Does Factor Timing Pass The Smell Test?

Academics and investment practitioners have long performed deep dives on momentum as a source of excess return premia. Analysis that focuses on value, small-cap and other accounting-based factors (book to market, capitalization, etc.) is a more recent addition to the research canon, but the seminal papers by Fama and French from the 1990s aren’t spring chickens either. In recent years a new line of inquiry that combines the two disciplines seeks to answer the question: Is timing of the standard risk factors based on momentum signals a worthy pursuit in its own right? A growing list of researchers are finding reasons to answer “yes,” or at least throw cold water on dismissing the idea outright.

The latest effort is a study that considers a “managed factors” strategy, which reportedly outperforms an equivalent “plain factor” portfolio. “The value-added induced by momentum-based timing is pervasive, survives transaction costs, is different from performance enhancements that are generated by traditional timing strategies, and carries over to multi-factor portfolios,” report Volker Flögel (Quoniam Asset Management) and two co-authors in “Momentum-Managed Equity Factor.”

Factor timing, like all forms of market timing, is a controversial subject, at least in some circles. But an accumulating pile of research and empirical investment evidence opens the door for embracing factor timing in some degree. The devil’s in the details, as always, but as Flögel, et al.’s analysis suggests, one of the more productive applications appears to be analyzing factor premia through a momentum (timing) lens.

The Momentum-Managed Equity Factor paper focuses on seven equity factors:

  • Market (MKT)
  • Size (SMB)
  • Book-to-market-equity (HML)
  • Momentum (WML)
  • Investment (CMA)
  • Operating profitability (RMW)
  • Volatility (SMV)

The timing signal is calculated at the end of each month on the expected next-month excess return of the “plain” factor. The forecast is derived from a simple econometric analysis of an expanding window of past monthly excess return premia. On this basis, “a managed factor then systematically scales the excess returns of the respective plain factor up or down over time,” the authors explain.

As we’ll see, the overall results are encouraging for thinking that market-timing signals drawn from the autocorrelation data (a form of momentum) of monthly excess returns lays the groundwork to earn a premium over the plain factors.

Squeezing More Alpha From Factor Trends

The table below summarizes the paper’s most significant observations. The key takeaway: managed factors tend to outperform their standard (plain) counterparts.  More specifically, in the case of six of the seven factors considered, the managed variety deliver higher mean excess returns. For example, the size factor (SMB, or small minus big) earned an average 11.7% for the sample period, nearly double the plain factor’s 6.2%.

The outlier in the study: the momentum factor (WML, or winners minus losers), which earns similar returns in managed and plain forms. That’s ironic, given the study’s focus. Apparently, trying to manage a plain momentum factor with an additional momentum overlay is going a step too far. Note, however, that the plain momentum factor is already a force to be reckoned with a 13.3% mean return, which leaves the other plain factors in the dust. Perhaps that’s why the extra momentum aspect hits a wall. There are limits to how much alpha you can squeeze out of any given methodology.

Another notable result: the worst performer for average excess return for the plain factors is the portfolio formed on operating profitability (RMW, or robust minus weak). That’s not surprising, given the existing research, advise Flögel, et al. Filtered through the managed factor lens, however, RMW comes alive by doubling its excess mean return.

The overall results show that significant additional alpha can be generated when factors are managed using momentum signals, even after adjusting for volatility. Panel C in the table “shows that the mean excess returns and Sharpe ratios grow considerably for five out of seven factors when timing is employed,” the researchers note. The biggest return increase is in the size factor (SMB), which adds 5.57 percentage points to its mean excess return through the managed factor model, closely followed by a 5.52-percentage-points gain for operating profitability (SMV).

Even more impressive, the results in favor of the managed factors survive after backing out transaction costs and running the analysis on multi-factor portfolios.

Something New Under The Sun?

By the authors’ reckoning, the study’s results represent “a new timing strategy which exploits the predictive power of the returns of equity factors over the previous month.”

