Largely because retail investors have too much information and little time to process it efficiently, they tend to be rather conformist and particularly prone to herd behaviour, says Lin Peng, a professor at the City University of New York.

“Excess co-movements” is the term he uses in his study, Investor Attention, Overconfidence and Category Learning, to describe the unfortunate tendency of investors to purchase stocks based on the overall industry and not the prospects of individual companies. If agriculture is a hot commodity, for instance, most investors will rush to purchase company shares related to the industry, regardless of the company’s financial situation. This follow-the-crowd approach generally leads to below-average returns for many investors, Peng suggests.

The notion that others are buying and making money seems to be a constant lure. In fact, something as simple as a name change can be sufficient to attract unschooled investor attention. For instance, a Purdue University study, entitled Changing Names with Style, documents how mutual funds regularly enjoy a significant jump in contributions simply by changing their names to take advantage of the current hot investment trends.

The year after the name change, such funds earn average cumulative excess inflows of 28% without any increase in performance compared with their pre-name change results, reports professor Michael J. Cooper, the study’s author. The increase in inflows is similar across funds that change their underlying investment style and those that do not, suggesting that investors are “irrationally” influenced by where they perceive others to be having success.

In a previous study, Cooper and his co-authors documented similar bumps in stock price and trading activity of companies that changed their names to include Web-oriented designations such as “.com” and “.net” during the tech bubble.

Is this herding a retail phenome-non, or does it include those that trade for a living? In Herd Behaviour in Financial Markets, Marco Cipriani, a professor at George Washington University, and Antonio Guarino of University College London, studied the behaviour of financial market professionals.

In order to determine what part of a trader’s action is simply following the crowd and how much is based on the trader’s own information, the researchers tested for herding behaviour in a controlled laboratory setting. Among the participants, 28% were traders, 47% were market analysts, 9% were investment managers, 9% were investment bankers and 6% were account managers. Eighty-four per cent were male; 16% were female.

Four sessions were held, with each participant in one session. All participants formed their initial opinion of the subject company and were then placed in either Treatment I or Treatment II. In the first, all participants heard some news affecting the firm’s earnings, but did not know whether it would lead to an upgrade or downgrade. In Treatment II, however, not all participants were given this news and some received incomplete information. In both cases, participants were able to observe the behaviour of other participants.

In Treatment I, the participants had a tendency to go against the market. However, some preferred to abstain from trading altogether, even though they had an information advantage.

Forty-six per cent followed their private view; 20% disregarded the private information and bought or sold based on one of the two signals; 19% disregarded the private information and bought or sold based on both signals, ignoring their private information to follow the herd; and 12% chose not to trade at all.

In Treatment II, herd behaviour increased, but only slightly. In fact, the degree of conformity was almost identical to the results in Treatment I.

But, overall, the level of herding among traders observed in the laboratory is less than what herding theory predicts, the authors note.

This last study suggests less of a propensity to herd by market professionals and even more of a tendency for them to go against the market and adopt a contrarian stance. IE