Finance: How to profit from others' credulity

In his novel A Tenured Professor, the distinguished economist JK Galbraith describes how a Harvard academic develops a technique called the Index of Irrational Expectations to profit from the stock markets.

The idea behind this economic model is that investors - both institutions and individuals - have a tendency to get carried away by optimism. In real life, this kind of behaviour has been termed the psychology of crowds, and it has been called on to explain all kinds of boom that lead to bust, from the South Sea Bubble to the US Savings and Loans crisis.

Essentially, many investors join the buying spree because they see that a particular sector or stock has rewarded others handsomely and they cannot bear to miss out themselves. The trick, of course, is knowing when to sell. Those that get in early and pull out ahead of the pack tend to do well, whereas late arrivals are caught out, as anybody who suffered when the likes of Polly Peck and Coloroll plummeted to earth in Britain in the late Eighties and early Nineties can testify.

The protagonist of Galbraith's book took the view that just because many people would lose when reality caught up with a stock, did not mean that nobody could win. Accordingly, he devised a strategy of using economic analysis to work out when euphoria was dangerously high and then took a "short position" in the shares concerned by borrowing to buy stock at current prices - and making a profit by replacing it when the price fell.

Much the same sort of thinking appears to be behind recent research by two Chicago business school academics and a former colleague now at Harvard. Their paper, "A Model of Investor Sentiment" argues that there is evidence that in some cases investors react too little to such news as higher-than- expected earnings or a large acquisition; in others, they react too much.

The authors - Nicholas Barberis and Robert Vishny, respectively assistant professor of finance and finance professor at Chicago, and Andrei Shleifer, a former Chicago faculty member now at Harvard - propose a model based on ideas from psychology to explain how investors make mistakes when they process new information.

This thinking challenges - in much the same way as Galbraith's Prof Marvin did - the efficient markets theory, or the idea that information is immediately and accurately reflected in the share price. This suggests that in a variety of markets, sophisticated investors can earn superior returns by taking account of this under- and overreaction to taking extra risks.

Previous research, including that by another Chicago business school academic, Richard Thaler, had indicated that investors react too sharply to company news.

As Barberis explains in an article in the Chicago Graduate School of Business research publication Capital Ideas, "if an investor ranks thousands of stocks based on how well they did over the past three to five years, he or she can then make a category for the biggest 'losers', the stocks that performed badly, and another for the biggest 'winners'. What you will find is that the group of the biggest losers will do well on average over the next few years. So it is a good strategy to buy these previous losers or undervalued stocks."

He and his fellow researchers argue that one of the ways in which these findings can be explained by investor overreaction is that if a company consistently reports earnings that beat expectations, shareholders can become excessively optimistic about the company's prospects and so push the stock price to unnaturally high levels. Later, the investors realise that they were unduly optimistic and the price corrects itself downwards.

Similarly, loser stocks may simply be stocks that investors have become excessively pessimistic about. As the misperception is corrected, the stocks earn high returns.

Under-reaction, according to Barberis and his colleagues, can occur, for example, when a company announces good earnings, but the stock price does not initially go as high as it should. Over the next six months, though, this "mistake" is gradually corrected as the price drifts upwards. Anybody buying stock immediately after the announcement will enjoy higher returns.

The fact that the same sort of thing happens with bad news leads the team to question the efficiency of financial markets. "Our idea is that these market anomalies - under-reaction and overreaction - are the results of investors' mistakes," says Vishny. "In this paper, we present a model of investor sentiment - that is, of how investors form beliefs - that is consistent with the empirical findings."

This way of looking at investor behaviour fits in with two important psychological theories: the "representative heuristic", and "conservatism". The first refers to the fact that people tend to see patterns in random sequences - ie investors may see patterns in financial data that are not really there. The second is that people are slow to change their opinions in the face of new evidence - for example, investors who see earnings going up think they have spotted a trend, and only gradually update their views as a result of new evidence.

The authors accept that their paper is not welcomed by those in academic finance who follow the efficient market theory and explain that higher returns simply reflect higher risks.

"We are taking the alternative approach, saying that perhaps there isn't any increased risk, and the under-reaction and overreaction phenomena can be explained by genuine mistakes that people are making," says Barberis.