Month of the year effect and January effect in pre-WWI stock returns: Evidence from a non-linear GARCH model

This paper investigates seasonal anomalies in the mean stock returns of Germany, the UK and the US during pre-World War I (WWI) period. The anomalies studied are month of the year effect and the January effect. The empirical research is conducted using a non-linear GARCH-t model, and monthly returns. Results obtained provide evidence of the January effect and the month of the year effect on the UK and the US returns. The German returns shows the month of the year effect but no January effect. Given the lack of tax treatment of capital gains/loss before 1914 by these countries, the results fail to provide merit to the tax-loss selling hypothesis of the January effect. Since we apply value-weighted returns in all cases, results obtained also fail to provide support for the small firm effect. Copyright (C) 2001 John Wiley & Sons, Ltd.