Prenatal testosterone predicts financial risk taking: Evidence from Latin America

•Prenatal testosterone and risk were related in a homogeneous non-Caucasian population.•The relation is significant for the left hand but not for the right hand.•The relation is weak and better captured by binary than by ordered models.•The relation was found for both 2D:4D and rel2.•Higher prenatal exposure to testosterone increases the choice of risky lotteries.

Most research on the relation between steroid hormones and risky behavior has been done for Caucasian populations using 2D:4D. Some articles with ethnically mixed samples were not conclusive. We studied the relation between prenatal exposure to testosterone (T) and financial risk in a Colombian population (89 male, 34 female). Colombian population is the result of miscegenation among Amerindian, Spanish and African groups. Prenatal exposure to T was proxied by 2D:4D and rel2 digit ratios. Risk behavior was elicited by the choice of a lottery in a 50%-50% Eckel and Grossman task with actual monetary payoffs. People with higher prenatal T exposure (lower left-hand 2D:4D, rel2 ratios) tend to choose more risky lotteries, although the effect is weak and better captured in binary-choice than in ordered-logit models. Results are statistically significant for the left hand but not for the right hand. An increment of one standard deviation in the 2D:4D/rel2 of the left hand diminishes the probability of taking the riskier lotteries by 8%/10%.