sexta-feira, junho 08, 2012

Not so expert

Buttonwood, The Economist

ASK enough people for advice, they say, and you will eventually find someone who will tell you what you want to hear. But the need for advice burns so strongly that people become blind to its quality. There is a remarkable tendency to trust experts, even when there is little evidence of their forecasting powers. In his book “Expert Political Judgment”, Philip Tetlock shows that political forecasters are worse than crude algorithms at predicting events. The more prominent the expert (ie, the more they were quoted by the news media), the worse their records tended to be. There is also an inverse relationship between the confidence of the individual forecaster and the accuracy of their predictions.

The remarkable tendency for individuals to rely on expert advice, even when the advice clearly has no useful component, was neatly illustrated in a recent academic paper* about an Asian experiment. Undergraduates in Thailand and Singapore were asked to place bets on five rounds of coin flips. The participants were told that the coins came from fellow students; that these would be changed during the process; that the coin-flipper would be changed every round; and that the flippers would be participants, not experimenters. Thus there was a high likelihood that the results would be random.

Taped to the desk of each participant were five envelopes, each predicting the outcome of the successive flips. Participants could pay to see the predictions in advance, but they saw them free after the coin toss had occurred.

When the initial prediction turned out to be correct, students were more willing to pay to see the next forecast. This tendency increased after two, three and four successful predictions. Furthermore, those who paid in advance for predictions placed bigger bets on subsequent coin tosses than those who did not.

Paying for financial advice might not seem quite as bizarre as paying for coin-toss predictions, but there are some similarities. Nobody can reliably forecast the short-term outlook for economies or stockmarkets; Warren Buffett, the world’s most successful long-term investor, thinks it is not worth trying to do so. But plenty of economists and strategists earn a good living doing just that. The average active-fund manager fails to beat the stockmarket index; no reliable way has been found for selecting above-average managers in advance. Yet investors are still willing to pay for the services of active managers.

The sheer complexity of modern financial markets and the torrent of information that is published each day are a boon to the providers of financial advice. Investors may feel that they simply do not have the time to analyse all the data, and they therefore need to rely on the advice of professionals. This is true even if they think the markets are a “rigged game” played for the benefit of insiders; it still makes sense for them to pay for an insider’s view.

There may be another, psychological, reason why investors want to pay for advice: the avoidance of regret. If you choose to put all your money into technology stocks on the back of your own research, and such stocks collapse, you only have yourself to blame. But if you have listened to the advice of an expert, then the decision is not your fault.

Some financial advice may be extremely useful. Many advisers steered their clients away from Bernie Madoff’s fraudulent funds. Investors also need to be made aware of the benefits of diversification and of the effect on their portfolios of tax rules and regulations. There is also evidence that market valuations revert to the mean over the long term, so pointing out when markets look historically cheap or dear can help.

The problem for the industry is that such advice will not be needed very often, and that limits the potential fees. So instead investors are bombarded with endless research on why stock A is better than stock B, why one currency is bound to outperform another and so on. Clients end up churning their portfolios, even though the costs erode their returns.

Perhaps the financial-advice industry survives because the idea that the future is unknowable is just unsatisfying. Some forecast—any forecast—is therefore comforting. Mr Tetlock suggests that “we believe in experts in the same way that our ancestors believe in oracles; we want to believe in a controllable world and we have a flawed understanding of the laws of chance.”

* “Why Do People Pay for Useless Advice? Implications of Gambler’s and Hot-Hand Fallacies in False-Expert Setting”, by Nattavudh Powdthavee and Yohanes Riyanto, Institute for the Study of Labour, May 2012.

3 comentários:

samuel disse...

Piada corrente no meio de prestação de serviços: Consultor é aquele cara que pede o seu relógio emprestado para ver as horas e se esquece de devolvê-lo!

Anônimo disse...

http://www.dilbert.com/dyn/str_strip/000000000/00000000/0000000/000000/40000/9000/200/49206/49206.strip.print.gif

Gustavo Sauer disse...

Irvin Fisher, o primeiro economista famoso dos EUA, e segundo M.Friedman, o melhor economista do século, ficou famoso com suas previsões catastróficas a respeito do crash de 29. Não conseguiu prever o crash e por meses continuou dizendo que era apenas uma queda momentânea e que a recuperação viria logo.