Only one-fourth of U.S. families own stock. This paper examines whether the consumption of stockholders differs from the consumption of nonstockholders and, if so, whether these differences help explain the empirical failures of the consumption-based CAPM. Household panel data are used to construct time series on the consumption of each group. The results indicate that the consumption of stockholders is more volatile and more highly correlated with the excess return on the stock market. These differences help explain the size of the equity premium, although they do not fully resolve the equity premium puzzle.

This paper presents and implements statistical tests of stock-market forecastability and volatility that are immune from the severe statistical problems of earlier tests. It finds that although the null hypothesis of market efficiency is rejected, the rejections are only marginal. The paper also shows how volatility tests and recent regression tests are closely related, and demonstrates that when finite sample biases are taken into account, regression tests also fail to provide strong evidence of violations of the conventional valuation model.