Undercover Economist

Why high-frequency traders are like rutting stags

“They were displaying classic symptoms of mania. They were overconfident, they had racing thoughts, they had diminished need for sleep and heightened sexual appetite.”

John Coates, a former Wall Street trading floor manager, was describing to me not drug addicts or rutting stags, but the male traders he had supervised during the dotcom bubble.

The similarities are not just skin deep. Successful traders and dominant stags are indeed high on something: testosterone. Spikes in testosterone levels are both the cause and the consequence of a profitable day on the trading floor. After a good day, traders find their systems flooded with testosterone, which encourages them to take more risks the following day and, up to a point, to make more money.

The same testosterone surges and streaks of success and failure can be seen in bulls, stags and other sexually competing male mammals. Female traders – who remain rare – don’t act in the same way.

Coates – a research fellow at Cambridge university who quit Wall Street to study the boundaries of economics and neuroscience – made a splash last year by publishing research (with the neuroscientist Joe Herbert) showing all this. Now, in a new paper with the endocrinologist Mark Gurnell and the economist Aldo Rustichini, Coates has been asking whether traders’ behaviour is influenced by high levels of testosterone (and other “androgenic” steroids) they may have been exposed to in the womb.

This is not an outlandish hypothesis, given what endocrinologists already know: high levels of pre-natal testosterone seem to be correlated with confidence, a tolerance for risk and quick reaction times, as well as sporting prowess. So it is not unreasonable that high-frequency traders, who take billion-dollar positions for minutes or even seconds, might benefit from having developed in a conducive womb.

One might ask how Coates, studying the behaviour of traders who were typically in their twenties, could know what had happened to them in the womb. In fact, there is a convenient and widely used proxy: the ratio of the index finger to the fourth finger – or the 2D:4D ratio. Low 2D:4D (a relatively long fourth finger) is evidence of high exposure to testosterone in the womb.

Coates, Gurnell and Rustichini found what they were expecting: that high-testosterone foetuses grew up to be excellent high-frequency traders. What was surprising was the huge size of the effect. Traders with a low 2D:4D ratio made six times as much money as those with high 2D:4D ratios. In an environment when the best traders earned more than £4m a year, this is hardly a trivial discovery.

The high-testosterone advantage seems to come from two sources. First, pre-natal testosterone seems to shape brains with quicker reactions and a greater ability to concentrate. Coates and his colleagues found that low 2D:4D traders did particularly well in volatile markets, when speed of action was paramount. They weren’t just “better”, they were better in a way that gave them an edge in frenzied times. In contrast, experienced traders have a more generalised advantage: they make more money than inexperienced traders in quiet times as well as volatile ones.

Second, pre-natal testosterone amplifies the “rutting stag” behaviour Coates and Herbert had already discovered. It seems to pave the way for a more intense response to later surges in testosterone: once the low 2D:4D traders get on a roll, they really start winning. By the same token, losses are also self-perpetuating for such traders, because they drain away testosterone and with it, the willingness to take risks.

Also published at ft.com.