Economist Writing Every Day

Last week I wrote about the “marriage market“. In many ways, the marriage market is like a labor market: there are search costs, match quality, competition for mates, and so on. When one side of the market becomes more abundant, that side become less picky — their minimum willingness to accept goes down.

Today we examine war as a shock that makes women more abundant than men. The marriage market predicts this shock will mean a smaller fraction of women will marry but those that do receive a smaller share of the benefits from marriage. Also, a larger fraction of men will marry and receive a larger share of the benefits from marriage.

Ran Abramitzky, Adeline Delavande, and Luis Vasconcelos investigate marriage in pre-and-post World War 1 (WW1) France where an estimated 16.5 percent of the French male population died or were missing in WW1. You can see from the map below that some “departments” suffered greater losses than others. Across France, the sex ratio was nearly even before the war and after the war, “If we focus on singles, widows, and divorces who were 30 years old or younger but of marriageable age, there were approximately 2 men for every 3 women.”

We can see for women born between 1891 – 1895 the distribution for year of first marriage is bimodal. The hollowed out area is because of WW1. You can also see a large spike in marriages right after the war for women born from 1896 – 1900. This suggests a sense of urgency to get married. In this game of musical chairs, you don’t want to be the one left without a chair.

Many women who would have married remained single. In the following graph, you can see the fraction of women who are single at age 50 by different birth cohorts. The women of marriageable age around WW1 were more likely to be single. At the same time, for men, there was an increase in marriage. The dearth of men meant that some men who would have otherwise remained single became marriage material.

One way to conceptualize men receiving a larger share of the benefits from marriage — due to their relative scarcity — is “marrying up”. The authors then create a variety of social class measures. For example, one social class measure uses marriage certificate data and reported bride and groom occupations. Another uses the Father occupations for the bride and groom. Using differences in mortality across departments to instrument for sex ratios, the authors report, “Overall, our results show that the increase in male scarcity in post-WWI France enabled men to improve their position in the marriage market and marry women from higher social classes when compared with prewar standards.”

The supply-and-demand model is a helpful metaphor in the dating and marriage market precisely because it gives us important predictions about how changes in environment or economic circumstance will affect outcomes. Next week I will review the book “Date onomics” and then will recount a talk I’m giving on February 10th to the Economics Club on the “Economics of Romance” but will continue to explore the comparative statics of the dating market until the end of February.