Two scholars at Ozyegin University, in Istanbul, make the case that hedge funds have a uniquely good record in one specialized sort of alpha harvesting: They are good at finding overpriced growth stock securities and trading to their advantage on the basis of that finding, especially when other sorts of investors, including non-HF institutional investors, are moving aggressively in the wrong direction, into those overpriced growth stocks.
The scholars, Mystafa Onur Caglayan and Umut Caliker, respectively an associate professor and an assistant professor of finance at the faculty of business at Ozyegin, begin from the established literature on the book-to-market effect. There is a substantial return difference between high book-to-market and low book-to-market stocks (colloquially known as value stocks and growth stocks, respectively). Value stocks do better than growth stocks.
Scholarship since Fama-French
Indeed, this observation dates back more than 20 years, to a landmark Fama-French paper in 1992. As Andrew Beer summarized the Fama-French understanding of the "growth factor" in an entry in AllAboutAlpha last year, "investors chase the latest highfliers, while boring 'value' stocks are overlooked and underpriced." Further, Fama and French said, the growth factor is more important when the market is bearish than when it is bullish, which of course makes it especially attractive to risk-shy portfolio managers.
Beer, CEO of Beachhead also suggested that the Fama-French view might be out of date, that one ought not rely on the growth factor unreflexively, because the underlying anomaly may have been priced away over the course of the two decades since its discovery.
Caglayan and Caliker don't believe that it has been priced away. They cite, for example, an article that appeared in the Journal of Financial Economics earlier this year, by R.M. Edelen et al., "Institutional investors and stock market anomalies," that seems to show not only that the value-over-growth anomaly survives, but that it is surviving not merely (as one might expect) on the strength of naïve individual investors. It is surviving because institutions, too, tend to over-value growth stocks.
Why does the anomaly survive? In Edelen's view, it is because institutional investing generally entails agency issues such as excessive turnover, and managerial incentives that skew toward risk taking.
Caglayan and Caliker begin where Edelen leaves off. This agree with Edelen, and with a 2010 study by Hao Jiang of Michigan State University, that institutions have done worse than simply fail to resolve this inefficiency: institutions have perpetuated it. Jiang's explanation for this is the alleged over-reliance of institutions on "intangible information," that is, on narrative qualitative information that is "orthogonal" to what one finds on a balance sheet. Institutions look to intangibles because they believe that is the best ingredient for secret sauce. In fact though, that focus favors investment in growth firms, and keeps such growth firms over-valued.
Caglayan and Caliker ask the next question: are all institutional investors alike in this? Or are there important differences, especially between hedge funds and other sorts of institutions?
They find that hedge funds "have a strong ability [to detect] overpriced growth securities and trade them to their advantage when non-hedge funds more aggressively in the opposite direction. That is, those over-priced growth stocks that are heavily bought by non-hedge funds and contemporaneously sold by hedge funds experience significant losses in the following year, possibly contributing to superior performance of hedge funds."
A One-Way Street
Separately, the same authors report, there is evidence that hedge funds are also good at finding overpriced securities within the universe of value stocks, and sell those too just as the less nimble institutions are buying. The evidence for this is weaker than is the statistical evidence that they are doing so in the growth-stock space, but it is nonetheless significant. In a footnote, the authors write that "[for] value stocks heavily purchased by non-hedge funds and sold by hedge funds simultaneously, we observe a monthly three-factor alpha of -0.42% with a t-stat of -2.17 and a four-factor alpha of -0.33% with a t-stat of -1.63."
All of this is consistent with a hypothesis, long common in the literature on modern portfolio theory, that the persistence of over-pricing is always more likely than the existence of underpricing, simply because short selling can be trickier than buying. Arbitraging away inefficiencies, in other words, is not a two-way street.