How do hormones such as testosterone and cortisol affect our ability to work, rest and play? A stock-market trader turned neuroscientist has the answer.
One day in 2000, a Wall Street trader called John Coates went to visit Linda Wilbrecht, a neuroscience PhD student in the Laboratory of Animal Behaviour at The Rockefeller University on the Upper East Side of Manhattan. Wilbrecht and Coates had met casually six months earlier on a flight from Paris to New York. The two North Americans got talking and soon found they had something in common: they had both studied at Oxbridge. "My life is a series of accidents,"
Coates tells Wired. "After studying philosophy, politics and economics [focusing on philosophy] I got a full scholarship to Cambridge. But I was put in the economics department by mistake – and if I turned the scholarship down, I'd have to reapply the following year. Who turns down Cambridge? After my PhD, I was offered a job at Goldman Sachs because a manager wanted someone to discuss philosophy with. Then I met Linda and she invited me to visit her lab. Oddly, I had no interest whatsoever in the brain at the time."
Coates was unprepared for what he saw. "Within 20 minutes of being in the lab I was hooked," he says. "Hearing the word
'plasticity' applied to the brain horrified and fascinated me."
After his first visit, Coates started to attend the neuroscience lecture series at the Caspary Auditorium on the Rockefeller campus.
There he saw neuroscientist Bruce McEwen, a pioneer in the study of the effects of hormones on the brain, and primatologist Robert Salpolsky reporting on his work with baboons in Kenya. When the markets were slow, Coates would nip away from his desk and walk to the neuroscience laboratories, where Wilbrecht would sometimes give him small jobs, such as pipetting samples, and show him how to measure testosterone in serum and how to stain new cells.
Gradually, Coates got hooked on experimental science. All his life he had studied purely theoretical subjects such as philosophy and economic theory. He never enjoyed it. Miring himself in a mountain of data, watching the signal emerge from the noise, be it in the laboratory or on the trading floor, was a challenge that he seemed to embrace.
Coates had been working as a trader since 1989. During the 90s, he had been at the centre of the dotcom boom. "It was like watching
A Midsummer Night's Dream," he says. Traders were euphoric and investors delusional. Few heeded Federal Reserve chairman Alan Greenspan's prescient warning in 1996: "How do we know when irrational exuberance has unduly escalated asset values, which then become subject to unexpected and prolonged contractions, as they have in Japan over the past decade?" Coates was not even interested in high-tech IPOs, so he was very much an outside observer. "I would see people get on a winning streak on the trading floor and go lunatic. It happened to me as well. For weeks, even months, you feel like the hero of the floor. Every trader who has made money knows what this feels like.
You think you're infallible."
Coates felt that while Wall Street was questioning the nature of irrational exuberance, the scientists at Rockefeller were getting close to explaining it. "Economists assumed that all behaviour was conscious and rational," he says. "They were ignoring the fact that signals from the body, both chemical and electrical, affect how we take financial risks."
Canadian-born John Coates is partly bald, with the physique of someone who swims regularly. When Coates talks about his research his manner is passionate and persuasive, his speech interspersed with hearty laughs. He lives in London and works at Cambridge University as a research fellow in finance and neuroscience. In the summer, he leaves Europe to spend holidays with his family in a cabin on a lake in Canada, where only a single phone-line snaking through the woods keeps him in touch with the rest of the world.
Every day he writes for hours on end, like a modern-day Henry David Thoreau. "Difference is, Thoreau went to the pub every night,"
Coates tells me. "Here there's nothing for miles in every direction."
Coates conducted his first study in the financial district of London. At a trading floor in the City, he recruited 17 traders from a mid-sized firm and followed them for two weeks. The subjects were high-frequency traders who held their bets – which could go up to two billion dollars (£1.25bn) – only for a short period of time, sometimes just for a matter of seconds. Every trader was sitting in front of six or seven computer screens displaying live information on currencies, commodities, bonds, stock-index futures, live news-feeds and ongoing commentary about daily economic statistics by a resident economist. Twice a day the traders would register their profit and loss statement and give saliva samples, from which cortisol – the hormone we release when we are stressed – and testosterone were measured. "I was just looking for preliminary data that would convince me that I wasn't wasting my time," says Coates. "I wasn't fishing for patterns, it was hypothesis-driven.
"This kind of field work didn't exist in finance."
The results were published in a 2008 report in the
Proceedings of the National Academy of Sciences of the United States of America. Coates found that, on days when traders made an above-average profit, their testosterone levels went up.