Not surprisingly, there are skeptics who question whether reforming and revising Mr. Market’s capitalization-weighted investment strategy will allow investors to outperform through time. But the skeptics are having a tougher time dismissing the idea in recent years as more researchers find reasons to embrace timing, if only slightly.

In 2016, for example, Rob Arnott (founder of Research Affiliates) outlined (with assistance from two colleagues) the evidence for another way to slice and dice timing a la factors. They report on “a contrarian timing approach—emphasizing factors or strategies trading cheap relative to their own historical norms, and deemphasizing the more expensive factors or strategies.” The conclusion: this approach “can improve performance, but should be used in moderation to avoid increasing portfolio risk from a loss of diversification.”

But AQR Capital Management’s Cliff Asness begs to differ: “If you choose your factors based on only past performance, even long-term past performance, that can be a recipe for poor results in a world where data mining is a problem.”

The Flögel, et al. research suggests otherwise, although it probably won’t change Asness’ mind, at least not much. Studies that uncover statistically significant alpha, after all, already come in a wide variety of flavors. The ability to replicate study results out-of-sample using real money, by contrast, is a rare bird.

However, Asness has also recommended investors “sin a little (perhaps very little)” with market timing via value investing and contrarian-based strategies. He recognizes the contradiction in this suggestion and his critique of Arnott and company’s mean reversion approach. But Asness defends his quasi-bifurcated stance by emphasizing that some factors deserve more attention than others. Even so, he recommends caution on making extreme bets, no matter how attractive they appear in real time.

The Milwaukee Company’s Classic Asset Allocation Rebalancing (“CAAR”) strategy takes a different approach to managing factors.  CAAR utilizes Mean Variance Optimization (“MVO”) to mathematically determine allocations to ETFs that target well-established market risk factors, such as value, momentum, quality, and high dividends. The objective is to identify the combination of factors that are projected to generate highest possible return for a specific level of risk, or equivalently offer the lowest possible risk for a given level of return. Each level of risk or return is calculated based on each factor-based ETFs performance over a trailing 12 month period.

Additionally, CAAR utilizes The Milwaukee Company’s Market Trend Indicator (“MTI”). MTI seeks to identify changes in the state of the market by comparing the trailing 12-month performance of the U.S. stock market to that of the U.S. bond market. If MTI identifies that stocks are in a bull market, neutral market, or bear market it will adjust the portfolio’s target allocations to overweight, target weight, or underweight equities, respectively.

While there is room for further study and discussion, the accumulating research on factor timing is making it tougher to argue that that the best way to tap into factor premia is to buy and hold. That can work, of course, but it’s also clear that a fully passive approach is no longer the only game in town.

By James Picerno, Director of Analytics

IMPORTANT DISCLOSURES:  PLEASE REMEMBER THAT PAST PERFORMANCE MAY NOT BE INDICATIVE OF FUTURE RESULTS.  DIFFERENT TYPES OF INVESTMENTS INVOLVE VARYING DEGREES OF RISK, AND THERE CAN BE NO ASSURANCE THAT THE FUTURE PERFORMANCE OF ANY SPECIFIC INVESTMENT, INVESTMENT STRATEGY, OR PRODUCT MADE REFERENCE TO DIRECTLY OR INDIRECTLY FROM THE MILWAUKEE COMPANY™, WILL BE PROFITABLE, EQUAL ANY CORRESPONDING INDICATED HISTORICAL PERFORMANCE LEVEL(S), OR BE SUITABLE FOR YOUR PORTFOLIO.  DUE TO VARIOUS FACTORS, INCLUDING CHANGING MARKET CONDITIONS, THE CONTENT MAY NO LONGER BE REFLECTIVE OF CURRENT OPINIONS OR POSITIONS.  MOREOVER, YOU SHOULD NOT ASSUME THAT ANY DISCUSSION OR INFORMATION CONTAINED IN THE MILWAUKEE COMPANY™ SERVES AS THE RECEIPT OF, OR AS A SUBSTITUTE FOR, PERSONALIZED INVESTMENT ADVICE FROM THE MILWAUKEE COMPANY™

Spread the love

Leave a Reply

Your email address will not be published. Required fields are marked *