Skip to content
What Causes What
MECHANISMUNSIGNED

Testosterone → Economic risk tolerance

The atlas has not yet labeled the sign from Testosterone to Economic risk tolerance.

Large randomized trials do not show that a testosterone dose moves lottery risk. A preregistered study of 1,000 men found no effect of a single nasal dose. Four weeks of testosterone in 200 postmenopausal women also left risk attitudes unchanged.

Randomized trial · Contested · N=1000

Supported by an experiment or a tightly identified design. Read the caveats.

Mechanism

The folk story is that testosterone raises status-seeking and reward drive, so people should take more financial risk. Dreber, Johannesson, Nave, Apicella and colleagues gave 1,000 men either 11 mg of nasal testosterone or placebo and, after 30 minutes, an adaptive 40-choice task of a lottery versus a certain payoff. They found no treatment effect on risk aversion or loss aversion, and no reliable link from basal saliva testosterone to those same measures. Zethraeus, Kocoska-Maras, Ellingsen and colleagues gave 200 postmenopausal women four weeks of testosterone, estrogen, or placebo, then asked for the certainty equivalent of a 50/50 gamble to win 400 kronor. Nothing moved. Stanton, Welker, Bonin, Goldfarb, and Carré ran two testosterone trials and one correlational study of economic risk and loss aversion and saw no consistent pattern. Nadler, Wibral, Dohmen and colleagues found the same null on economic risk-taking after a single dose in two experiments (91 and 242 men). Vartanian and colleagues restored testosterone during a 28-day energy deficit; risky choices about cash and lives did not change. The observational file is different. Stanton, Mullette-Gillman, McLaurin and colleagues saw a U-shape in 298 men and women: low or high endogenous testosterone went with risk-neutral certain-versus-gamble choices, mid-range testosterone with risk aversion. Huang, Zhang, and Zhuang saw higher basal testosterone track more play on a personal probabilistic gambling wheel in 69 men. The same slope appeared when winnings went to charity. Cortisol did not moderate that slope.

Caveats

Dreber studied only men aged 18 to 45, one dose, and one timing. A longer course, a different dose, or women could still differ. Zethraeus studied postmenopausal women; the 4-week oral course is not a nasal pulse. Huang is saliva and a wheel gamble, not Holt–Laury, and N is 69. Charitable risk is not this page; the personal arm is. Vartanian's sample is 50 men under severe energy deficit; the hormone was restorative. Lieberman's balloon task from the same trial had no money at stake and is not on this page. A methods paper on testosterone and ultimatum bargaining is not this page. Failed replications lower confidence. They do not change the design stamp. Seven papers do not make the stamp more certain.

Effect

qualitative. Dreber N=1000: null on lottery risk and loss aversion. Zethraeus N=200 women: null on certainty equivalents. Smaller RCTs mixed or null. Stanton 2011: U-shaped endogenous T. Huang: positive personal-gamble slope.

Nodes

Biology & endocrinologyTestosteronePsychology & behaviorEconomic risk tolerance

Named confounders

Cite this page

What Causes What. “Testosterone → Economic risk tolerance.” https://whatcauseswhat.org/edges/e-t-risk (atlas updated 2026-09-02).

https://whatcauseswhat.org/edges/e-t-risk

Papers