Testosterone on the trading floor: Peiran Jiao gets the measure of the men behind the markets

Neoclassical economics, says Peiran Jiao, dominated his undergraduate and even his doctoral studies; as a young scholar, he adds, he “didn’t know what behavioural economics was”. Today, however, the assistant professor of finance at Maastricht University School of Business and Economics and member of the SBE Human Decisions and Policy Design research theme is a behavioural economist himself. He’s also one of the authors of a recent high-profile study of financial decision-making that highlights the field’s ability to reveal valuable insights into the human (as opposed to Homo economicus) factor in economics.

Jiao’s paper, “The Bull of Wall Street: Experimental Analysis of Testosterone and Asset Trading“, co-authored with his doctoral supervisor Paul Zak and three fellow PhD students, indicates that testosterone not only drives competitive bidding, thereby producing larger stock market bubbles, but reduces trading performance and increases trader overconfidence. In the male-dominated world of the stock market, could it be the case that the people making split-second multimillion-dollar decisions are dangerously hormonal?

How did your research project on Wall Street trading and testosterone come about?

Peiran Jiao: “It grew out of an interest in neuroeconomics in general; and then we became interested in the effect of testosterone in trading behaviour. The majority of traders on Wall Street are male, and testosterone is apparently one of the most potent hormones that influence male decision-making, especially when it involves risk or competition. This trading environment is highly competitive, and you have to take risks, so it’s the best testbed for the effect of testosterone.

“In the existing literature, there are lots of discussions about how testosterone is correlated with risk-taking behaviour, competition or cognitive ability, but there was very little that had been done to view the causal link between testosterone and trading.”

How did you design the test?

PJ: “The design is a simplified stock trading environment. A very specific design choice was that we told the subjects the fundamental value of the asset, which is not true in the real world. In the real world, they don’t know a stock’s value, but they can calculate some proxies of fundamental value.

“We gave some test subjects testosterone, in the form of a drug called AndroGel, increasing their levels to something that was still within the range of daily variations, so that made the manipulation reasonable. Then we checked whether they exhibited different behaviour than those who received placebos.”

What did you find out?

PJ: “The biggest takeaway is that the people who had been given testosterone generated larger asset price bubbles, which is really bad if it happens in the real world. On the one hand, they were more aggressive in bidding for shares of an asset, and on the other hand, they were not so good at perceiving the declining fundamental value of the asset. In our experiment, those taking testosterone were slower to take the fundamental value into account than the placebo group. This is significant.

“The asset price bubble generated by the testosterone group was not only larger but also lasted longer than that of the control group. We found that the testosterone group had a worse incorporation of fundamental value in almost all aspects.”

Were there any additional findings that surprised you?

PJ: “Yes. AndroGel is a prescription drug, and we learned that a large proportion of Wall Street traders take it, mostly to deal with depression and stress. It’s conceivable that when these traders go to their doctor and say, ‘I feel bad and feel stressed,’ they are given testosterone…which is really bad! So at least one significant implication from our study is that you shouldn’t give testosterone to traders.”

Were these tests done only on men?

PJ: “Yes. The drug is only approved for use on males. We would like to have tested females as well, but we’re not allowed to.”

What would you say to those who suggest that the 2008 financial crash, and indeed problems in the world of finance in general, are created by reckless young men playing with dangerous implements?

PJ “There is an element of truth in that argument. Younger male subjects did really do worse in our experiment. I also have a subsequent study looking at how people make decisions based on their past experience, and there I also found that younger male subjects did worse. They relied too much on their previous gain and loss experience in making new decisions, even where they had better information to inform their new decisions.”

In the world of finance, would a fairer gender balance help?

PJ: “I think there is definitely some truth in this. We could argue that an implication of our testosterone study is that we should have some kind of division of labour between gender on such finance-intensive jobs. I read that in the US Marines’ aircraft carriers, the fighter jet pilots are male, but the fire control officers are female. I wouldn’t say that all fund manager jobs should be done by women, but you could argue that in a team of fund managers, you should make sure that there are both men and women, and also make sure to do something to let these male traders cool down, perhaps by a rotation of management roles.

“Age is also worth looking at, because younger males have higher testosterone than older males, so if testosterone is driving trading behaviour to some extent, then younger male traders will definitely behave differently. Age and experience matter as well.”

But don’t traders need to have that killer instinct?