Most surprisingly, the testosterone levels in the morning predicted how much money the traders would make that day: high levels forecast high earnings. At the same time, the traders' cortisol was unaffected by how much money they lost. Rather, cortisol levels were sensitive to the volatility in the market, which is a measure of risk and uncertainty. "Cortisol is likely, therefore, to rise in a market crash and, by increasing risk aversion, to exaggerate the market's downward movement," the report states. "Testosterone, on the other hand, is likely to rise in a bubble and, by increasing risk-taking, to exaggerate the market's upward movement. These steroid feedback loops may help to explain why people caught up in bubbles and crashes often find it difficult to make rational choices."
Coates first learned of steroid feedback loops during his regular visits to Rockefeller University. The testosterone feedback loop is known as the winner effect. The winner effect had been observed in nature for many different species, from cichlid fish to rhesus monkeys, and its physiology is well understood. When two animals square off in anticipation of a fight, they experience a rise in testosterone levels. This self-doping mechanism prepares the animal for competition, increasing the blood's capacity to carry oxygen, quickening the speed of reactions, and, via its effect on the brain, increasing fearlessness and appetite for risk.
In the aftermath, winners can emerge with a tenfold increase in the amount of testosterone circulating in their bodies, whereas losers' testosterone levels can be suppressed by the same order of magnitude.
This doping effect can sometimes last for months. Nature primes winners to keep winning and losers to keep losing. The winner effect is not exclusive to the savannah. Sports scientists have observed it in tennis, rugby, football and even chess. Winning athletes experience a post-game spike in testosterone. Even the fans of winning teams vicariously experience a testosterone surge. "Many economists dispute it," Coates says. "They seem completely unaware of the animal studies. They say that anyone who believes in the winner effect is suffering a cognitive bias, but they never mention what happens our physiology. We do have a self-doping mechanism lurking in our bodies and any sports scientist will tell you that rising levels of testosterone contribute to victory."
In a recent study, Coates analysed a database of 623,000 professional tennis matches, narrowing the sample down to matches between tennis players who were as closely matched as possible in rank and to matches that went to a tie break with decided by the narrowest possible margins – two points. "These matches were played so closely that they could have been won by a gust of wind, a cough in the stand," Coates says. He found that, for the men, the winners of the first set had a 60 per cent chance of winning the match. The statistics didn't show any correlation for the women players. "The protocol was a very clean test of a pure winner effect. That we didn't see it for women indicates it is probably a testosterone effect. The data was eye-popping."
Women produce, on average, about ten per cent of the amount of testosterone that men generate. According to Coates, they may therefore be less prone to excessive risks driven by the winner effect; their stress response may also be less sensitive to risk-taking failures.
During the dotcom boom, it always surprised Coates that the women traders seemed to be relatively immune to the euphoria that engulfed most male traders at the time.
Women seemed to know that a storm was coming. When it comes to the financial markets, Coates says, men are more hormonal than women. Male physiology makes men more attuned to high-frequency risk-taking. "Our latest studies suggest that women are not more risk averse than men," says Coates. "They merely prefer to have more time and information before they take risks."
This doesn't imply smaller profits – quite the opposite, in fact.
Studies of gender differences in investment behaviour consistently show that, in the long term, female investors consistently outperform their male counterparts. This is not, Coates stresses, an endorsement of one sex over another. "It's not that one group is better than the other," says Coates. "They're different. It's just that by diversifying the biology of the trading floor you would counterbalance the extreme tendencies."
In his second year as a junior trader at Goldman Sachs in Toronto, Coates had a bad losing streak.
Quebec, Canada's largest province, was having an independence referendum. Canadian bonds were devaluing, so Coates bought Canadian bonds and sold US bonds against them. He was confident that he was making the right trade. "I lost $22 million (£14 million), which was a hell of a lot for a junior trader," he says.
He stopped sleeping. He began to obsess about market rumours. New York started to look like Gotham City. "I became absolutely insane. My world fell apart."
The stress response is a problem that preoccupies Coates more than the problem of irrational exuberance. He was captivated, 13 years ago, by the ideas of Robert Sapolsky, a primatologist who first made the link between low social status and high levels of stress. Sapolsky had studied neuroendocrinology under Bruce McEwen at Rockefeller and, after that, spent nearly 30 years studying the social behaviour of baboon troops in Kenya. In 2000, while still on Wall Street, Coates heard a lecture by Sapolsky on his work with the baboons and, he recalls, left the room "knowing that was something I was going to end up working on". He wasn't even contemplating leaving Wall Street at that point, but Sapolsky's description of young baboons getting bullied by their seniors sounded too familiar. "During a crisis, middle managers at the bank behave like a stressed-out troop of primates," says Coates. "Bosses bully their juniors just like adult male baboons bite their juniors to relieve their stress. Some bosses think that they are not doing their job unless everyone is in a state of constant panic."
