Trading on Calendar Misconceptions: 14-Week Quarter Revisited

Abstract

This study examines the abnormal returns caused by the 14-week quarter anomaly, a phenomenon where US firms define their fiscal quarters as 13 weeks to maintain precisely 52 weeks in each fiscal year. Consequently, one day (two days in leap years) is omitted annually. To realign with the calendar year, a week is added back every 5 or 6 years, resulting in one 14-week fiscal quarter during those periods. Over a decade ago, Johnston et al. (2012) introduced the 14-week quarter anomaly. This thesis revisits and extends their study. The main findings of the thesis are that the 14-week quarter anomaly still persists, even after the publication of Johnston et al. (2012), and that it is more pronounced for firms with higher idiosyncratic risk. Specifically, investors can gain a 3.8 percentage point (16.1 pp annualized) abnormal return by merely buying and holding stocks during the 14-week quarters. The thesis contributes to the literature on market efficiency, arbitrage, and financial reporting. The thesis provides new evidence on the existence and persistence of the 14-week anomaly, which challenges the efficient market hypothesis and suggests that investors are not fully rational or informed. The thesis also provides new insights into the role of idiosyncratic risk in explaining the 14-week quarter anomaly, which implies that arbitrage is limited or costly for these firms. Furthermore, the thesis sheds light on the implications of accounting policy choices for firm returns and investor behavior. This thesis is built upon an extensive review of existing literature and examines data from 591 US firms spanning the years 2005 to 2023. Regression analyses employed in this study utilize a fixed effects method, account for industry- and year-effects, and employ heteroskedasticityrobust standard errors clustered at the firm level.