Momentum In Industries

Summary

• There's a large body of evidence supporting the benefits of momentum strategies, both time-series (trend or absolute) and cross-sectional (relative) momentum.
• New research analyzed the performance of industry momentum strategies, providing a test of pervasiveness.
• Findings demonstrate the efficacy of industry momentum strategies using various holding periods, providing test of robustness.

Momentum is the tendency for assets that have performed well (poorly) in the recent past to continue to perform well (poorly) in the future, at least for a short period of time. There are two types of momentum: cross-sectional momentum (a measure of relative performance) and trend-following (or time-series) momentum (an absolute measure of performance). As presented in “ Your Complete Guide to Factor-Based Investing,” both types have been found to be persistent across time and economic regimes; pervasive around the globe and across stocks, bonds, commodities and currencies; robust to various formation periods; and have well-documented behavioral explanations for why they have persisted post-publication. And with the use of patient trading strategies, they survive transaction costs. In another test of pervasiveness, the December 2018 paper “ Factor Momentum Everywhere” demonstrated that there is also momentum in factors.

Klaus Grobys and James Kolari contribute to the literature with their study “On Industry Momentum Strategies,” published in the Spring 2020 issue of The Journal of Financial Research. They investigated the asset pricing implications of different industry momentum strategies, sorting U.S. industry portfolios into quintiles. The first group comprised industries that had the lowest returns in the previous month, and the fifth group comprised industries with the highest returns in the previous month. They built 48 U.S. industry portfolios using data from Kenneth French’s website. Portfolios were value weighted. A zero‐cost strategy was constructed that was long (short) the fifth (first) portfolio group. Their sample period was July 1926 to February 2018. They built three strategies: a 1-0-1 strategy (portfolio formed based on prior month returns and held for one-month); a 6-1-1 strategy (portfolio formed based on prior six-month returns, excluding the most recent month, and held for one month); and a 12-1-1 strategy (portfolio formed based on prior 12-month returns, excluding the most recent month, and held for one month). Traditional industry momentum strategies cited in the literature had examined the 6-1-1 and 12-1-1 strategies. Following is a summary of their findings:

• Industry portfolios outperforming in the previous month subsequently generated on average significantly higher returns than those underperforming in the previous month. Returns increased monotonically as they moved across quintiles. The winner-minus-loser portfolio returned 0.62 percent per month (t-stat of 4.6) with a Sharpe ratio of 0.49.
• The 6-1-1 portfolio returned 0.57 percent per month (t-stat of 3.5) with a Sharpe ratio of 0.48; and the 12-1-1 portfolio returned 0.80 percent per month (t-stat of 5.2) with a Sharpe ratio of 0.55. In both cases, returns basically increased monotonically as they moved across quintiles.
• Skewness risk was lowest for the 1-0-1 strategy.
• Managing the portfolio by targeting (scaling) volatility added value, reduced skewness and increased the Sharpe ratio.
• Combining signals reduced risk (there was low correlation of tail risk between the 1-0-1 and the 6-1-1 and 12-1-1 strategies) and increased the Sharpe ratio. Risk‐ managed strategies generated average payoffs that were statistically uncorrelated with bear market states, the market factor, and the market factor conditional on bear market states.
• Controlling for traditional industry momentum did not diminish the significance of the 1-0-1 strategy or its economic magnitude.

Summary

Given the risks of data mining, before investing in a factor, one should be confident there is sufficient evidence to support the belief that the premium found in the historical record is likely to be predictive of future results. To gain that confidence, you should have evidence that the premium was statistically significant and unique/independent as well as that it showed persistence, pervasiveness, robustness, implementability (survive transactions costs) and has risk-based or behavioral explanations for why it should be expected to continue. By examining the 1-0-1 strategy, Grobys and Kolari contribute to the literature by demonstrating evidence of the robustness of time-series momentum as well providing further evidence of its pervasiveness across industries. They also demonstrated that historically combining a shorter-term momentum measure with more traditional measures improved results.

Analyst’s Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.