ABSTRACTEarlier evidence concerning the relation between stock returns and the effects of size and earnings to price ratio (E/P) is not clear‐cut. This paper re‐examines these two effects with (a) a substantially longer sample period, 1951–1986, (b) data that are reasonably free of survivor biases, (c) both portfolio and seemingly unrelated regression tests, and (d) an emphasis on the important differences between January and other months. Over the entire period, the earnings yield effect is significant in both January and the other eleven months. Conversely, the size effect is significantly negative only in January. We also find evidence of consistently high returns for firms of all sizes with negative ear