Industry Anomalies: An examination of asset pricing anomalies through an industry-specific framework

Abstract: The finance literature has discovered a large number of anomalies in the cross-section of stock returns over the past three decades. This thesis examines whether some of the most robust anomalies also appear within industries, and whether some are more prominent than others within specific industry sectors. We construct Fama-French type industry portfolios sorted on variables which have historically provided for higher expected returns in the aggregate market data: size, value, operating profitability, investment and momentum. We find comprehensive evidence that patterns of anomalous average returns observed in the U.S. stocks at large persist within industry subsets of the market, and that they exhibit differing premiums across industry sectors. We also find that industry-specific anomaly-based trading strategies can be formed in a way that delivers risk-adjusted outperformance over the historical market premium, but that said strategies tend to also be largely explained by the aggregate Fama-French risk factors. Lastly, we provide evidence that the industry-specific anomalous premiums vary across different time samples and put forward a case for further research on their persistence and predictability, as well as on optimal holding periods of their respective trading strategies.