When people feel stressed, a region of the brain triggers the release of cortisol, a hormone that puts the body in heightened state of alert. Cortisol mobilises nutrients into your bloodstream, rapidly increases its levels of glucose, providing muscles with a burst of energy, it shuts down all nonessential bodily processes, such as digestion, the reproductive system and the production of testosterone. An acute cortisol response to a challenge is fundamental to survival. However, the stress response was designed to be short-lived. The problem arises when stress does not go away and cortisol builds up in your body over months and years. In the 70s, neuroscientists found that if you repeatedly expose lab mice to uncontrollable stressors, such as electric shocks, after a while they will fail to leave their cages even if the door is left open.
Biologists call this state "learned helplessness". It's a stark illustration of the extent to which high levels of cortisol can dramatically change our brain and subsequently our behaviour: you feel you no longer have control over your own fate, even if a way out is right in front of you. You become risk-averse and despondent. You give up. "It's so dysfunctional it seems like a flaw in evolution," Coates says.
For him, the tragedy is that we become used to thinking that stress is all in our head, when in fact, it might be our body. "In therapy, we are talked into seeing stress as a challenge, not a threat. I'm not convinced talking gets at the physiology. If the demands made on you are greater than your resources – you are going to get stressed and no amount of talk therapy is going to change that equation."
In the same way that testosterone can drive financial bubbles, Coates believes that high levels of cortisol on the trading floor can amplify periods of irrational pessimism. Coates is currently running experiments to understand if traders too can be afflicted by learned helplessness during a crisis and how cortisol can affect financial decisions. At Addenbrooke's Hospital in Cambridge, he recruited volunteers and, by administering tablets, artificially raised their cortisol levels by about 55 per cent over a period of ten days – the state of someone who is moderately stressed. He then had them play a game that gauged their risk preferences before and after their cortisol levels were raised. "It absolutely nuked their appetite for risk,"
Coates says. "By manipulating the level of cortisol in their bodies, we can almost tune how risk averse people are. Stress hormones act like a dial. In economics, there's an unstated assumption that our risk preferences are innate. This suggests the opposite. Our risk preferences fluctuate with the state of the body."
At the core of Coates's hypothesis – that testosterone is the molecule of irrational exuberance and cortisol the molecule of irrational pessimism – lies a deeper, more challenging idea: that a lot of the decisions that we assume are conscious processes are actually strongly manipulated by signals from our bodies. This is an idea that has been explored in detail by sports physiologist Christian Cook, who is considered a pioneer in the study of hormones in athletic performance. In the early stages of his career, he studied the effect of human encroachment in the stress levels of exotic and domestic animals, from sheep to polar bears.
He then worked as a physiologist for the New Zealand rugby team and America's Cup Yachting, where he looked at ways for athletes to optimise their training and studied the role of natural hormones in athletes' recovery from intense exercise. "In the early noughties we still had the stereotypical vision of the balloon-muscled man introducing 100 times his normal physiological levels of testosterone into his body," he says. "It biased our thinking and people assumed testosterone only makes muscle."
This picture, however, was too simplistic. "Obviously, if you put that much testosterone into your body your muscles will grow. But within a normal biological range, its main direct role is behavioural.
Testosterone gives you more confidence and motivation and that makes you work harder, which indirectly influences muscle growth."
The role of testosterone, he found, was not so much that it gives you something extra for free, but that it allows you to express more of what you're capable of. If you have higher levels of testosterone you can jump higher and pull more weight not necessarily because you have more muscle power, but because your hormonal levels allow you to express that power more freely. Cook speculates that this is the reason some athletes still use steroids: they have become psychologically addicted to the feeling of confidence they get from the artificial testosterone.
