Testosterone replacement therapy is popular. In men with a deficiency, injections should provide pep, a stronger libido and lower anxiety. Now many with normal levels hope to achieve the same benefits, as well as slower ageing, even though the jabs are unproven and risk side-effects including infertility. Adherents include Joe Rogan, a podcaster; Robert F. Kennedy junior, America’s health secretary; and Robbie Williams, a singer portrayed as a monkey in a recent biopic.
The motley crew will be excited to learn of new work by Peter Eibich of the Paris Dauphine University, Ricky Kanabar of the University of Bath and Alexander Plum of the Auckland University of Technology. In a working paper, the trio suggest a link between testosterone levels and earnings in British men. Their findings indicate that moving a man from the 25th percentile of circulating testosterone to the 30th would lead to a rise in monthly earnings of £277 in 2015 prices, or 10% of average wages.
To obtain these results, Messrs Eibich, Kanabar and Plum made use of the UK Household Longitudinal Study. This not only provides long-term data on social and economic outcomes, but also gathers biological and genetic information. By looking for single nucleotide polymorphisms (SNPs), genetic signatures associated with levels of testosterone, the researchers were able to calculate a “polygenic score”, which they showed did indeed reflect hormonal balances as measured in blood samples.
The relationship between pay and testosterone only applies to men with lower testosterone levels. On top of this, gains were only seen in the highest-earning quarter of men. The relationship is mainly driven by variations within occupations, rather than between them, indicating high- and low-testosterone men do not choose different jobs. Instead, testosterone seems to be affecting performance in some way.
Despite their best efforts, the researchers were unable to find anything that might explain this relationship. For example, it does not appear to be the case that testosterone affects the number of hours worked or performance-related pay. Nor is there any visible impact on educational attainment, cognitive functioning or the “big five” personality traits (agreeableness, conscientiousness, extroversion, neuroticism and openness to experience). Although more testosterone leads to a higher risk tolerance, adjusting for this does not explain its impact on wages.
Could medical research provide clues? Levels of testosterone in therapeutic doses appear to have little or no link with aggression. Yet they do raise sensitivity to status-related cues and, for low-status folk, the desire to compete for a higher rung on the ladder. Caring more about what other people think is not the sort of alpha-male behaviour beloved by Mr Rogan. It might, nevertheless, help explain the pay mystery. ■
For more expert analysis of the biggest stories in economics, finance and markets, sign up to Money Talks, our weekly subscriber-only newsletter.