LOADING
Momentum is the empirical observation that stocks that have performed well over the past 3 to 12 months tend to continue performing well, and stocks that have performed poorly tend to continue performing poorly. Jegadeesh and Titman (1993) provided the seminal academic evidence, showing that buying past winners and shorting past losers generated significant profits. The momentum premium has been one of the most robust anomalies in finance, documented across 40+ countries and multiple asset classes including bonds, commodities, currencies, and real estate.
The standard momentum signal is the total return over months 2 through 12 (skipping the most recent month). The most recent month is excluded because of the short-term reversal effect: stocks that rallied sharply in the past few weeks tend to experience a brief pullback. This "skip month" construction is nearly universal in academic and practitioner implementations. Stronger signals can be constructed by using risk-adjusted returns (residual momentum) or by combining price momentum with earnings momentum (the tendency for positive earnings surprises to be followed by more positive surprises).
The behavioral explanation for momentum centers on two cognitive biases: underreaction and overreaction. Investors initially underreact to new information, such as a positive earnings report, allowing the stock to gradually drift upward. As the price trend becomes obvious, momentum traders and trend followers pile in, eventually pushing the stock beyond its fundamental value (overreaction). The momentum premium is captured during the underreaction phase, while the eventual overreaction creates the conditions for the crash risk that momentum strategies are exposed to.
The primary risk of momentum strategies is the momentum crash. During sharp market reversals, particularly the transition from bear to bull markets, recent losers rally explosively while recent winners collapse. The most dramatic example occurred in March 2009, when momentum strategies lost decades of accumulated profits in a matter of weeks. This crash risk is one reason why momentum premia are persistent: rational investors demand compensation for bearing it. Risk management techniques such as dynamic hedging, volatility scaling, and combining momentum with value (which is negatively correlated) can mitigate but not eliminate crash risk.
Cross-sectional momentum (comparing stocks against each other) and time-series momentum (comparing a stock against its own history) are related but distinct concepts. Time-series momentum, also called trend following, goes long assets with positive recent returns and short assets with negative recent returns. It has a strong track record in futures markets and tends to perform well during prolonged market dislocations. Combining both types of momentum can create more diversified and robust strategies.