Testosterone Does Not Shape Men's Economic Choices, Large Trial Finds

A preregistered randomised trial shows why organisations should treat biological explanations of behaviour with caution until the evidence is robust.

Testosterone Does Not Shape Men's Economic Choices, Large Trial Finds

For years, behavioural science has linked testosterone to risk taking, generosity, and competitiveness. A new preregistered randomised trial, the largest of its kind, tests that claim directly. Led by researchers at the Stockholm School of Economics and Nipissing University, with Wharton’s Gideon Nave among the authors, the study gave men either a single intranasal dose of testosterone or a placebo. It then measured their choices across standard economic tasks. No statistically significant effect appeared on any of the nine main outcomes.

The findings matter well beyond endocrinology. They caution organisations against building talent, risk, and governance assumptions on biological explanations drawn from small early studies. Leaders, policymakers, researchers, and anyone who commissions or relies on behavioural evidence should read this as a case study in evidence quality.

Key Insights

  • Men given testosterone made the same economic choices as men given a placebo across risk taking, generosity, willingness to compete, and fairness preferences.
  • The trial found no statistically significant effect on any of its nine main outcomes.
  • Earlier studies suggesting an effect were often based on small samples, which the authors say limits how far those results can be trusted.
  • Preregistering the analysis plan before seeing the data reduced the room for biased or selective reporting.
  • The authors argue that most behaviours reflect many psychological and biological factors, each with a small effect.
  • The study reinforces a wider pattern in behavioural science in which results from small samples often fail to replicate.

What Made the Difference

The design of this trial is what gives its null result weight. It was substantially larger than earlier experiments, participants were randomly assigned to testosterone or placebo, and the authors registered their analysis plan before observing the data. Together these choices limit the flexibility that can allow chance findings to appear meaningful.

Participants received a single intranasal dose and waited 30 minutes for the hormone to take effect. They then completed well established behavioural economics tasks measuring risk taking, generosity, willingness to compete, and fairness preferences. Across nine main outcomes, the testosterone group did not differ from the placebo group.

The broader dynamic is one of scale and transparency. As Nave notes, most behaviours are likely shaped by many interacting factors, each contributing a small share. Small studies can easily overstate any single factor, and the effect can disappear once the test becomes more rigorous.

Why It Matters for Organisations

Organisations regularly encounter confident claims that a single trait, hormone, or demographic characteristic explains how people take risks, negotiate, or compete. When such claims shape hiring, promotion, or team design, weak evidence can quietly become policy.

This study is a reminder for boards and executives to ask how strong the evidence behind a behavioural claim really is. Sample size, preregistration, and replication are practical indicators of reliability, and they belong in the due diligence process alongside financial and legal review.

For policymakers and institutions that fund or commission research, the implication is similar. Investment in adequately powered, transparent studies is more valuable than a steady stream of small studies that produce striking but fragile headlines.

From Research to Reality

Organisations can act on this by building an evidence standard into how they use behavioural research. Before adopting an insight about decision making or leadership, teams should check whether the finding has been replicated, whether the sample was sufficiently large, and whether the analysis was specified in advance.

Risk and talent frameworks should be designed around structures, incentives, and processes rather than assumptions about biological predisposition. Where governance depends on how individuals behave under pressure, the design question is which systems support consistent, well informed decisions.

Commissioners of research, including universities, NGOs, and government entities, can also raise the bar by requesting preregistration, clear reporting of methods, and openness about null results. Null findings are informative, because they prevent resources from being directed towards effects that do not exist.

Key Takeaways

A substantially larger, preregistered experiment found no effect where earlier small studies had suggested one.

Men given testosterone made the same economic decisions as those given a placebo across all nine main outcomes.

Setting the analysis plan before seeing the data limits selective reporting and strengthens confidence in the result.

Most behaviours reflect a combination of psychological and biological factors, each contributing a modest share.

Organisations should apply checks on sample size, replication, and transparency before acting on behavioural claims.

“”Findings reported from small experiments can vanish when tested more rigorously.”  — Magnus Johannesson, Professor of Economics, Stockholm School of Economics

Source: “Does Testosterone Affect Men’s Economic Decisions?”, Knowledge at Wharton, January 2026, based on Dreber, A. et al. (2025), Proceedings of the National Academy of Sciences

At Global Consultancy, we translate research like this into evidence-based strategy. 

Get in touch to explore what these findings mean for your organisation.

3 Comments

Leave a Comment

Your email address will not be published. Required fields are marked *

0 %
Title

Latest Insights

  • All Posts
  • Global Reports
  • Research

Category

  • Global Reports
  • Research

Explore More Insights

Key findings at a glance.

Topic

  • AI
  • Behavioural Economics
  • Finance
  • Gender & Work
  • gender and work
  • HR strategy
  • inclusion
  • interpersonal discrimination
  • organisational behaviour
  • retail
  • unconscious bias
  • weight bias
  • Women Who Went First
  • workforce
  • workplace discrimination