over time an investor needs to invest in strategies that have fewer holdings and, of course, a winning investment strategy. On the other hand, if the investor’s goal is to match the benchmark over time then it is more cost effective to buy an index fund from Vanguard for 9 basis points.
We make no secret about the fact that our relative strength strategies have high active-share (most above 90%). Most of our strategies have 20-25 holdings. While others serve the closet-indexing market, we have chosen to serve the active-investors market.
