On the explanatory power of firm-specific variables in cross-sections of expected returns

This paper pertains to the controversy surrounding the explanatory power of certain firm-specific variables such as size and the book-to-market ratio in cross-sections of average stock returns. To investigate whether these firm-specific variables capture the sensitivity of returns to unobserved systematic risk, two sets of principal component factors are used. The first set is constructed from individual stock returns and the second set is from size- and book-to-market-sorted portfolio returns. The evidence from the first set of factors shows that size and the book-to-market ratio have little to do with factor betas. The evidence from the second set of factors shows that the forces underlying size and the book-to-market ratio are indeed systematic risks, although they explain very little return variation at the firm level, and that the betas of size- and book-to-market-sorted portfolio returns with respect to the corresponding systematic factors do explain the size and book-to-market effects.