The gender difference in risk preferences may be bad news for male investors.Federico Caputo/Getty Images/iStockphoto
In Japan, women typically handle the family finances. I'm starting to think that this is a good idea.
Economists have known for a long time that men, on average, are more risk tolerant than women. In 2009, Rachel Croson of the University of Texas-Dallas and Uri Gneezy of the University of California-San Diego reviewed the literature on the topic, and found that this gender difference pops up again and again, everywhere we look. When researchers give experimental subjects real opportunities for risk taking, men tend to be less cautious. When they ask hypothetical questions about gambling decisions, men tend to take more risks. When they look at people's pension-plan allocation decisions, they find that men tend to put more of their money into risky assets. Although there is the occasional study finding specific situations where the difference goes away, the vast weight of evidence says that the gender difference – on average – is real.
But so what? Risk preferences differ from person to person – that isn't exactly news. In a well-functioning financial market, risk should be correlated with expected return – on average, more risk-tolerant people will make more money in the long run. So there's no reason we should automatically regard this gender difference as favouring female investors over male ones.
The problem is that risk-taking behaviour doesn't always come from the rational calculations that economists typically envision. In standard economics, risk aversion comes from the simple fact that money matters more to you the less you have of it. Winning $50,000 on a coin flip would be a great windfall for the average person, but losing $50,000 would be a disaster. So the average person is highly unlikely to bet $50,000 on a coin flip. It's only rational.
But in the real world, there are undoubtedly more reasons why people avoid risk. The shame of failure and the fear of regret are undoubtedly powerful psychological deterrents to risk taking. Overconfidence in one's own abilities or excessive optimism about one's own chances can make people more tolerant of risk. The problem with these emotional causes of risk tolerance is that unlike rational calculation, they can lead people to take risk with no compensating return.
The gender difference in risk preferences may therefore be bad news for male investors. Indeed, in 2001, U.S. finance researchers Brad Barber and Terry Odean looked at individual investors' brokerage accounts. They found that men tend to trade much more than women, so that their returns are eaten up by trading costs. On average, women had better investment performance even though they took fewer risks. Fidelity, looking at its own far bigger data sets, has found that women diversify more than men, trade less and earn the same average returns as men, all with lower risk. This doesn't mean that women are categorically better investors than men; each might have skills that the other could benefit from.
So maybe the Japanese way is the right way. Let Mom manage the family finances instead of Dad, and the family might have a more secure financial future.
But all this raises another interesting question: Is excessive male risk taking the product of the testosterone, or is some of it caused by social constructs and gender roles?
Enter Francesco D'Acunto, a finance researcher at University of California-Berkeley's Haas School of Business. In a new experiment, Prof. D'Acunto tests whether risk-taking behaviour can be changed by giving people cues that make them focus on their masculine identity. His results suggest that social gender roles have a big effect on investment behaviour.
In one experiment, Prof. D'Acunto simply had subjects read about "masculinity," and tested their risk tolerance before and after with a gambling-type task. He found that just this simple reading assignment caused men, on average, to become substantially more risk tolerant than they had been before. But women were unchanged.
In other experiments, Prof. D'Acunto had subjects think about other things commonly associated with masculine identity – for example, in one case he had them recall a time when they had power or control over others. The results were similar – priming the pump of masculine identity caused men, but not women, to start taking more risks.
Prof. D'Acunto found that not only did men take more risks, they became more optimistic about purely random outcomes. In other words, there is some indication that the increase in risk tolerance came from a distortion of rationality. Men, when they concentrate more on masculinity, might simply become overconfident.
Now, Prof. D'Acunto's experiment is just one study. It needs to be replicated before we can regard it as scientifically sound. But the implications of this finding would be interesting. It would mean that trading firms might want to avoid a "bro" culture. It would mean that individual investors might want to be wary of financial media websites that promote a hyper-masculine image.
And of course, it would mean that trading firms might want to hire more female traders – something they should probably do anyway.