Explaining the divorce rate

Justin Wolfers writes:

I had an op-ed in yesterday’s New York Times noting a very simple fact: those married in the 1990s have proved less likely to divorce than those wed in the 1980s, which were less likely to divorce than those wed in the 1970s. The Divorce Facts are that divorce is falling, and marriages are more stable.

Here is an academic paper that tries to look at data to test a couple of theories about the effect of unilateral (aka no-fault) divorce law on the divorce rate. Different states adopted unilateral divorce at different times, so it is possible to look at how the divorce rates changed relative to the changes in law. There were of course other changes in our society that are difficult to measure. I'll try to explain some of this in plain English.

1. Coase theory. The Coase theorem says that shifts in property rights can be counterbalanced by private contracts. This theory predicts that changes in divorce law would have no effect on the divorce rate.

Suppose a husband wants a divorce and the wife does not. Under unilateral divorce, he may not get the divorce because the wife may somehow sweeten the marriage contract in order to induce him to stay. Likewise, under the old (mutual-consent-required) divorce law, he might somehow pay off the wife to induce her to agree to the divorce. Assuming that these inducements are readily available, easily negotiated, and legally enforceable, then the theory says that husbands and wives will find socially optimal arrangements, regardless of divorce law.

2. Naive probability model. Suppose that there is a probability of 40% that a married person will want a divorce after 10 years of marriage, independent of all other factors, including whether the spouse also wants a divorce. Then the divorce rate should have jumped from 16% to 64% after changing to unilateral divorce.

Out of every 100 couples, 40 men want divorce and 60 men do not. Of those 40 marriages, 40*40%=16 of the women want divorce and 24 do not. Of the other 60 marriages, 60*40%=24 of the women want divorce, and 36 do not. So if mutual consent is required, then only 16 will get divorced. But there are another 24+24=48 marriages in which exactly one party wants a divorce, so you expect 16+24+24=64 divorces under unilateral divorce.

The article finds that the unilateral divorce caused an immediate jump in the divorce rate, but may not have had any long term effect. It may have also caused a small drop in the marriage rate.