In the 1960s debate [over no-fault divorce], no one thought the divorce rate would change, but it changed enormously and led to a divorce culture.
A lot of social conservatives, like Allen, believe that the evidence strongly supports the view that no-fault divorce caused a permanent change in the culture — a “divorce culture,” in people get divorced at the drop of a hat.
In fact, the best evidence indicates that any changes to the divorce rate caused by no-fault divorce laws were temporary.
Douglas Allen — yes, the same Douglas Allen — and Maggie Gallagher wrote a useful review of the empirical research on no-fault divorce laws and divorce. Allen and Gallagher, both of whom are conservative on marriage issues, cite two papers in particular as the “high water mark” of divorce research, saying that “Friedberg’s study stood as the high-water mark of the no-fault divorce literature
until the arrival of Wolfers (2006).”
Much of the debate over no-fault divorce and divorce rates seemed to be over with the publication of Friedberg’s (1998) seminal work in the American Economics Review. This paper created a panel data set of every divorce in the United States from 1968 to 1988. It used sophisticated econometric techniques to control for state endogeneity and changes in behavior over time. She tested for different legal classifications, and performed a series of robustness tests. In the end she found that no-fault divorce laws led to a 6% higher divorce rate and that they accounted for about 17% of the increase in divorces over the time period studied. She also found that the change was permanent, and exogenous. Differences between states and changes over time, however, accounted for most of the divorce trends. She concluded: “The results above make it clear that unobserved covariates and unobservable divorce propensities — which may include for instance, social attitudes, religious beliefs, and family size — are the main determinants of divorce.”
Friedberg’s study is excellent, but it had one unavoidable limitation: Because it was conducted so soon after many states instituted no-fault divorce laws, it was not able to distinguish between permanent changes and temporary changes. When Wolfers replicated Friedberg’s study in 2006, extending it with up to date data (pdf link), he found that the increase in the divorce rate Friedberg had observed disappeared after about ten years.
A clear finding from this analysis is that the divorce rate exhibits interesting dynamics in response to a change in legal regime. [...] The data broadly indicate that divorce law reform led to an immediate spike in the divorce rate that dissipates over time. After a decade, no effect can be discerned. [...] It should be clear that unilateral divorce laws explain very little of the rise in the aggregate divorce rate.
Allen and Gallagher classify Wolfers as evidence that no-fault divorce laws led to a change in divorce rates, and technically they are correct. However, for most people, the difference between a temporary change in divorce rates and a lasting change in divorce rates is essential.
If “after a decade, no effect can be discerned,” then it is not legitimate to claim that the divorce law reforms of the 60s and 70s created a permanent “divorce culture,” and were a disastrous change. Rather, it seems as if the change to the laws had virtually no long-term effect on the divorce rate; the law changed in response to the culture changing, but it did not itself create cultural change.