By isolating stock-specific momentum from earnings announcements, the study shows it predicts returns across the US, Europe, and Japan. This approach avoids factor momentum bias, supports underreaction theories, and revives interest in the post-earnings drift.
Interested in having your article published in the Financial Analysts Journal? Find out how.
Abstract
This study aims to investigate the recent controversy surrounding the existence of stock-specific momentum. Stock momentum consists of both factor- and stock-specific components, but the risk associated with factor momentum might hinder the impact of stock-specific momentum. Using earnings announcement returns that occur during the formation months of the stock momentum strategy, the study captures a component largely unaffected by factor momentum, thereby mitigating the bad-model problem. This stock-specific momentum source predicts future returns, does not reverse in the long run, and is pervasive, as similar results are found in the US, Europe, and Japan over the last 30 years.