We perform the longest study of long-run reversal in commodity returns. Using a unique dataset of 52 agricultural, industrial, and energy commodities, we examine the price behavior for the years 1265 to 2017. The findings reveal a strong and robust long-run reversal effect. The returns of the past one to three years negatively predict subsequent performance in the cross-section of returns. The effect is robust to extensive subsample and subperiod analysis, and not driven by statistical biases, extreme events, or macroeconomic risks. Our findings support the explanation that the long-term reversal originates from supply and demand adjustments following price changes. Finally, the phenomenon is elevated in more volatile commodities and in periods of high return dispersion.

We investigate the investability of commodity risk premia in China. Previously documented standard momentum, carry and basis-momentum factors are not investable due to the unique liquidity patterns along the futures curves in China. However, dynamic rolling and strategic portfolio weights significantly boost the investment capacity of such premia without compromising its statistical and economic significance. Meanwhile, style integration delivers enhanced performance and improved opportunity sets. Furthermore, the observed investable premia are robust to execution lags, stop-loss, illiquidity, sub-period specifications, seasonality and transaction costs. They also offer portfolio diversification for investors. Finally, investable commodity premia in China reveal strong predictive ability with global real economic growth.

This paper examines profitable trading strategies that jointly exploit momentum and reversal signals in commodity futures. While the single-sort momentum strategies returns 11.14% per annum, on average, a consistent reversal pattern of momentum profits is pronounced from 12 to 30 months after portfolio formation. Combining the observed reversal pattern with the momentum signal, our double-sort strategy returns 20.24% per annum, which significantly outperforms single-sort strategies. The proposed strategy is robust to seasonality effects and sample adjustments in commodity futures. The profitability of the double-sort strategy cannot be explained by standard risk factors, term structure, market volatility, investor sentiment, data-mining or transaction costs, but appears to be related to global funding liquidity. As a consequence, the double-sort strategy in commodity futures may be employed as a portfolio diversification tool.

The Nelson-Siegel framework is employed to model the term structure of commodity futures prices. Exploiting the information embedded in the level, slope and curvature parameters, we develop novel investment strategies that assume short-term continuation of recent parallel, slope or butterfly movements of futures curves. Systematic strategies based on the change in the slope generate significant profits that are unrelated to previously documented risk factors and can survive reasonable transaction costs. Further analysis demonstrates that the profitability of the slope strategy increases with investor sentiment and is in part a compensation for the drawdowns incurred during economic slowdowns. The profitability can also be magnified through timing and persists under alternative specifications of the Nelson-Siegel model.