May 14, 2008

The Economics of Love-Part 2: Courtship

In the last post we saw that differences in gender preferences over the partner confirm old, stylized cliché: women like wealthy men, while men like physically attractive woman. If we consider that, roughly speaking, beauty is inversely related to age, while wealth profile increases positively over years, these ex-ante preferences seem to lead to an equilibrium where, within the couple, the man is older than the woman.

In this nice chart from the UN (2000) you find all the relevant statistics and the indication that it is indeed a world pattern. The smallest difference in mean ages was 0.3 years in Belize (Central America) and the largest difference was 8.6 years Congo and Burkina Faso (Middle, Western Africa). In 2000 in the United States it was 2.3. In western Europe it is about 2.5 years, and in southern and eastern Europe about 3.5 years, similar to that of Japan. In India and the Middle East is between 4 and 5 years. In Central America and South America between 3 and 4 years. In African countries, this gap ranges between 5 and 10 years.

It was argued that this equilibrium is consistent with explanation from evolutionary psychology according to which a younger female, from a men’s point of view, corresponds to a higher likelihood of successful pregnancy, while an older man, from a woman’s point of view, signals a higher potential in providing financial resources to raise the offspring.

The market for partners is dominated by informational asymmetries and you don’t want anybody, you want the right one, the best one. How do we find though the matching equilibrium? Let’s say there are two types of partners, high quality males and females (HQm,f) and low quality ones (LQm,f), and within each group, there is a ranking of qualities from highest to lowest. In a perfect information world, a HQm type will match with a HQf type; as long as all the same quality HQm,f’s have matched with each other, then the LQm,f will start matching, until everybody has found a partner. What happens when you don’t know what is the quality of your partner?How can you tell the difference between a HQm vs LQm?

In a paper published on the Journal of Political Economy (see reference), Theodor Bergstrom and Mark Bagnoli provide an answer that has to do with the concept of signalling. In a world where a male is considered the “resource provider” within the couple, then information about his own capabilities may be revealed only after he has spent some time in the workforce. From a woman point of view, where her main capabilities are related to childbearing, there is little additional signal to convey as time goes by. Men who expect to prosper, will delay marriage until they are able to attract the best available partners. The most desirable females will instead marry relatively earlier. In the long run, unsuccessful men will marry earlier in life than successful men. As all women marry relatively early, the best ones will marry older men, while the less desirable ones will merry young males with lower earning potential.

The model can explain the observed stylized facts cited above: both the equilibrium age difference between men and women, but also the differences observed across countries. Where labor market opportunities for women are higher (developed countries), the age difference should be smaller: as women also compete for the best partners, as their role become more similar to that of the men, thus becoming less specialized in childbearing, their earning potential also becomes a signal to the partner. The age gap in the couple would still be higher in developing countries, where this proximity is genders’ roles is far from achieved.


This model has another clear empirical prediction: more successful men should get married later in life. As such, it has some important implications for all the graduate students of the world: studying may be interpreted as a rational choice also from a dating perspective, since it increases your marketability to attract better partners. Good news, aren’t they? Well, however, since also finishing your studies on time is also a signal of future successfulness, then it is also rational to wait until you are really done with them before dating anybody, otherwise your signal may not be credible...as any good economic choice, there is always a trade-off!

  • Theodore C. Bergstrom and Mark Bagnoli , “Courtship as a Waiting Game”, The Journal of Political Economy, Vol. 101, No. 1, (Feb., 1993), pp. 185-202

3 comments:

Cam said...
This comment has been removed by the author.
Cam said...

I have to say that when I applied to do a PhD, my chief motivation wasn't that it would eventually 'help me pick up girls'. In Australia, since sporting achievement is more glorified than intellectual attainment, having a PhD is more a signal that you are freakish nerd, rather than an attractive quality. Proof perhaps that the outcomes of rational actor models aren't universally transposable across countries and cultures.

cosi said...

Great piece, Salvi! Cam, I see your point. However, I think that we PhD students have slightly more sophisticated preferences than people with lower education. For example we value intellectual matching very much. This criterion will tend to exclude top models or lifeguards with no brains from the preference set of males and females respectively. Since we have low interest in forming couples with people who grossly mismatch our intellectual capacity, we shouldn't care too much about those who think we are freaking nerds.