The Many Facets of Stock Momentum: Distinguishing Factor and Stock Components

Stock momentum—the practice of buying recent winners and selling recent losers—has long served as a foundational strategy in systematic investing. However, the source of momentum profits has been contested in academic literature, with critics attributing returns primarily to tilts toward broad factors such as value, size, and industry exposure rather than to genuine stock-specific effects. A 2025 paper by Gerard and Jehl published in the Financial Analyst Journal challenges this critique by isolating and demonstrating the persistence of a distinct momentum component driven by individual company fundamentals.

The research distinguishes between two momentum sources: factor-driven momentum and stock-specific momentum. Traditional 12-month momentum strategies carry substantial exposure to market, value, and industry effects, creating systematic risk that accounts for a meaningful portion of returns. The authors isolate the earnings-announcement component—the portion of 12-month momentum derived exclusively from price reactions in short windows surrounding each firm's earnings announcements over the prior year. This component demonstrates predictive power for future returns independent of broad factor movements, across the US, European, and Japanese markets over a three-decade period.

The earnings-announcement momentum strategy exhibits materially different risk characteristics than conventional momentum. It shows much lower systematic factor exposure and a high correlation between long and short portfolio legs, indicating that returns are driven by balanced, company-level information processing rather than directional factor bets. This structural difference makes the strategy a more isolated expression of firm-specific momentum, reducing drag from unintended factor tilts.

Notably, the persistence of longer-horizon earnings-based momentum contrasts sharply with the documented decay of short-term post-earnings announcement drift (PEAD) strategies, which have weakened significantly in recent years. This bifurcation suggests that the market may have become increasingly efficient at pricing individual earnings surprises in short windows, while the aggregated signal from multiple earnings reactions over a 12-month period retains exploitable predictive content. For institutional investors managing systematic momentum exposure, these findings support a shift toward earnings-announcement frameworks as a means of capturing momentum returns with reduced factor loading and potentially lower crowding risk.