Thursday April 17 2008
“Rush of hormones may be behind credit crunch,” said The Times today. They reported that a study has found that traders make greater profits on days when their testosterone levels are high. The Guardian also covers the story, saying that researchers measured levels of cortisol and testosterone in 17 city traders. Testosterone levels in the morning could predict the success of the trader that day, while cortisol levels increased the more volatile the market was. It says that the researchers have a theory that “men gain a testosterone boost when they win” giving them an advantage in the next competition, and this recurs with each bout. Too much testosterone can make men take irrational risks however, and this can burst the bubble and the market crashes.
This small observational study did not look at the changes over time, so it is not possible to test any theories about cause and effect and how it happens. There are several questions left unanswered by this study including; whether the traders with higher levels of testosterone generated greater profits in the way described, whether being successful generated higher levels of testosterone rather than vice versa, or if testosterone is involved in a causal way at all.
There are likely to be many causes for the credit crunch, it seems implausible to attribute such a complex phenomenon to hormones.
Where did the story come from?
Doctor John Coates and a colleague from the Department of Physiology, Development and Neuroscience at the University of Cambridge carried out the research. Sources of funding are not declared. The study was published in the (peer-reviewed) medical journal: Proceedings of the National Academy of Sciences of the United States of America.
What kind of scientific study was this?
In this cross-sectional observational study, the researchers recruited 17 male traders from a mid-sized trading floor in the City of London. The volunteers all worked on the same floor of about 260 traders, the majority of which were male (four were female). They ranged in age from 18 to 38 years of age. There was no control group. The traders were not paid to take part and volunteered after learning of the study through flyers distributed on their floor. The flier invited anyone interested to a one-hour talk explaining the project. They were told that they would receive the results and the findings of the study. An initial questionnaire revealed that none of the subjects were taking medication that could have affected their hormone levels, they were all non-smokers, and none drank more than 1-2 cups of tea or coffee a day. It is unclear what the pattern of coffee and tea consumption was and how this was related to hormone levels.
On a normal business day, the traders sit in front of a bank of computer screens which show live prices of currency, commodity, bond, and stock index futures. They also have live news feeds, a risk-management system and an intercom relaying a commentary from an in-house economist. According to their level of experience, each individual traded in figures ranging from £100,000 to £500,000,000.
The traders were followed for eight consecutive business days. At 11am and 4pm each day, (before and after the bulk of the days’ trading) the researchers took a 3ml sample of saliva to measure the levels of the hormones testosterone and cortisol. About half the volunteers needed to chew gum to stimulate the production of saliva.
At the same time as the saliva collection, the traders recorded their profit and loss in the computerised risk management system and this data was used to calculate the average profit and loss for the day. This was combined with the official end of day figures for each trader, collected from the brokerage firms.
The volunteers also filled out a questionnaire on what they ate and drank throughout the day, and anything else that might influence their hormone levels. There is a natural variation in the testosterone and cortisol levels of healthy people throughout the day and these hormones are influenced by food and drink intake.
What were the results of the study?
The researchers report, “a trader’s morning testosterone level predicts his day’s profitability”. A plot is given for the average profit and loss of the 17 traders compared to their 11am testosterone levels. The researchers say that 14 of the 17 traders had higher profit and loss on high days than they did on low testosterone days. The remaining three subjects had negligible differences.
The researchers also looked at the data the alternative way around and found that daily testosterone (the average of the 11am and 4pm samples) was significantly higher on days when traders made more than their one-month daily average than on other days.
No relationship was found between the level of cortisol and the level of profit and loss recorded by the traders.
What interpretations did the researchers draw from these results?
The researchers conclude that their results “suggest that higher testosterone may contribute to economic return, whereas cortisol is increased by risk”.
They propose a theory explaining how the rise and fall of testosterone and cortisol affects people’s thinking and behaviour. This theory revolves around the idea that, should levels of hormones remain acutely elevated, or increase as the financial market’s volatility rises, this could shift risk preferences and even affect a trader’s ability to engage in rational choice.
What does the NHS Knowledge Service make of this study?
This is an observational study that documented variations in levels of two hormones and the profitability of traders at a trading floor in the City. The study had the advantage of using objective measures for the hormone testing and daily and historic profit and loss records. However, there are several limitations to this type of study:
- It only sampled the volunteers for eight days. This, as well as the small size of the study, reduces the amount of data that the researchers were able to collect and therefore the confidence in the result.
- The researchers admit that it had the further drawback of being conducted during what turned out to be a period of low volatility. This may have reduced the range and magnitude of the observed results thereby reducing the chance of finding a positive result.
- There is a strong possibility that the variability in hormone levels are tracking some other, unconsidered and unmeasured, factor that is related to the financial success of the traders. Though the researchers did try to account for some of these potential “confounders”, such as coffee intake and major events in the personal lives of the traders, it is not clear how the answers were fed into the statistical analysis. There were also dietary and sleep factors that were unexplored.
From the results of this small study, it is not possible to say that the testosterone or cortisol levels in individuals have a meaningful effect on financial markets.
Sir Muir Gray adds...
Many women will say this confirms what they have observed for years.