These findings neatly corroborate the animal studies on the winner effect. But when it comes to humans, Cook makes another, subtler point. Testosterone is not related to winning. It's related to our perception of winning. In one of his studies, Cook took a group of highly trained rugby players and gave the coach a shortlist of standard phrases from which to choose, such as "You did that poorly, why couldn't you do that right?" or "Well done, that's how you do it, you performed really well." Half the group received a positive feedback, the coach positively reinforcing things that were done well, whereas the other half listened to the coach negatively reinforcing things that had been done badly. Players who received positive feedback had a 30 percent higher testosterone response than the players who'd received negative feedback. This effect lasted several days until the next match, when the players who had had positive feedback performed better than the players who got criticised. "What we found in rugby is that you could be recovered physically but not be ready to compete," says Cook. "Recovery is far more than the ability to physically perform. It's also the ability to mentally make yourself perform." This optimal state – what Cook calls "readiness to compete" – is dependent on the player's hormonal balance and appropriate recovery. Training programmes are designed to be intermittently stressful. Any kind of physical exertion is accompanied by a release of cortisol in the body, which allows it to more quickly mobilise the necessary nutrients to the working muscle tissues. Once that stress response is turned off, the body turns on a testosterone response to cope with the stress and rebuild energy stores. The body adapts to the physical stress through a rhythmic alteration between a period of physical loading followed by a period of recovery and adaptation. That adaptation increases an individual's capacity to handle stress. Elite athletes have an unusual ability to cope with challenges: their initial stress response is strong, but abates quickly.
Cook is keen to stress that testosterone and cortisol are part of a complex system that dictates how we respond to stress. This stress can be alleviated – by exercise, for example – but it is far from a panacea. "Exercise promotes physiological resilience, but it's pointless to say, 'Go exercise'," he says. "People's bodies respond differently. A lot of our training is wrong because it enforces the same programme for everyone. It's what I call
'survival of the fittest', but it's not necessarily survival of the best."
During his first experiment with financial traders, in addition to the physiological data that he collected, Coates would hand them multiple-choice questionnaires about their day at the trading floor, their health and their concerns. They turned out to be useless. "One trader didn't even use the multiple-choice answers available. He just wrote down 'euphoric, stressed, fucked off, despondent'," Coates recalls. "That page alone was worth keeping as a testament to the inability of questionnaires to get at anything precise." Traders could not interpret or explain their own financial bets during the day. Their cortisol levels, on the other hand, mimicked with astonishing precision the degree of volatility in the markets. "Psychologists often don't think about the brain and neuroscientists often don't think about the rest of the body,"
Bruce McEwen says. "People have to realise that hormones are not only controlled by the brain, they act on the brain. They affect all sorts of behaviours – certainly trading."
Coates sees it as his mission to spread the notion that in order to understand our own behaviour we have to understand our own biology. Economists, Coates writes, tend to view the assessment of financial risk as a purely intellectual affair, but we prepare for financial risk physically by bringing forth a biological reaction.
When this happens to traders, they come to suffer an irrational exuberance or pessimism. "As a result," he writes in his book,
The Hour Between Dog and Wolf, "traders are walking time-bombs, and banks invariably light the fuse, dangling huge risk limits and bonus payments that have exceeded $100 million."
The elegance of Coates's pioneering studies is that he was able to cut through the messy nature of our physiology and provide clear relationships between objective and yet disparate measures from the universe of high finance, and our interior hormonal world; how market volatility is connected to cortisol and how profit and loss statements are connected to testosterone. "John's work most closely matches my work with wild baboons in Kenya," Robert Sapolsky says. "It's based on subjects in the real world rather than some artificial experimental setting. His interpretations are smart, subtle and appropriately careful. What I like about his work is that he is showing that, yes, external events such as absolute numbers in stock-market outcomes can impact physiology, but that the meaning of those events do so more powerfully."
Coates is currently running eight experiments, spanning a range of topics from difference in risk-taking between men and women, to the study of stress in the workplace. He is adamant that a deeper understanding of our physiology should inform not just how we manage our trading floors, but also how we design all workplaces.
His understanding of physiology is why he contends that trading floors should be spaces with biological diversity, populated by both young and old, male and female – and why he also believes that bank management should extend its period of assessment from one to five years.
The fundamental lesson is that there is no such thing as a pure human intellect that is not tainted by the hormonal waves and other signals that arise from the rest of the body. Traders on Wall Street, Olympic athletes and baboons in Kenya are just the extreme examples of the one biology that pervades us all. "As a species, we share the same biology with other animals, but express it uniquely," Sapolsky said during a lecture at Stanford University in 2010. "We can have two humans sitting at a table doing nothing more physically taxing than one of them moving a little piece of wood on the table. And if it happens that these two individuals are at a chess tournament, then they are able to keep [up] a blood pressure for six hours [at a level] that you normally only see in a marathon runner, while doing nothing more than thinking. And this is outrageous because when you look at these chess Grand Masters who've just taken down an opponent, they will have the exact same physiology of some wild baboon in the savannah who has just ripped open the stomach of his worst rival."
This article was originally published by WIRED UK
