Fama and French Dish, Diss, and Disagree

In this Journal of Indexes (JOI) interview (download below), two of the world’s smartest investors weigh in on the strategies that have made Dimensional Fund Advisors one of the most successful fund firms in the world.  Better known as finance professors, Eugene Fama and Ken French helped design DFA’s stock-selection and fund-management techniques.  In this interview, Fama and French revisit their original conclusions, poke fun at the arrivistes who have since claimed to have discovered their ideas, and, once again, show why traditional stock-picking is almost always for the birds.

Notable points:

• Fama and French now disagree about the cause of the "value effect"–the phenomenon in which value stocks have outperformed growth stocks over the long term (for more on the value effect, see this post).  French no longer believes that this outperformance represents "distress risk" and, instead, thinks there is some "mispricing" involved (investors underestimating the prospects for value stocks and overestimating them for growth stocks).  Fama, ever the efficient market man, begs to differ.

• Both men are shocked that commentators believe that today’s hot product–"fundamental indexing"–is actually something new.  The academic literature, they point out (and DFA), has described the strategy for 15 years.  The advantage of "fundamental indexing" is merely that it captures the value effect.  As French puts it: "It’s not even another way [of doing this.]  It’s the same way.  I have a friend who calls [fundamental indexing] value investing for clients who don’t understand ratios."

• On the "dividend-weighting" strategy used by Jeremy Siegel and WisdomTree: This, too, is nothing new.  Dividends are just another way of sorting stocks to capture the value premium.  Unfortunately, using dividends as the value screening metric is inferior to using book-to-market ratios.  How do Fama and French know this?  Because they’ve tested all the value metrics.

• Fama still thinks the market is efficient, but French now concedes that some people can probably beat it.  Returns are so random, however, that he just doesn’t have enough information to figure out who they are (implication: neither do the rest of us.)

• On whether the value/growth mean-reversions can be timed (Specifically, as I asked in this post:  Can investors use the ratio between value and growth to predict the performance of each asset class?)  According to French, there’s no evidence that this works.

• On ETFs: Great tools, as long as they’re cheap and passive.

• On behavioral finance: Investors make the mistake of assuming that, now that they know some of the mistakes investors make, they won’t make those mistakes.  Unfortunately, they’ll probably just make others.

Download fama_french_journal_of_indexes_3_07.pdf