Swedroe: The Mystery Of Momentum

September 03, 2014

Momentum is large, persistent and—most of all—mysterious.

Momentum is a well-established, empirical fact. Its premium is evident in more than 87 years of domestic market data, in more than 20 years of out-of-sample evidence beginning from the time of its original discovery, in statistics from 40 other countries, and in the performance of more than a dozen different asset classes.

In fact, the momentum premium has been both larger and more persistent in the U.S. since 1927 than the other three stock premiums—equity, size and value. Over the last 87 calendar years (1927-2013), the annual momentum premium was 8.4 percent. It was positive in 78 percent of the years during that period.

By comparison, the figures for the equity, size and value premiums are not quite as strong or persistent. Over the same time frame, the equity premium was 8.2 percent, and it was positive in 68 percent of those years; the size premium was 3.1 percent, and it was positive in 56 percent of those years; and the value premium was 4.9 percent, while showing a positive return in 62 percent of those years.

While no one questions that the source of the equity premium is in taking greater risk with stocks, there is much debate in the academic community about the source of the other three premiums—especially value and momentum. There are numerous papers providing both behavioral and risk-based explanations for the value premium. For example, the risk side explanations include:

  • Value stocks have more leverage, both financial and operating
  • Value stocks have higher volatility of earnings and dividends
  • Higher returns to value stocks are systematically related to business cycle fluctuations as indicated by the macroeconomic variables of default and term spreads
  • Value companies have more irreversible capital
  • Value companies have more uncertain cash flows
  • Value companies are more subject to risk during monetary policy contractions, or periods associated with increasing risk of labor capital

On the other hand, behavioralists believe the value premium results from pricing mistakes. These occur when investors persistently overprice growth stocks (especially small growth stocks) and underprice value stocks. There are many academic papers that favor this view.

Explanations include the presence in the market of investors that tend to confuse the more familiar growth companies with safe companies and investors who prefer investments with lotterylike distributions; in other words, investors who are hoping to find the next Google. These investors opt for the small chance of winning a large sum, and they underestimate how quickly abnormal earnings (either faster or slower than average) revert to the mean.

When it comes to momentum, there are a few papers advocating a risk-based account, but most of the literature favors a behavioral explanation, either because of under- or overreaction. As Tobias Moskowitz points out in his paper, “Explanations for the Momentum Premium,” investor “under-reaction results from information travelling slowly into prices. That causes momentum.