Can Momentum Beat Buy-and-Hold?

Momentum, the tendency for assets that have performed well in the recent past to continue outperforming in the near term, is one of the most robust and extensively replicated empirical anomalies in finance. First formally documented by Jegadeesh and Titman (1993), who showed that buying recent winners and selling recent losers produced statistically significant returns over 3 to 12 month holding periods in U.S. equities, momentum has since been replicated across asset classes (bonds, commodities, currencies, real estate), geographies (developed and emerging markets), and time periods extending back over 200 years of available data.

The standard cross-sectional momentum strategy ranks stocks by their past 12-month return (skipping the most recent month, which exhibits a reversal effect) and holds the top decile long while shorting the bottom. Time-series momentum, rigorously studied by Moskowitz, Ooi, and Pedersen (2012), takes a different approach: instead of comparing assets against each other, it asks whether each asset's own past return is positive or negative and takes a long or short position accordingly. This approach is accessible to retail investors because it doesn't require shorting and can be implemented using sector ETFs.

The practical question for a long-only investor is whether sector-level time-series momentum, implementable with liquid ETFs and modest transaction costs, generates meaningful risk-adjusted excess returns after costs over a long historical window. This is a more conservative test than academic factor studies, which often use hypothetical portfolios with unlimited shorting and ignore implementation friction.

We use monthly GICS sector total returns from January 2000 to December 2023, sourced from FRED sector return indices and cross-validated against SPDR sector ETF (XL*) NAV data. The momentum signal for each sector is the 12-1 month return: the cumulative return over the past 12 months excluding the most recent month.

At the start of each month, we classify each of the 11 GICS sectors as "in" (positive 12-1 momentum) or "out" (negative 12-1 momentum). The strategy allocates equally to all sectors with positive momentum and holds cash (assumed to earn the prevailing 3-month T-bill rate) for sectors with negative momentum. If no sector has positive momentum, the portfolio is 100% cash. We compare this to a passive buy-and-hold allocation of 1/11 per sector, rebalanced monthly.

Performance metrics include: compound annual growth rate (CAGR), annualized Sharpe ratio (using monthly T-bill as the risk-free rate), maximum drawdown (peak-to-trough decline measured on a daily basis where daily data is available, otherwise monthly), and Calmar ratio (CAGR divided by maximum drawdown). We apply a round-trip transaction cost assumption of 0.10% (0.05% per leg), which is conservative relative to modern ETF bid-ask spreads and reflects typical institutional trading costs.