Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/58361 
Year of Publication: 
2012
Series/Report no.: 
Working Paper No. 12-7
Publisher: 
University of California, Department of Economics, Davis, CA
Abstract: 
We correlate choice under risk in Holt-Laury lottery tasks for gains and losses with salivary testosterone, estradiol, progesterone, and cortisol, the use of hormonal contraceptives, menstrual cycle information as well as the digit ratio (2D:4D) in more than 200 subjects. Risk aversion is negatively correlated with testosterone and positively correlated with cortisol, a stress hormone, for gains only. In males, testosterone is negatively correlated with risk aversion for gains only. In females, cortisol is marginally significantly positively correlated with risk aversion for gains only. No other significant correlations between risk aversion and salivary hormones are observed. In females, testosterone and progesterone are positively correlated with reflection, i.e., risk aversion for gains and risk seeking for losses. Testosterone is negatively correlated with consistency of preferences in females, while estradiol is negatively correlated with consistency of preferences in males. No significant correlations between risk aversion and the menstrual cycle or the digit ratio are observed. Females on hormonal contraceptives are more likely to make consistent choices although this may be due to a selection effect. Risk aversion is positively correlated with being female for losses only. Yet, if we control for salivary hormones we are surprised to find a negative correlation between female and risk aversion for gains.
Subjects: 
Hormones
Menstrual cycle
Contraception
Digit ratio
2D:4D
Gender
Risk behavior
Endocrinological economics
Holt-Laury
Risk aversion
Risk seeking
Reflection effect
Prospect theory
JEL: 
C91
C92
D44
D81
D87
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.