“Value” investors are the tortoises of the investing public, following the teaching and examples of Ben Graham and Warren Buffett. By purchasing funds that invest in companies whose share prices lag their fundamental “book” value, they expect eventually to outperform the “hares” —i.e., investors in “growth” companies with fast-rising share prices.
Martin Lettau, an economist at UC-Berkeley’s Haas School of Business, was a value fund investor for years. Then he studied the book-to-market ratios of the stocks—not just in his own funds but in a cross-section of active US funds from 1980 to 2016—and learned that he (and other value investors) have usually been rabbits in disguise.
Martin Lettau
“Many active value funds are closet growth funds,” Lettau told RIJ in an interview. “On average, they hold more growth stocks than value stocks. There are few exceptions. DFA’s [Dimensional Fund Advisors] value fund hold mostly value stocks. Only these funds are what I would call a ‘value fund.’ Growth funds, on the other hand, hold what you would expect them to.”
The upshot is that if you’ve been investing in value funds and are waiting for the value premium to show up and vindicate your patience, maybe you should stop holding your breath and look closer at your funds’ holdings.
“Characteristics of Mutual Fund Portfolios: Where are the Value Funds?” co-authored by Lettau, Sydney Ludvigson and Paulo Mandel, all of Berkeley (NBER Working Paper 25381) analyzed 2,638 actively managed funds in the CRSP/Thomson-Reuters Database. Of those, 574 identified as value funds and 1,130 as growth funds. The sample also included 955 exchange-traded funds and 114 hedge funds. The number of active funds, they note, rose eight-fold, from 185 to 1,424, between 1980 and mid-2016.
The authors reviewed several characteristics of the stocks held by the funds, including the book-to-market ratio. They sorted the stocks into five book-to-market quintiles, from low to high. Then they assayed each fund to measure the percentage of its underlying stocks that belonged to each quintile.
“‘Value funds’ are missing from the US equity market,” they concluded. Counter-intuitively, a larger percentage (24%) of the stocks in value funds came from the lowest book-to-market quintile than from the highest (13%). More than half of all value funds held a larger share of low book-to-market stocks than high book-to-market stocks. Only 7% of value funds held more than 25% of their portfolios in high book-to-market stocks.
“While there are over 1,000 mutual funds with consistently low b-to-m ratios, there are virtually no high b-to-m funds in our sample… Even funds with an explicit ‘value’ objective hold a larger share of low b-to-m stocks than high b-to-m stocks in their portfolios,” the paper said.
“One rationale for having value funds is to exploit the value premium. That is, to get a higher return in the long run than growth funds. Our results show that even the funds that call themselves value funds don’t exploit the value premium,” Lettau told RIJ.
It wasn’t a question of whether or not fund managers have abandoned the book-to-market metric in the age of hard-to-value tech stocks. “We looked at a whole list of measures of value and growth,” Lettau said. “We saw the same pattern for any measure.”
“This pattern is true going back to the 80s, so it’s been consistently true. It’s not style drift on the part of fund managers. It’s a consistent pattern. The purpose of our paper was to survey these facts, which are not well-documented.”
Why don’t most value funds hold value stocks? Lettau says that this question is still unanswered and further research is needed. When he presented his findings to an audience at one of the largest asset management firms, the fund managers there didn’t deny his findings. They suggested however that they are only giving the investing public what it wants. “They said, ‘We’re demand-driven.’
“Apparently it’s easier to sell sexy growth funds rather than boring old value funds. But that goes against the value premium, which says that, on average, cheap stocks will perform better in the long run. But, for whatever reason, investors seem to demand growth and glamour.” Just as surprising to Lettau, seemed to be the fact that the value benchmarks that value fund managers use, such as the CRSP value indices, did not necessarily include only what he considered true value stocks.
Dimensional’s value funds are the exception, according to Lettau. In response to an RIJ query, Dimensional’s co-head of research, Marlena Lee, sent a written description of her company’s process when building value funds. “Are we surprised by Lettau et al. finding for our value funds? No,” she wrote. “In our value strategies, one of their primary focuses is purchasing stocks with low prices relative to a fundamental measure of firm value. Price-to-book ratio is the primary way we do this. Among those stocks, we use their profitability, market cap, momentum characteristics, etc. to improve expected returns. Lettau et al. find that our value funds have a strong focus on stocks with low price-to-book ratios because that is how we have designed them.
“Valuation theory suggests that low market prices relative to fundamental measures of firm size combined with high expectations of future cash flows to shareholders should indicate high expected returns… However, the theory doesn’t tell you the best way to measure and pursue the value premium,” Lee wrote.
Lee sent RIJ a table comparing the 15-year net returns (as of 12/31/2018) of five DFA value funds with the average returns of the corresponding Morningstar Categories, and with the DFA fund’s rank among the funds in each of those categories. Performance was higher than the Morningstar average in each category and the rank was in the top 10% or better for four out of five of the DFA funds. The DFA US Large Cap Value fund showed an annualized return of 7.85% over 15 years. The Morningstar average was 6.52% a year.
“Among the variables we use to identify stocks with higher expected returns are price-to-book, profitability, and market capitalization. These variables can be thought of as capturing different angles of value investing, because they help us identify stocks that have low market prices relative to current fundamental measures of firm value and what future cash flows investors expect to receive.”
In her letter, Lee supported Lettau’s idea that the book-to-market ratio has not become less meaningful or obsolete as a value indicator during the tech-stock era. DFA’s analyses “have not shown any downward trend in the informational content of BtM,” she wrote. “In developed and emerging markets outside the US… the explanatory power of price to book in the past 15 to 20 years is similar to its long-term average.”
Lettau has shared his paper informally with a range of reviewers. “The reaction has been varied,” he said. “Some people are surprised by the degree of the tilt. The fund industry already knows there’s more interest in sexy growth funds. What is surprising is that there are virtually no high book-to-market mutual funds. Different market participants knew vaguely about this, but not the extent of the tilt. In academia, this wasn’t known.
“I had a value fund in my own retirement portfolio. I thought I would capture the value premium. But after writing the paper I looked closer at my fund and saw that it didn’t do what I thought it was doing. Even if the fund says ‘value,’ there’s no likelihood that it’s a value fund.”
© 2019 RIJ Publishing LLC. All rights reserved.
