Earlier research into the relation between stock returns and the effects of size and earnings/price ratios has not provided clear-cut results. New findings, based on the longer, 1951–86 sample period, indicate that both E/P and size effects are significant when estimated across all months. January differs from the other months of the year, however; both the E/P and size effects are significant in January, but only the E/P effect is significant during the rest of the year.
Estimated regressions of portfolio excess returns on a predetermined variable proxying for risk premiums indicate that the variable is able to predict returns on size and E/P portfolios, especially in January. Furthermore, the strength of the predictability increases as firm size decreases and E/P increases. The size and E/P effects, and the interaction between them, take on different shapes, depending on the level of the predetermined variable.
