It’s not me, it’s you

Every four years, I publish my World Cup forecasts as an exercise in showing that economists think they can meddle in everything, even though they have no clue and should better leave the topic alone. And every so often, I come across an economic study where I think they are inadvertently doing the same thing.

Enter José-Victor Rios-Rull, Shannon Seitz, and Satoshi Tanaka, who built a model to explain changing marriage and relationship preferences among men and women over time. To do that, they use data from Americans born in 1870, 1930 and 1950 to estimate a model of their preferences and the ‘utility’ they get out of a potential partner of the opposite sex.

Let’s ignore for a moment the odd choice of birth cohorts and the fact that this model has 22(!) parameters and an unknown discount factor to calculate the net present value of finding a partner in the future. Let’s instead focus on their baseline estimate of the ‘utility’ of a partner split by gender.

Partner attractiveness by age in the model baseline

The dashed red line shows the view men have of women of a different age. They tend to think of young women aged between 18 and 26 as more attractive than women over the age of 26. Meanwhile, they tend to think of girls below the age of 18 as providing lower ‘utility’ as a partner. Well, I very much hope they do, because having an underage partner is illegal. In that regard, I am really concerned about how small the step down in utility is for underage partners. Indeed, if I look at the red dashed line, I would say that in the eyes of men, all women are useless, creating effectively zero ‘utility’.

Meanwhile, women clearly differentiate (solid blue line). Older men have greater ‘utility’ to them, while boys under the age of 18 have large negative utility. Meanwhile, young men aged 18 to 26 have effectively zero ‘utility’ for women. I just let that sit here for you to ponder.

Then, they use this baseline model and look at how marriage behaviour has changed over time. I think I will just extensively quote from their abstract because it speaks for itself (in more ways than I care to explain):

“Women’s premium for older husbands collapsed across cohorts; men’s preferences barely changed. Love that survives its early years becomes permanent, but the odds of surviving fell from 97% to 44%. Divorce costs fell sixfold and depend on life stage. A horse race across behavioural channels shows that the match quality process – not mate-age preferences – is the primary dimension of generational change.”

Happy mating, everyone.