This paper attempts to shed light on the asset pricing questions reaised by recent empirical research. Fama and French, among others, find that variables that are supposed to explain cross-sectional returns, specifically risk parameters that emerge from asset pricing models, have little explanatory power. On the other hand, firm characteristics such as size and book-to-market ratios, that do not fall out of any asset pricing model, are quite successful in explaining the cross-sectional distribution of historical returns. This has led to the suspicion that Fama-French type results are due to data problems, such as selection bias or data mining. In this paper, a new, direct estimate of expected returns is constructed using the discounted cash flow model and the IBES data. Consistent with previous empirical work, these new estimates of expected returns are found largely to be uncorrelated theoretical risk parameters, but to be significantly correlated with firm size and book-to-market ratios. However, the correlation between DCF expected returns and book-to-market ratios is found to be negative, as opposed to the positive correlation between book-to-market ratios and historical returns.