Key Takeaways
- • Commodity carry—the return accruing on long futures positions when all prices remain constant—demonstrates strong predictive power for subsequent commodity returns across a 23-commodity panel spanning 2000–2026, with evidence particularly robust when adjusted for inflation, seasonality, and volatility.
- • Real carry metrics incorporate adjustments for implicit subsidies including convenience yields and hedging premia, materially improving signal precision compared to unadjusted nominal carry metrics.
- • Naïve profit-and-loss analyses based on real carry strategies generate material economic value both as standalone trading signals and as overlays to commodity long exposure, with relative carry signals demonstrating greater potency than directional signals.
- • The research is underpinned by proprietary Macrosynergy methodology requiring access to J.P. Morgan DataQuery and JPMaQS quantamental indicators, with full audit and replication possible via downloadable Jupyter notebooks.
Summary
Commodity carry operates as a measurable trading signal with demonstrated predictive power for future commodity price movements, according to proprietary research from Macrosynergy updated through June 2026. Carry is defined as the return that would accumulate on a long futures position if prices along the futures curve remained static—a metric determined by the shape of that curve. When commodity futures curves exhibit backwardation, carry is positive; contango produces negative carry. The fundamental insight is that this curve shape encodes information about implicit subsidies and risk premia within commodity markets, particularly convenience yields (the benefit of holding physical inventory) and hedging premia (compensation for risk-bearing).
The research demonstrates that real carry metrics—adjusted versions accounting for inflation, seasonal patterns, and volatility—substantially enhance signal precision relative to nominal carry. These adjustments refine the raw carry signal by filtering out distortions that obscure the true return opportunity. Across a broad sample of 23 commodities analyzed from 2000 onward, such adjusted carry metrics exhibit statistically significant predictive relationships with subsequent futures returns.
Beyond predictive accuracy, the research documents concrete economic value generation. Stylized portfolio strategies constructed on real carry signals produce material performance gains when deployed independently or layered onto commodity long-only exposure. Notably, relative carry signals—which compare carry across multiple commodities to identify relative value—have proven more economically potent than directional signals that rely on price trends alone. This finding suggests that the cross-commodity information embedded in carry differentials contains sharper alpha potential than momentum-based approaches.
The research is reproducible through proprietary Jupyter notebooks available for download, with full audit capability contingent upon access to J.P. Morgan's DataQuery platform and JPMaQS quantamental indicators. The methodology builds on Macrosynergy's prior analysis in "Commodity carry as a trading signal – part 1," which introduced foundational concepts of carry mechanics in commodity futures markets.
The research demonstrates that real carry metrics—adjusted versions accounting for inflation, seasonal patterns, and volatility—substantially enhance signal precision relative to nominal carry. These adjustments refine the raw carry signal by filtering out distortions that obscure the true return opportunity. Across a broad sample of 23 commodities analyzed from 2000 onward, such adjusted carry metrics exhibit statistically significant predictive relationships with subsequent futures returns.
Beyond predictive accuracy, the research documents concrete economic value generation. Stylized portfolio strategies constructed on real carry signals produce material performance gains when deployed independently or layered onto commodity long-only exposure. Notably, relative carry signals—which compare carry across multiple commodities to identify relative value—have proven more economically potent than directional signals that rely on price trends alone. This finding suggests that the cross-commodity information embedded in carry differentials contains sharper alpha potential than momentum-based approaches.
The research is reproducible through proprietary Jupyter notebooks available for download, with full audit capability contingent upon access to J.P. Morgan's DataQuery platform and JPMaQS quantamental indicators. The methodology builds on Macrosynergy's prior analysis in "Commodity carry as a trading signal – part 1," which introduced foundational concepts of carry mechanics in commodity futures markets.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.