Is the value premium disappearing? The answer to that question could shake the foundations of the asset-management industry.
First, for the uninitiated, a little background. In the late 1970s and '80s, a lot of investors and researchers confirmed what market analysts had claimed for a long time – that certain stocks seemed cheap relative to their value on paper. These stocks tended to perform better, on average, than the stock market as a whole. That was a bit of a puzzle, since basic finance theory says that it shouldn't be that easy to beat the market. If you can just buy stocks that are cheap relative to their book value, and wait for them to go up, the market isn't very efficient, is it?
To beat the market that easily and consistently, the strategy should incur some kind of systematic risk – risk that you can't get rid of with diversification. In the early 1990s, future Nobel-winning finance researcher Eugene Fama and his longtime co-author Kenneth French came up with a partial answer to the puzzle. Some stocks, they said, were so-called value stocks that tended to trade at prices below their book values. During certain periods, these stocks all tend to rise in value, but other times they tend to all fall together. Over a long period of time, the rises outweigh the falls, so that these value stocks earn a premium. But if you invest in value stocks and you get caught in one of the bad periods, you're in trouble. Hence the value premium persists, because trading against it – by simply loading up on all the value stocks you can grab – incurs some risk.
That answer wasn't completely satisfying. Why do value stocks all tend to rise in certain periods and fall in others? One possibility, according to asset-management mogul Cliff Asness, is that value stocks are "crappy companies." Shaky companies might find themselves suddenly squeezed for credit under certain conditions, while other companies are still able to borrow. That might explain the value premium. But it turns out that the times when value underperforms and outperforms are not clearly tied to the business cycle. The mystery remains.
The model of Professors Fama and French of a risk-based value premium has become standard throughout the asset-management industry. But there's another, more disquieting possibility. It may be that the value premium is caused not by risk, but by systematic inefficiencies in the financial market.
There are many reasons other than risk why investors might choose to avoid cheap-looking stocks. One reason is herd behaviour – if investors think that other investors are avoiding a certain stock, they might assume that stock is bad, and similarly avoid it. Some stocks might simply be boring and unglamorous. Some might be obscure companies in overlooked industries, or information about them might be more difficult and costly to obtain.
These factors all might have been more prevalent in the 1960s and 1970s, when data were sparse and expensive, trading costs were much higher and the number of professional investors was more limited. As financial markets improved, we may have seen the entrance of more investors who are willing to do the hard work of digging up obscure and boring companies, and who are willing to go against the herd. If the value premium really was a systematic underpricing rather than a true risk premium, then the gradual development of financial markets would be expected to shrink this premium over time.
There are signs that this is happening. Although value stocks did well in the early 2000s, they have dramatically underperformed since the crisis, even though the market has boomed. Of course, that might simply be a particularly long period of underperformance – we might expect to see value bounce back soon enough. But in fact, the decline has been going on for quite a bit longer than that – the value premium has been falling since the mid-1990s. Coincidentally, that is exactly when the Internet and computerized trading systems made it possible to invest in stocks much more cheaply, and to gather information much more easily.
That would mean that markets are getting more efficient – at least, in this one particular way. But it would also mean that market efficiency takes a very long time to establish itself. If big, systematic mispricings such as the value premium can survive for decades before they are finally traded away, it means that other flaws in the market might be equally long-lived. For example, the momentum factor – another mainstay of standard finance theory – might also be a market flaw that will eventually be shown the exits.
If the market is that inefficient, it also means that stock prices are, in some deep sense, "wrong" – that they are not the best available estimate of a company's value. That would suggest we should be relying on markets less than we do for such things as executive compensation. So watch to see whether the value premium comes back. If it doesn't, it means it might never have been about risk in the first place.