Testosterone had no effect on women's risk taking behaviour. Credit: Punchstock

Women given testosterone for a month were no more likely than women not receiving the hormone to engage in risky financial decisions, according to researchers in Sweden. The findings could suggest that women are a safer pair of hands on the stock-market trading floor than men — or throw into doubt earlier findings about the effect of the hormone on men.

A spate of recent studies have found correlations between testosterone levels and risky behaviour in men, including one that found that male securities traders with more testosterone in their saliva made riskier financial decisions1.

But now a team led by Magnus Johannesson, an economist at the Stockholm School of Economics, has found no such effects in a group of 200 post-menopausal women. The women were administered testosterone, oestrogen or a placebo for four weeks and asked to play a series of economic games that measure the player's propensity to take risks, their trust and their willingness to share resources. In the 'dictator game', for example, a player can decide how much of a pot of money they will share with a charity and how much to keep for themselves.

The team thought that the testosterone-taking women would behave more like men, giving less to charity and accepting more risk in an investment game. Yet their results, which are published in the Proceedings of the National Academy of Sciences, revealed no meaningful differences between the women who had taken testosterone or oestrogen and the placebo group2. "My assumptions have changed a lot," Johannesson says.

Stabilizing influence

Other studies have identified differences between how men and women make financial decisions, suggesting that exposure to sex hormones during prenatal and adolescent development may matter more than it does later in life.

John Coates, a neuroeconomist at the University of Cambridge, UK, was the lead author of the study of male traders. He points out that a person's sensitivity to circulating testosterone could be a function of the amount of prenatal testosterone they were exposed to. "So increasing testosterone in women might have much less effect than increasing it in men," he says. "The results are roughly what I would expect."

It would be necessary to run an experiment of a similar scale with young men, says Coates, to better understand the extent of a link between hormones and economic decisions. The results of such experiments may also depend "on whether the experiment was conducted in a lab, with low monetary rewards, or in the field, so to speak, with meaningful amounts of money and risk".

Either way, adds Coates, "Women might be a stabilizing influence on the markets."

Johannesson agrees that the impact of sex hormones on financial decision-making is "still an open question". Yet he points out that the statistical power of his study is high because it included so many individuals.

"I'm relatively pessimistic of finding an effect in men," Johannesson says. He writes in the paper that it is possible that previously published links between testosterone and risk-taking are "spurious". Studies that do not find a correlation between sex hormone levels and economic behaviour may simply have a harder time getting published. "Negative correlation results don't get published," he says.