I had an interesting argument with Charles Murray at yesterday’s Cato Book Forum.  While he expressed fundamental agreement with my views on nature and nurture, he thought parental marital status was an important exception.  Children of divorce do worse than children whose parents remain married; children of never-married parents do worse than children of divorce.  At least at first, Murray seemed to see these disparities as entirely causal: getting married causes your kids to do better in life; getting divorce causes some (but not all) of that benefit to go away.

I objected that divorce and single parenthood are not random.  People who divorce are on average more impulsive and quarrelsome.  Single parents are on average more impulsive and less achievement-oriented.  Since these traits are heritable, we’d expect children of divorce and children of single parents to have worse outcomes – even if they were adopted at birth by Ozzie and Harriet.

I also mentioned that if Murray were right, he shouldn’t express fundamental agreement with me.  After all, about 40% of divorces end in marriage, and about 40% of kids are born out of wedlock.  So if marital status matters as much as Murray says, my results hold only for (1-.4)*(1-.4)=36% of the population.

In his response, Murray seemed to admit that some of the apparent effect of marital status on kids wasn’t causal.  But he insisted that much of it was.  How could we otherwise explain the simultaneous decline of the traditional family and the rise of so much social pathology?  I failed to fit in a full response, but here’s what I would have said:

1. Lots of factors changed since the 50s besides family structure: Expansion of the welfare state, increase in rights of the accused, skill-biased technological change, feminism, decline of religion, etc.  Why single out family structure, especially given all the adoption and twin evidence that sheds so much doubt on such explanations?

2. Many social pathologies have been falling for the last two decades, even though in Murray’s eyes, family structure is now worse than ever.  (Divorce is down a bit, but single parenthood is way up).

3. Even if family structure explains the changes, the causation could easily work across families rather than within them.  Maybe the problem is living in a world where lots of kids don’t have dads around – not whether or not your dad is around.  If so, Murray’s point is relevant for policy-makers but not for individual parents.

4. Given the prevalence of single parenthood and divorce, adoption and twin methods should pick up much bigger family effects than they do.

A few hours after the talk, Garett Jones reminded me that there’s at least one paper that tries to adjudicate my dispute with Murray: O’Connor et al’s “Are Associations Between Parental Divorce and Children’s Adjustment Genetically Mediated?” published in Developmental Psychology in 2000.  The study uses the Colorado Adoption Project to measure the causal effect of divorce.  The results are mixed:

In biological families, children who experienced their parents’ separation by the age of 12 years exhibited higher rates of behavioral problems and substance use, and lower levels of achievement and social adjustment, compared with children whose parents’ marriages remained intact. Similarly, adopted children who experienced their (adoptive) parents’ divorces exhibited elevated levels of behavioral problems and substance use compared with adoptees whose parents did not separate, but there were no differences on achievement and social competence. The findings for psychopathology are consistent with an environmentally mediated explanation for the association between parent divorce and children’s adjustment; in contrast, the findings for achievement and social adjustment are consistent with a genetically mediated explanation involving passive genotype-environment correlation.

Read the whole piece for details.  The main problem with the study is that it focuses on 12 year-olds.  We still don’t know how divorce affected kids’ adult outcomes.  But as a general rule, long-run effects are smaller than short-run effects.

I have a knee-jerk horror of divorce.  But if you asked me, “What’s so bad about it?” I’d still downplay what social scientists call “adult outcomes.”  Instead, I’d focus on the parent-child relationship.  Getting a divorce won’t ruin your child’s life, but it is fairly likely to forever maim or destroy the way your beloved child feels about you.*

* I tried but failed to find any academic research on this specific point, so I’m going with common sense plus the academic literature on appreciation.