There are fashions for everything: clothes, hairstyles, video games and hashtags on Twitter. And that applies to stockmarkets as well. Who can forget the enthusiasm for TMT (technology, media and telecom) shares in the late 1990s?
A new paper in the Journal of Portfolio Management suggests that this tendency may provide a strategy for outperforming the stockmarket, based on the popularity of individual stocks.
The authors defined the most popular stocks as those that saw the most trading in their shares as a proportion of their market value. These are most likely to be the companies that are in the news, perhaps because they have a hot new product or because many analysts are recommending them.
For a time these stocks may benefit from the so-called momentum effect — a phenomenon whereby stocks that have recently risen in price continue to perform well in the short term (usually a matter of months). However, this popularity may drive such stocks up to excessive valuations from which future returns are bound to be disappointing (in other words, they inevitably lose momentum sooner or later).
The chart shows the annualised return of American stocks, based on their popularity over the preceding year. Over a period of more than 40 years, the paper finds, stocks in the least popular quartile outperformed those in the most popular segment by seven percentage points a year.
As a control, the authors combined the popularity measure with another well-known effect, related to the volatility of individual stocks. Stocks that are more volatile than the market (rising or falling 10% when the index moves 5%) are described as having a “high beta;” stocks that are less volatile than the market are “low-beta.”
Under the capital asset-pricing model that is at the heart of academic finance, high-beta stocks should offer high returns to compensate for the higher risk. In fact, a 2011 study by GMO, a fund-management group, explains why low-beta stocks actually outperform their more volatile rivals over the long run.
The authors of the new paper combined the popularity and beta characteristics. They found that popularity was by far the dominant effect: whether a stock was popular was more important when determining its return than whether it was volatile. The same was true when controlling for other factors, such as the size of the company and the starting valuation.
The finding is significant. Academics have explained the long-term outperformance of small companies (the size effect) or those with below-average valuations (the value effect) in terms of compensation for extra risk. Small firms are more likely to go bust than large ones; cheap-looking stocks are usually cheap for a reason.
Both effects are thus theoretically compatible with the efficient-market hypothesis. But it is very hard to see how the momentum or popularity effects can be squared with the hypothesis, which supposes that all public information is already reflected in share prices and thus should be no help in determining future price movements.
The psychological reasons for the popularity effect are not hard to discern. Financial assets are not like other goods; when they rise in price, demand has a tendency to increase, not decrease. An investor who hears that a friend or neighbour had made money out of a particular stock will want to jump on the bandwagon.
The authors of the paper quote Ben Graham, the doyen of share analysts, as saying the market is not a weighing machine but a “voting machine whereon countless individuals register choices which are partly the product of reason and partly the product of emotion.”
Even professional fund managers may have good reasons for following a fad. They may want to show, in their reports to clients, that they have been smart enough to buy the hottest stocks of the year. In addition, clients have a natural tendency to fire managers who have performed badly, and transfer their assets to managers who have recently beaten the market. When that happens the new managers get cash, and they are likely to use it to buy their favourite shares — by definition, those that have recently performed well. This may exacerbate the momentum effect.
In turn, this may explain why the average manager does not outperform the market, even though apparently exploitable anomalies exist. Professional fund managers have their favourites; they just hang on to them for too long.
Click here to subscribe to The Economist
This article was from The Economist and was legally licensed through the NewsCred publisher network.