Library
Avoid the institutional heavyweights
By Bob Tattersall
Earlier this year, I wrote an article explaining the difficulties facing investors who attempt to capture the value premium in the stock market (Capturing the wealth premium: Feb. 22, 2008). Although there is extensive evidence that small-capitalization stocks and value stocks (defined as low price to book value) outperform large-cap and growth stocks at the individual security level, most of this positive differential disappears at the index or portfolio level.
A study by two U.S. professors quoted in this article found that there was no benefit to value investing in the large-cap sector. There was a benefit to value investing in the small-cap sector, but in their opinion, investors failed to capture even this benefit because of higher transaction costs and higher management fees associated with small-cap mutual funds.
My conclusion was that: “if you hope to capture the value premium in the market, then you must be willing to own a large percentage of small-cap names which are off the radar screen of the institutional investors.” By chance, the current issue of the Financial Analysts Journal (March/April 2008) has an article by Ludovic Phalippou, which picks up on this theme of institutional ownership and value investing.
The author’s opening argument is that for the value premium to persist, there must be consistent pricing errors in a certain group of stocks and limited or expensive opportunities to arbitrage away these errors. Making a heroic leap of faith, Mr. Phalippou assumes that institutional investors are less likely to make these pricing errors and so he ranks his U.S. stock database first by degree of institutional ownership. There was a wide variation in this measure, with the lowest 10 per cent of the universe reporting an average institutional ownership of only 1 per cent, while the top 10 per cent reported ownership of 68 per cent. His definition of the value premium is the difference in average return between the cheapest 25 per cent of stocks as measured by price-to-book value and the most expensive 25 per cent measured on the same basis.
The table which plots increasing levels of institutional ownership against the value premium is fascinating and the regularity of the relationship is striking. According to the author, the value premium decreases steadily from an extremely high 185 basis points a month for the lowest institutional ownership decile to a negligible 13 basis points a month for the highest ownership decile.
Since institutions tend to favour large-cap growth stocks, there is presumably a high correlation between small size, low price-to-book and low institutional ownership, which makes it difficult normally to sort out the dominant influence on the value premium. Mr. Phalippou, however, is able to sort his database three ways so as to hold each of these elements constant and comes to some surprising conclusions. He notes that “small-cap stocks with high institutional ownership did not exhibit a significant value premium, even if they were very small.” In addition, low price-to-book value stocks failed to deliver a value premium, regardless of their market cap, if they also had high institutional ownership. In other words, these are not unrecognized bargains: They are well-researched stocks trading at an appropriate price.
Although the article observes that “size, therefore, has no marginal explanatory power over institutional ownership in explaining the value premium”, there is no question that most of the low ownership stocks are also small-cap. This particular study covers U.S. stocks for the period 1980 to 2001. The breakpoint where institutional ownership begins to fall off rapidly and the value premium begins to emerge was about $200-million market capitalization. As a practical matter, for most individual investors it may be easier to screen first for low market capitalization and then exclude those that already have high institutional ownership. You will miss out on a few large-cap low ownership value stocks, but the numbers involved will likely be small.
The insights provided by this article are helpful, but not likely to radically change your approach if you are already a value investor. There is a high level of overlap between small-cap stocks, low price-to-book stocks and low institutional ownership stocks, so it is not surprising that screening on any of these variables will generate similar results. The article does make it clear, however, that there is probably little reason to devote much research effort to a stock which already has high institutional ownership, no matter how small the market cap or how low the price-to-book ratio.