Decision Making, Financial Risk Aversion and Behavioral Biases: The Role of Testosterone and Stress
- Nofsinger, Patterson and Daigler
- A version of the paper can be found here
- Want a summary of academic papers with alpha? Check out our Academic Research Recap Category
This study examines testosterone and cortisol levels of finance graduate students participating in three investment trials. The trials involve a portfolio asset allocation task with the last trial also including a series of rebalancing tasks in an increasingly competitive environment. Their testosterone and cortisol levels are positively correlated with financial risk in a competitive environment. However, testosterone is also positively related to more diversified portfolios. Specifically, higher testosterone participants choose higher risk asset allocations to earn a higher risk premium, but also chose more diversified portfolios to reduce unsystematic risk. Lastly, task success leads to higher levels of post-trial testosterone.
For example, Jia, Van Lent and Zeng (2015) found that male CEOs with higher testosterone levels are more likely to be associated with financial misreporting (read more: Manliness implies misreporting). Also, Coates finds that higher testosterone levels lead to more risk taking (anyone who has been around teenage boys knows this already!).
Hormones also play a critical role in the cognitive processes that guide financial decision-making. Researchers are trying to understand 1) which specific financial decision-making tasks are influenced and 2) the dynamic nature of those effects.
The effects can be dynamic because the financial success or failure associated with decisions can in turn affect hormone levels, in a feedback loop. Some have suggested this feedback loop, and the dynamic interaction of behavior, outcomes, and hormones can explain the occurrence of stock market bubbles.
In this paper, John Nofsinger and company, add to this literature by studying how testosterone and stress affect 1) goal-oriented decisions, as with investment or asset allocation decisions, and 2) completion-oriented decisions, as with speculative decisions relating to trading or rebalancing tasks.
They conduct three trials with graduate students at the University of Toronto, using a sample of 39 graduate students from a financial software course. The participants were familiar with financial trading simulation procedures and also had some basic financial knowledge. The trials were conducted at the end of the course, and the participants were told that their performance would be linked to a monetary reward and would have a grade impact.
Testosterone and Financial Decision Making Experiments:
First, saliva samples (used to measure the levels of testosterone and cortisol) were collected immediately before the trial (Note that we only highlight the first two trials in this post).
Trial 1: Asset Allocation Decision Making (Basic Mode)
- Participants were given a virtual endowment of $500,000 to invest. The stated goal was to grow this to $1.5 million by the end of the simulated 20-year time horizon, which meant they would have to achieve an annualized return of at least 5.65%. They were required to make an initial allocation decision.
Trial 2: Asset Allocation Decision Making (Competitive Mode)
- Before Trial 2 began, participants got to see the results of the first trial. They learned 1) whether they had achieved the goal in the first round; and 2) their ranking within the group.
- Then they were required to repeat the portfolio allocation task, which is the same as the first. The only difference was that participants were more stressed after learning their absolute and relative performance from the prior round.
- Trial 3: Rebalancing Task
- In Trial 3, subjects had the opportunity to rebalance their portfolios after 5, 10 and 15 years.
A second saliva sample was obtained at the end of the trials.
In the first trial, 34 of the 39 portfolios met the goal of achieving the $1.5 million target. After reviewing the results of the first task, the participants repeated the portfolio allocation task, but in competitive mode. Since most participants achieved the goal for the first trial, we would expect that they would choose the same allocation in the second task. However, only 5 did so.
Most participants increased their allocations to riskier, higher expected return assets for the second trial. Since the only difference between the tasks was a more competitive environment for Trial 2, the authors attributed the increased risk taking as a response to the higher level of competition. Subjects wanted to do well versus their peers.
The authors examine whether such behavioral changes are related to hormone levels. They found that in competitive mode, participants with higher levels of testosterone and cortisol tended to choose higher risk asset allocations to earn a higher risk premium. But they also tend to choose more diversified portfolios to reduce unsystematic risk. So while risk was increased, it was not reckless risk, but calculated risk.
In Trial 3, the authors found that the higher the returns achieved, the higher the testosterone in the subjects. This seems to support the idea that outcomes can drive changes in hormones.
Robert Shiller has argued that speculative bubbles are driven by increasing prices, which affect investor optimism, which drives more buying. The evidence in this paper is consistent with this idea and suggests that winning begets more risk-taking. Perhaps this can help explain “momentum” or other effects in the markets? Who knows…
But one thing is clear — there are a lot of cool studies coming out in the academic literature related to our brain chemistry and decision-making. Psychologists have determined that the environment can have direct affects on our decision-making, even when we are faced with the same facts. Now researchers are drilling down to the chemical level to try and determine exactly how our brains operate and influence our decisions. This piece of research is another link in that research vein.
- The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Alpha Architect, its affiliates or its employees. Our full disclosures are available here. Definitions of common statistics used in our analysis are available here (towards the bottom).
- Join thousands of other readers and subscribe to our blog.
- This site provides NO information on our value ETFs or our momentum ETFs. Please refer to this site.