By Mark Hulbert, MarketWatch
CHAPEL HILL, N.C. — You’d have made more money over the last two months by betting on the stock market’s prior losers than on the winners. That’s just the opposite of what Wall Street’s well-known “momentum effect” expects.
While this reversal was partially caused by temporary factors relating to the new bull market that began in late March, it also reflects more enduring changes in the market that could very well diminish the momentum effect’s profitability going forward.
That would represent a serious setback to what once was one of Wall Street’s most profitable strategies. Since 1926, a portfolio of the stocks with the best trailing-year returns performed far better than a second portfolio containing the stocks with the worst—by a margin of 10.6 annualized percentage points, according to data from Ken French, a Dartmouth professor.
This historical pattern has been turned on its head since the stock market’s /zigman2/quotes/210599714/realtime SPX -0.56% March 23 lows, however. According to French’s data, the low-momentum portfolio since then through the end of April has beaten the high-momentum portfolio by 10.7 percentage points. And note that this margin of outperformance is unannualized; on an annualized basis, the low-momentum stocks’ advantage over that period is over 70 percentage points.
At least in part this reversal was to be expected, since momentum typically fails in the first couple of months after a bear-market bottom. That makes sense, of course. When the economy decides not to jump off a cliff, the best-performing stocks will be those basket cases that otherwise were near-certain bets of going bankrupt.
Recent research suggests that a longer-term shift in the markets may also be playing a role. That research finds both that the momentum effect is in large part caused by the behavior of retail investors, and that they represent a smaller and smaller proportion of overall trading volume.
The study, “ Retail Investors’ Contrarian Behavior Around News and the Momentum Effect ,” began circulating in February in academic circles. Its authors are Cheng (Patrick) Luo, the lead data scientist at Farallon Capital Management; Enrichetta Ravina, a professor at Northwestern University’s Kellogg School of Management; and Luis Viceira, a professor at Harvard Business School. The researchers reached their conclusions upon analyzing the trading records of 2.8 million individual accounts at a major discount brokerage firm from 2010 through 2014.
They found that, on balance, individual investors react in a contrarian way to earnings surprises. That is, they tend to sell stocks that have had a positive surprise and buy stocks with a negative surprise. Their behavior leads to momentum because it means that stocks underreact to their earnings surprises. As the market eventually corrects this underreaction, the positive-surprise stocks keep on winning and the negative-surprise stocks continue losing.
This aspect of human behavior is not new. What has changed is the declining proportion of trading volume from individuals. Because of that, Ravina told me in an interview, we’d expect the momentum effect to decline along with it.
And that’s exactly what we’re seeing, as illustrated in the chart at the top of this article. It plots the difference in annualized trailing 20-year returns of the highest-momentum and lowest-momentum stocks. Notice the distinct downtrend over the last four decades.
The returns reflected in this chart don’t take transaction costs into account. Since the momentum portfolios undergo lots of transactions, it’s entirely possible that, after transaction costs, high-momentum stocks enjoyed no actual advantage over low-momentum stocks over the last 20 years.
There are at least two major investment implications of this research. The first is that you may need to reduce your expectations for momentum approaches in coming years. The second is that you shouldn’t automatically bet that the market’s reaction to an earnings surprise will soon be reversed.
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at email@example.com.