- Fama-French US Small Value Index (ex-utilities) -- 13.6 percent
- Fama-French US Small Cap Index -- 11.8 percent
- Fama-French US Large Value Index (ex-utilities) -- 10.3 percent
- Fama-French US Large Cap Index -- 9.5 percent
- Fama-French US Large Growth Index (ex-utilities) -- 9.0 percent
- Fama-French US Small Growth Index (ex-utilities) -- 8.7 percent
Similar findings have been found for penny stocks, stocks in bankruptcy and IPOs -- all have experienced lower returns than we would expect. Behavioralists explain these outcomes as the result of the "lottery effect," or investors searching for the next Google (GOOG) or Apple (AAPL).
Dimensional Fund Advisors (DFA) designs its funds differently than index funds. It attempts to maximize the benefits of indexing, such as low costs, broad diversification, low turnover and relatively high tax efficiency. It also tries to eliminate or minimize the negatives of indexing, such as forced turnover, realization of short-term capital gains and inclusion of all securities within an index. For example, DFA has long used screens to eliminate IPOs, stocks in bankruptcy and penny stocks from its eligible securities lists. A recent internal study sought to find a way to minimize or eliminate the "black hole" problem.
DFA's researchers found that it could improve returns by adding the following screens. First, it screens for the 25 percent of the small-cap stocks with the lowest book-to-market (BtM), and then screens for the top 25 percent by price-to-earnings and/or price-to-cash flow ratios. Using simulated data covering the period 1979-2009, DFA found that if it screened out about 10 percent of the roughly 2,500 stocks from its list of eligible securities, it would have improved returns by about 1.4 percent.
Amazingly, while its current small-cap universe produced an annualized return of 11.9 percent (versus 11.3 percent for the Russell 2000), the extreme small-cap growth stocks actually lost 1.6 percent per year. Note that the weighted average BtM of their small stock universe was 0.57, while the excluded stocks had a weighted average BtM of just 0.16.
DFA's action is another example of it not blindly follow a belief in market efficiency, but instead basing strategies on academic research as well as common sense.