PJ: “In sport, you wouldn’t want to regulate athletes’ testosterone, because testosterone is good to some extent in situations where you want people to be competitive and risk-taking. But in finance, the case is quite different. They’re trained, they know all the algorithms and formulas, but if they’re driven by hormones, they may still make bad decisions. Neurobiologists explain the effect of hormones as not changing whether people will do something, but that it can affect the probability they will revise their behaviour when the situation requires them to do so.”

Speaking of behaviour, why did you choose to be a behavioural rather than a neoclassical economist?

PJ: “My transition to becoming a behavioural economist happened when I read George Akerlof and Robert Shiller’s book, Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism. Before that, I didn’t know what behavioural economics was, because almost all the courses I took in China as an undergraduate were neoclassical economics, as was the case in the first year or two of courses in my PhD programme in the US. Then my macroeconomics professor told me that we should read Animal Spirits, and it made me really rethink my philosophy.

“It probably wouldn’t have had such a great impact on me if it had been another book of behavioural economics, because Animal Spirits is about behavioural biases that may lead to differences in macroeconomic outcomes. It was amazing to me how these behavioural biases can be so systematic and can lead to even significant differences when it comes to aggregation, because one of the arguments that neoclassical economists had is that when you aggregate these behavioural biases they are smoothed out. The book made really interesting and relevant points about this.

“That made me want to see how these behavioural biases can also be systematic and generate significant aggregate outcomes in finance. So, my testosterone paper is about how these biases induced by testosterone can generate aggregate market outcomes on asset prices… so you see my journey right there!”

With Richard Thaler winning the Nobel Prize last year, is behavioural economics now central to the discipline?

PJ: “We’re slowly moving towards incorporating behavioural economics into the mainstream, but there’s still a long way to go. Thaler’s contribution is mainly to bring people’s attention to the field and to demonstrate that there are many different kinds of behavioural biases that were not taken into account in economic models. Another thing that he has done really well is to generate some meaningful and systematic policy suggestions from these findings.

“But in order to incorporate behavioural economics in mainstream economics models, we need to build systematic models of behaviour. Right now, there are just so many behavioural biases that you don’t know, when you have a new problem at hand, which model you should apply. Classical economics is elegant in doing that; they know exactly what model to use in each situation. Now we’re adding behavioural components to their models, one at a time. But what is the right thing to do? We still don’t have a very good idea of how to make parsimonious adjustments to existing neoclassical economic models so that we can use one consistent behavioural model to study many different behaviours.

“Neoclassical economists are moving in the direction of acceptance. Initially they wouldn’t accept any of the behavioural biases, but nowadays, more and more studies show that certain biases are very consistent, very systematic, in most people, and not just in the students that were traditionally used in experiments.”

Is the financial world interested in findings such as yours?

PJ: “They definitely are paying attention: they’re looking for academic research that may have an impact on their own performance, of course. I think that’s why this paper of ours attracted so much media attention, because in financial institutions and in financial media, there are people constantly looking for these very relevant pieces of research.”

Was it a difficult choice to leave the University of Oxford for SBE and Maastricht University?

PJ: “I think it’s a perfect fit for me. One of the reasons why I came to Maastricht is that the Finance department has such diversified research values. They value all kinds of projects. There are people working on health-related issues, and people working on charity, for example, and not necessarily directly related to traditional focuses of finance.

“Oxford is excellent, as everyone knows. But their system of colleges and academic departments is a tricky organisational structure. It’s good that you can interact with people in the same college who are from different fields, but it can also be the case that you don’t have lot of interactions with people from the same academic department. People in the Economics department tend to come together only when there is a meeting or a seminar.

“Here, there is more room to develop my own research projects and more opportunities to meet with people who have similar research interests, as with the Human Decisions and Policy Design research theme. People with similar research interests coming together is really nice.

“I think the idea of research themes is a good one. In one of the first theme meetings I went to, some researchers presented their tentative grant proposals, and then other people gave comments and suggestions. That can be really helpful, if you talk with people from similar backgrounds but different subjects or different disciplines. As the theme leader, Lisa Brüggen made a lot of effort right from the start, talking with everyone to get to know their respective research projects and build connections. I think that’s really efficient. It’s much better than having people just randomly meet with each other, or organising some event in the pub where you can talk to a passing person: this is more effective.”

Would you say that your Bull of Wall Street study is the most significant research you’ve been involved in so far?

PJ: “Well, no, I wouldn’t… because the most significant research is always the next one!”