Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency¶
Why this mattered¶
Jegadeesh and Titman’s 1993 paper mattered because it made intermediate-horizon momentum a central empirical fact for asset pricing. Earlier market-efficiency debates had often focused on whether prices followed random walks or whether apparent predictability could be explained by risk, size, or other known effects. This paper showed, with a simple and replicable portfolio design, that buying recent winners and selling recent losers produced significant returns over 3- to 12-month horizons, and that these returns were not readily explained by conventional measures of systematic risk or by delayed reactions to common factors. The result was difficult to dismiss as a technical anomaly because the strategy was economically intuitive, statistically strong, and directly tied to the core claim of market efficiency: that publicly available past prices should not predict abnormal returns.
The paper also shifted what researchers could ask. Momentum became a benchmark anomaly, not merely a curiosity, forcing asset-pricing models to account for both continuation and later reversal: the same winner-minus-loser profits that appeared during the first year partly dissipated over the next two years. That pattern opened a path between two competing explanations. If momentum reflected compensation for risk, models needed to identify a priced risk that rose and fell in this distinctive way. If it reflected mispricing, researchers needed theories of investor behavior that could produce underreaction in the short run and correction or overreaction in the longer run. The paper’s evidence around earnings announcements strengthened this second line of inquiry by linking momentum profits to information processing rather than only to mechanical price trends.
Its influence is visible in later breakthroughs across empirical finance, behavioral finance, and investment practice. Momentum became one of the canonical challenges to the traditional efficient-market view, alongside size, value, and post-earnings-announcement drift, and it helped motivate behavioral models of underreaction, overconfidence, and gradual information diffusion. It also became a practical “factor” in quantitative investing and later multi-factor asset-pricing work, where any credible model of expected returns had to confront the persistence of winner-loser spreads. In that sense, the paper did not just document a profitable trading rule; it helped turn return predictability into a disciplined research program.
Abstract¶
ABSTRACT This paper documents that strategies which buy stocks that have performed well in the past and sell stocks that have performed poorly in the past generate significant positive returns over 3‐to 12‐month holding periods. We find that the profitability of these strategies are not due to their systematic risk or to delayed stock price reactions to common factors. However, part of the abnormal returns generated in the first year after portfolio formation dissipates in the following two years. A similar pattern of returns around the earnings announcements of past winners and losers is also documented.
Related¶
- cite → Risk, Return, and Equilibrium: Empirical Tests — The momentum paper cites Black, Jensen, and Scholes' CAPM tests as the benchmark risk-return framework challenged by winner-loser stock returns.
- cite → A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix — The momentum paper cites Newey and West's HAC covariance estimator to support inference robust to heteroskedasticity and autocorrelation in return regressions.
- cite → The Cross‐Section of Expected Stock Returns — The momentum paper cites Fama and French's size and book-to-market results as cross-sectional return patterns that momentum must be distinguished from.
- enables ← Risk, Return, and Equilibrium: Empirical Tests — Fama-MacBeth empirical asset-pricing tests supplied the risk-return testing framework used to evaluate whether momentum profits violate market efficiency.
- enables ← A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix — Newey-West HAC covariance estimation enabled robust inference for autocorrelated stock-return strategies such as momentum portfolios.