Male income and reproductive success

As happens occasionally, I have just come across an article that I should have seen years ago. In an article titled Natural Selection on Male Wealth in Humans, Daniel Nettle and Thomas Pollet look at data from a variety of societies, ranging from subsistence groups to industrialised societies, and show that the link between male income and reproductive success is strong and ubiquitous.

In their main analysis, they use longitudinal data for a group of British men born in a single week in March 1958. These men were tracked through to age 46. With around 96 per cent of reproduction for men occurring before this age, the number of children at this age provides a good guide to reproductive success.

Among the men in the sample, the effect of male income on reproductive success was positive, with a selection gradient of 0.10. This means that for every standard deviation increase in income, there is a 0.1 standard deviation increase in reproductive success. This may not sound like a lot, but it is strong selection compared to that observed in many other species. This effect was mostly driven by whether people had any children or not, rather than through variation in family size.

The authors also controlled for education by splitting men into three separate education groups (which has a negative effect on fertility for both men and women), which tended to strengthen the observed effect of income.

Where we see even stronger selection, however, is in the historical data. Nettle and Pollet took their results and compared them to selection gradients calculated from a range of other papers. They found that for contemporary United States and Sweden, English testators in 1540 to 1850, Norwegian farmers in 1700 to 1900, and a range of pastoral and hunter-gatherer groups, there was a positive selection gradient based on male wealth or hunting ability.

Further, while the strength of selection in the three contemporary societies was similar to each other, it was lower than in the pre-Industrial societies. The authors suggested that this weakened pressure was largely due to an increase in the level of monogamy and reduced variation in effective family size, rather than changing preferences for income. As the authors note however, while weakened, the selection in modern societies is still significant.

An underlying question to these results is whether the selection of phenotypic features (income) has any relationship to underlying genetic qualities. Both income, and many of the factors underlying income, such as IQ, have been found to be heritable, so there is likely to be evolutionary changes to the composition of the population.

The other question is why is there this link. The most obvious explanation is that women prefer men with more resources, both for the resources themselves and for the fact that they signal quality. That the effect of income was largely related to whether someone had zero or more children, rather than size of family, might be suggestive of this. To have more than zero children, you need to attract a mate. However, while doubting the idea, the authors did not rule out an explanation that men who have children earn more to support them.

Economists might try to pin this relationship down to the decisions of the male with higher income. If children are normal goods, a man should want to buy more as their income goes up. There is a range of issues with this argument, and I am going to address them in a post next week.