Determinants of the cross-section of expected stock returns in Japan

This dissertation consists of two essays which evaluate whether the cross-section of expected stock returns in Japan is more consistent with the recent risk or non-risk based theories. The first essay investigates whether the Fama and French (1993) size and book-to-market factors are risk proxies. If these factors are true proxies for risk they can be used to price assets across countries in a world where capital markets are at least partially integrated. I find that U.S. and Japanese size and book-to-market effects are not related. Japanese assets with high loadings on the Fama and French factors do not earn higher returns. To evaluate whether the results could be due to lack of integration between the U.S. and Japanese capital markets, the pricing implications are examined in Canada with a similar conclusion. These results are not consistent with the view that size and book-to-market are priced risk factors.;The second essay examines whether the relationship between the ownership of Japanese equities and the cross-section of expected stock returns is consistent with models of investor under- and over-reaction. After controlling for widely-accepted variables affecting stock returns such as size and book-to-market equity, stocks that earn higher returns exhibit large decreases in individual ownership during the year in which the increase occurs. Conversely, stock price increases are accompanied by increases in bank, corporate, foreign, and institutional ownership. After controlling for the change in individual ownership, stocks with a high level of individual ownership also earn lower returns. These findings are consistent with irrational-based theories of under- and over-reaction as well as simple theories which argue that trading by banks, insurance companies, and corporations differentially impact stock prices in Japan. Additional contentions of irrational pricing models are tested and generally reject the notion that under- and over-reaction patterns are entirely due to irrational behavior on the part of individual investors.