Managed Futures vs Factor Investing: A 100-Year Perspective

Rethinking traditional asset allocation

June 2026. Reading Time: 10 Minutes. Author: Nicolas Rabener.

SUMMARY

  • Managed futures generated higher returns than a simple market-neutral multi-factor portfolio
  • Naturally, the factor selection matters
  • Combining both generated the largest diversification benefits

INTRODUCTION

In our recent research report, Bonds vs. Managed Futures: A 100-Year Perspective, we showed that managed futures delivered significantly higher returns and lower drawdowns than U.S. government bonds over the past century. The contrast was particularly striking in real terms, where bonds suffered a 61% drawdown. These findings stand in sharp contrast to conventional asset allocation frameworks, in which bonds occupy a central role while managed futures remain largely absent.

The limited adoption of managed futures is especially puzzling given the extensive body of academic research supporting the strategy. A similar pattern can be observed in factor investing: investors frequently rely on evidence derived from long-short factor research, yet allocate predominantly to long-only implementations.

In this report, we examine the long-term performance of managed futures and market-neutral multi-factor investing over the past century.

PERFORMANCE OF MANAGED FUTURES VS MARKET-NEUTRAL MULTI-FACTOR INVESTING

To compare the long-term performance of managed futures and market-neutral multi-factor investing, we use a backtested managed futures index developed by AQR, which provides nearly 100 years of data. As the AQR index is available only through 2020, we extend the series using realized returns from the AQR Managed Futures Strategy Fund (AQMIX) thereafter.

For market-neutral multi-factor investing, we construct an equal-weighted portfolio of the long-short size, value, and momentum factors, rebalanced annually using data from the Kenneth R. French Data Library. Both indices are based on backtests that exclude transaction costs and therefore likely overstate realized returns.