How Anomalous are the Anomalies in Common Stock Returns? Fresh Evidence from Pre-World War I Germany

Abstract

Using a new set of monthly stock price data for a random sample of German companies, we investigate the pattern of common stock returns on the Berlin Stock Exchange between 1904 and 1910, when it ranked among the very top markets worldwide. We find that the CAPM performs poorly in this context: beta does not relate significantly to returns, while additional factors do. However, the anomalies we uncover differ substantially from those found in the United States more recently. Specifically, we find that book-to-market ratios relate negatively to returns, while size relates positively (but weakly) to returns. Also at odds with U.S. experience, earnings ratios are insignificant (though positive) in predicting returns, and a momentum portfolio earns returns not different from zero, on average. In addition, we explore the impact of bank directors sitting in company boards, since these bankers are thought to have wielded influence over firms and possibly over the price-setting process at the exchanges. Controlling for selection bias, however, we find that bankers had no consistent effect on common stock returns.