Too Small or Too Low? New Evidence on the Four-Factor Model

In 1992, Fama and French published a landmark paper in which they provided, by means of a cross-sectional analysis, strong evidence of explanatory power by size and book-to-market factors, compared with little or no ability by the market factor to explain differences in equity returns. After this, a large body of literature came out evidencing the beta model’s weakness in explaining asset returns. Empirical works have mostly used US data, and most of them reject the beta Capital Asset Pricing model (CAPM — see, for example, Grinold, 1993). In another paper, Fama and French (1993), using a time-series approach, found basically the same evidence. However, further evidence for their model (Lakonishok et al., 1994; Haugen, 1995), highlighted the role of investor overreaction (De Bondt and Thaler, 1985) in explaining the value anomaly. Based on the overreaction/under-reaction argument to information, Jegadeesh andTitman (1993) and Rouwenhorst (1998) document the existence of a momentum anomaly: over a medium time horizon, firms with high returns over the previous three months to one year continue to outperform firms with low past returns over the same period.