The Size Effect Is Alive and Well, and Hiding behind Calendar Anomalies

Abstract

Empirical evidence from the literature documents that the size effect has gradually diminished since the early 1980s. In this article, the authors examine the stability of this anomaly between 1926 and 2014. The evidence they document indicates a substantial presence of the size effect across various calendar anomalies. The anomalies considered are the January effect, the Halloween effect, the Mark Twain effect, the other January effect, the seasonal affective disorder effect, and the turn-of-the-month, day-of-the-week, and week-of-the-year effects. Their findings hold true for different sample periods, various modeling specifications, and types of returns and may provide investment professionals such as portfolio managers, investment officers, analysts, and other market participants with a practical tool for their daily decision-making.

TOPICS: Factor-based models, factors, risk premia