Fama and French on small/value premiums and risk
What is the nature of the small (SmB) and value (HmL) premiums proposed by Eugene Fama and Kenneth French? Does this mean that markets aren’t efficient, or is there a Small and a Value risk (“distress risk”) factored in to these premiums that leaves markets efficient? In a 2007 interview with Journal of Indexes, Fama and French discuss this question. And disagree.
Journal of Indexes (JoI): The market has been through some wild rides since you wrote your seminal paper in 1992. Has anything altered the views you have vis-à-vis the sources of return in the market?
Kenneth French (French): I don’t think so…
Eugene Fama (Fama): No, I don’t think so.
French: Actually, I take that back. Initially, we thought the value premium was associated with distress risk. I’m not so confident of that any longer.
What’s clear is that the value effect is a catch-all for differences in expected return. Without pernicious assumptions about expected growth, any differences in expected return will show up in ratios like book-to-market, earnings-to-price, or cash flow-to-price. Take a company with a high expected return. When you discount its expected future cash flows back to the present, the high expected return (which is also the discount rate) gives you a low price relative to the expected cash flows. As long as the fundamental metric—the book value, earnings, etc.—is a reasonable proxy for the expected cash flows, you’ll also get a low price relative to the current fundamental. This simple discount rate effect implies that differences in expected return are almost certainly linked to ratios like book-to-market.
Notice that I didn’t say why there are differences in expected return; I just said that if there are differences in expected returns, they will show up in ratios like book-tomarket. That means the value effect is a catch-all that captures any differences in expected return—whatever their source.
I think the differences in expected return are the result of an amalgamation of different risks, plus some mis-pricing. And I don’t think we have the technology to distinguish between those two.
Fama: I don’t know about the mispricing. I don’t know that there is mispricing.
French: There has to be some mispricing. I’m certainly not saying it’s all mispricing. I like to tell people that it’s 87.38 percent risk.
Fama: I don’t know about the mispricing part. I think he’s wrong there.
French: To get back to distressed companies … the typical high book-to-market company is distressed. But if you hold book-to-market fixed and you sort companies by financial distress, you don’t get much difference in average returns. So something else is at work.
The rest of the article is worth a read, as they discuss fundamentally-weighted indexes, and more.
