Cross-sectional momentum is an investment strategy that ranks a universe of assets by their recent returns and buys the top-performing assets while avoiding (or shorting) the bottom-performing ones. It is the “relative” form of momentum investing — evaluating each asset not against its own historical return (as absolute momentum does) but against the returns of all other assets in the universe at the same time. The strategy exploits the empirical finding that assets that have recently outperformed their peers tend to continue outperforming over the next 3 to 12 months.

Academic Foundation

Cross-sectional momentum was formally documented by Jegadeesh and Titman in their landmark 1993 paper, which showed that U.S. stocks with strong 3-12 month past returns continued to outperform stocks with weak past returns over the following months. This finding was subsequently confirmed across global equity markets, other asset classes (commodities, currencies, fixed income), and various time periods — making cross-sectional momentum one of the most robustly documented return anomalies in academic finance.

Implementation

A typical cross-sectional momentum strategy: ranks all assets in the universe by their trailing 12-month return (excluding the most recent month, to avoid short-term reversal effects); buys the top-performing decile or quintile; shorts or avoids the bottom-performing decile or quintile; rebalances monthly or quarterly. The strategy works best applied to a broadly diversified universe — individual stocks, sectors, country indices, or global asset classes — rather than a small, highly correlated group.

Combining Cross-Sectional and Absolute Momentum

Research, particularly Gary Antonacci’s Dual Momentum work, shows that combining cross-sectional momentum (selecting relative leaders) with absolute momentum (only holding assets with positive absolute returns) produces a more robust strategy than either alone. The absolute momentum filter acts as a market timing overlay that reduces exposure to all assets during broad market downturns — even when cross-sectional leaders are still leading their falling peers downward.

---