Hollywood screen writers seem to love the statistic that 50% of marriages in the US end in divorce. For example, it seems mandatory when writing a stag party scene that at some point the groom takes his best man aside and says “Dude, what am I doing? More than 50% of marriages end in divorce!” To which his friend responds: “Not you guys. I don’t know what it is about you, man. You guys are just different.” They share a man hug, just so we know how sensitive they are, and immediately go back to doing belly shots.

Now there aren’t many economists writing screenplays (sad, I know) but really the best man should respond something like this: “Not you guys. If you condition for education and income divorce rates are actually quite low. You and [name bride here] both have a post-graduate education. I think you are going to make it man!” And THEN they go back to the strippers, because having post-graduate education actually means that you are more likely to visit a strip club, not less.*

And, of course, the dude would be right. If divorce rates are a good measure of success in marriage, then it seems that the more educated you are the more likely your marriage will be a happy one.

A new paper reports the following evidence.** If you have less than 11 years of education (less than high school) the probability you have ever been divorced is around 16%. With 12 years of education that probability drops to 10%. By the time you get to post-graduate education (more than 16 years) the probability of ever having been divorced is less than 3%. All of this is conditioned on income, so it isn’t higher incomes that are causing this differential.

Using data collected from sibling pairs, they also find that a 1 year increase in education rates decreases the probability of ever having been divorced by about 0.23%.

Now, there is one sure fire way to make sure you never get divorced and that is never to marry. The divorce rates by education level I quote above may seem low – and they are – but that is because while the authors condition on income they don’t condition on whether or not an individual has ever been married.

This would be a severe problem if more educated people married less often than everyone else. But despite the fact that more educated people marry later in life they are, in fact, more likely on average to be married. That fact means that these statistics understate the success of educated individuals on the marriage market compared to those with less education.

So why do educated people divorce less often? It could be that they are more competitive on the marriage market and so are more likely to marry a person who will make them happy. Maybe because they marry later they make better spousal choices. Perhaps divorce is more costly to those with high education and so they take the steps they need to ensure that the marriage persists. Or perhaps educated people are better negotiators or have more equality in marriage.

I always tell my students when they grumble about the cost of tuition that the next time they go to Tim Hortons to buy a coffee they should take a moment to thank the person behind the counter for helping to pay for their education. I tell them this to remind them that while they do pay some of their costs, the majority of their educational expense is paid for by the government. Of course we need an educated population for purely pecuniary reasons, which is why we spend this money, but if education improves the quality of marriage, and by extension child welfare, then maybe that makes educating the population a worthwhile investment.

* Brooks, Taggert (2007). “In Da Club: An Econometric Analysis of Strip Club Patrons.” Working Paper.

** Oreopoulos, Philip and Kjell G. Salvanes (2011). “Priceless: The Nonpecuniary Benefits of Schooling.” Journal of Economic Perspectives Vol. 25 (1): Pages 159–184.