We examine the methods commonly employed to estimate Tobin's q ratios and find them to be flawed in design and arbitrary in implementation. We propose an alternative procedure which is both simpler and more accurate. The key to the procedure is an improved measure of fixed asset replacement costs, through the proper identification of the vintages of fixed assets that are in place for a firm. Application of this procedure to a large sample of nonfinancial corporations indicates that existing methods generally produce downward-biased measures of q and can result in errors in the ordering of firms by their q's.

Research has shown that managers display self-serving behavior in a variety of discretionary information production decisions. We test whether such behavior is also manifest in discretionary information disclosure decisions — in particular, in the common stock return performance comparisons now required in corporate proxy statements. We find evidence that the industry and peer-company stock return benchmarks, and broader market indices, chosen by management for those comparisons are downward biased, thereby overstating relative reporting-firm performance. Cross-sectionally, the extent of the bias varies with key reporting-firm attributes, including firm performance and the character of firm ownership structure.

Tax‐loss selling by investors in common stocks near the end of calendar years has been proposed as an explanation for the turn‐of‐the‐year effect in stock returns. Past analyses of this hypothesis have relied on inferential data. We provide here some direct data from a compilation of over 80,000 actual common stock investment round trips by a sample of 3000 individual investors. We find strong evidence of a concentration of loss‐taking trades late in the year and milder evidence of a concentration just prior to the dates when investments become eligible for long‐term tax treatment.

Tax-loss selling by investors in common stocks near the end of calendar years has been proposed as an explanation for the turn-of-the-year effect in stock returns. Past analyses of this hypothesis have relied on inferential data. We provide here some direct data from a compilation of over 80,000 actual common stock investment round trips by a sample of 3000 individual investors. We find strong evidence of a concentration of loss-taking trades late in the year and milder evidence of a concentration just prior to the dates when investments become eligible for long-term tax treatment.

Tax-loss selling by investors in common stocks near the end of calendar years has been proposed as an explanation for the turn-of-the-year effect in stock returns. Past analyses of this hypothesis have relied on inferential data. The authors provide here some direct data from a compilation of over 80,000 actual common stock investment round trips by a sample of 3,000 individual investors. The authors find strong evidence of a concentration of loss-taking trades late in the year and milder evidence of a concentration just prior to the dates when investments become eligible for long-term tax treatment.