Stock Line

The Momentum Factor: An Empirical Update

Stock Line

The Momentum Factor has been well documented since the early 1990s[i]. As a quick reminder, the Momentum Factor is the tendency of stocks that have outperformed (or underperformed) the market over the last three, six, and twelve month periods to continue to outperform (underperform) the market. We have written extensively about the Momentum Factor over the years here and here.

Among the academic community, there is little dispute as to whether or not the Momentum Factor exists. In fact, many practitioners have taken the well-known Fama/French 3 Factor Model and have expanded it to include momentum when looking at the cross-section of stock returns overtime. Where there is dispute, is whether or not momentum can be captured in an actual investment strategy so that it can be of benefit to investors.

We articulated the reasons for the difficulty in capturing momentum before in an article, but the main point is that because momentum only persists for a short period of time, an actual investment strategy would need a tremendous amount of turnover in its holdings in order to keep up with it. Turnover in a portfolio creates numerous costs such as taxes, market impact, and actual commissions from trading.

Some investment companies attempt to isolate the momentum factor within their investment strategies, like AQR (Applied Quantitative Research) Capital Management. Their leadership has close ties to the University of Chicago and is very much aware of the dimensions of expected return, market efficiency, and modern portfolio theory. Nonetheless, they believe that they can effectively capture the benefits of momentum and thus better diversify their investors.

You also have investment companies, like Dimensional Fund Advisors (DFA), who believe that the costs associated with trying to isolate momentum are too great. Instead, they screen for momentum when trading within their investment strategies as to not buy or sell a security too early if it is exhibiting positive or negative momentum. This way, they are able to take advantage of the momentum factor, but without incurring substantial costs.

It is still too early to determine whether or not either approach to incorporating momentum into investment strategies is superior, but we believe in keeping investors updated if there happens to be new information.

Dimensional Fund Advisors recently examined the entire US mutual fund landscape to see if there was anything exciting to report[ii]. For the 15 years ending 12/31/2014, DFA’s research team ran regressions to find strategies that had exposure to momentum during that time. The Momentum Factor, as measured by the Fama/French momentum up-minus-down (UMD) factor, was 6.68% per year during that time. In other words, in theory, an investor who was solely long stocks that exhibited positive momentum and short stocks that exhibited negative momentum during that time would have earned 6.68% per year. This return was accompanied by an annual turnover of 316%, which comes out to an average holding period of slightly less than 3 months. Again, this is a common trait amongst strategies trying to capture momentum.

The real question is whether or not having exposure to momentum actually benefitted investors in practice, not theory. DFA categorized groups of mutual funds based on their degree of exposure to the momentum factor during that time. Here is a quick synopsis of their results:

  • On average, strategies with higher exposure to momentum were not associated with higher returns over this period. In fact, the group of mutual funds with the highest exposure to momentum lagged funds with the smallest exposure to momentum by 0.05%-0.77% per year.
  • On average, funds with higher momentum loadings had positive loadings on the SMB (small-minus-big) factor and negative loadings on the HML (high book-to-market minus low book-to-market stocks). In other words, strategies that were exposed to momentum were on average smaller than the overall market and more tilted towards growth stocks.
  • Turnover in the strategies that exhibited significant momentum averaged approximately 173%-204, which is very significant.
  • The expense ratios of the strategies with the most momentum averaged between 1.13% and 1.45%.

See the table below for a more visual representation of this information. The Mu (μ) simply indicates the amount of exposure to momentum. For example, μ>0.3 indicates the mutual funds with the highest exposure to momentum.

 

While this is not the end all be all analysis that will settle the debate on what portfolio managers should do in terms of momentum, it does highlight the difficulty associated with trying to capture market premiums in strategies that have high expense ratios and high turnover - as is characteristic of strategies trying to capture momentum. Although the Momentum Factor was positive during the 15 years ending 12/31/2014, this benefit did not get passed on to investors who were exposed to it. Our investors are already benefitting from exposure to momentum through the trading strategies implemented by Dimensional Fund Advisors. We will continue to monitor any research and performance figures to ensure that our clients’ portfolios are structured in a way that gives it the highest probability of reaping the benefits that the market provides.

Stay tuned.



[i] Narasimhan Jegadeesh and Sheridan Titman, “Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency,” Journal of Finance 48 (1993): 65-91.

[ii] Crill, Wes. Momentum Loadings and Mutual Funds. Dimensional Fund Advisors, July 1, 2015.