The US divorce rate remains roughly 50%. Because divorce is costly and carries social stigmas, the percentage of unhappy marriages is likely well over 50%. Apparently the two peak periods for divorce are 4-5 years into a marriage and 25 years in. Assuming this is right, and focusing only on preference-maximization, shouldn’t a libertarian paternalist prefer a default rule on which marriages expire after 4-5 years? At that point a couple could, of course, opt to remain married (by filing a form or an affidavit or something like that), but they would have to take affirmative steps to do so. This regime should have the effect of shifting the divorce rate from 50% to a higher, more preference-maximizing, rate.
One possibility is that the 50% divorce rate actually reflects rational decisions made by couples who weigh the costs (including social stigma) of remaining married against the benefits. There are two reasons to be skeptical that this is true.
- Behavioral Economics. The first reason, near and dear to a libertarian paternalist’s heart, is that multiple cognitive biases should be at play in undertaking a cost-benefit analysis of whether or not to get divorced, with the upshot that many people are overestimating the costs of getting divorced and underestimating the benefits. People unduly weight short-term costs and benefits over longterm costs and benefits. Additionally, “optimism bias” means people are likely to underestimate the costs of remaining married overall — optimistically believing that things will get better. Finally, a marriage’s unhappiness is often the product of small aggravations building over the course of years; according to behavioral economics, people tend to unduly discount small costs.
- The effect of law on social norms. The second reason is that creating a new default rule according to which people are automatically divorced after 4-5 years, absent their taking affirmative steps to remain married, is likely to itself go some way toward legitimizing divorce and thereby diminishing the social stigma associated with it. This is the “normative power of the actual” (see here, footnote 154, for a discussion of the etymology of the phrase). By implementing this policy — actualizing it — we would be taking a step toward removing the social stigma obstacle to preference-maximization.
Of course preference-maximization is not the only goal of public policy. There are a variety of reasons a policymaker might want to keep the divorce rate as low as possible even if it results in a higher incidence of unhappy relationships. But the utilitarian case for a legal presumption that marriage is not for life seems strong to me.