A new report from AEI and the Institute of Family Studies shows that the share of married adults, and especially married parents, are associated with higher per-capita gross domestic product and lower levels of violent crime.
Yesterday I was on a panel at the American Enterprise Institute about families and state-level economic performance. We were discussing a new report from AEI and the Institute of Family Studies that shows that the share of married adults, and especially married parents, are associated with higher per-capita gross domestic product and lower levels of violent crime.
It’s one of the best predictors of better outcomes that they studied. “In fact,” write authors W. Bradford Wilcox, Joseph Price and Robert I. Lerman, “it’s a better predictor of states’ economic mobility, child poverty, and median family income than are the educational, racial, and age composition of the states.”
Saying that family structures predict outcomes is not, of course, the same as saying that they caused them. If I look out the window and see icy streets, I can make a pretty good prediction that it’s cold outside. That does not tell me that the cold is being emitted from the ice.
And so, unsurprisingly, much of the discussion focused on causality. Does a strong economy help people form strong families, or do strong families make for a stronger economy? I don’t have to tell you why this question matters: The direction of the causation should shape countless government policies. Liberals tend to believe that the causation mostly runs from the economy to the family, and conservatives tend toward the opposite explanation.
Both arguments have some evidence to back them up. In the worst year of the Great Depression, marriage rates were down about 20 percent from their pre-crisis levels. No one thinks that the Great Depression happened because the world experienced a huge decline in the desire to marry, so we can be pretty sure that the causation ran from economic disaster to young people declining to tie the knot. We also have studies showing that people who get better-paying, more stable jobs are more likely to marry.
Unfortunately, we don’t have similarly elegant studies looking at the opposite question. Any social scientist who proposed dividing people randomly into two groups — one forced to marry, the other forbidden to — would hear some very sharp words from the folks who monitor research on human subjects. We can look at differences between people who are married and people who aren’t, and see that married men, for example, tend to have higher earnings and otherwise superior outcomes to those who are unmarried.
The problem is that there are probably selection effects: People who get and stay married may be different sorts of people from folks who don’t, in ways that researchers would have a hard time observing and controlling for. They may be more patient, more conscientious, better plugged in to an extended family network that helps them survive the slings and arrows of outrageous fortune.
But even if we have trouble teasing out a causal relationship in the data, there are nonetheless strong reasons to think that two-parent families may help promote economic growth. For starters, two parents living in one household can devote more income and time to their children than even dedicated parents living separately, simply because it’s cheaper to live together than separately, and more parental time can be spent with the children than is possible when the parents are living in separate homes. Moreover, multi-partner fertility sets up competition for paternal resources between families in a way that you don’t see in two-parent families.
Beyond that, single parenting can be more stressful than co-parenting; ideally having another person there means you can distribute the load. The adults in a two-parent household can also specialize in the tasks they are good at. And as any economist will tell you, specialization and gains from trade are among the major drivers of economic productivity. Instead of two people taking 45 minutes to cook dinner for themselves, one person can cook for both in the same 45 minutes, freeing 45 minutes of labor for something else.
Having a spouse facilitates this specialization more than having an unmarried partner does. If I decide to invest less in my career in order to invest more time in raising kids and making the household run smoothly, while you devote time to boosting your earning power, that can make all of us better off. But if we’re just cohabiting, I might not want to make that investment. The easier it is for one partner to leave, and walk off with the benefits they’ve gained from specializing, the less specialization and investment we would expect to see.
So even if we don’t have a good way to test the economic benefits of married, two-parent families, we do see a plausible mechanism by which marriage could enhance economic performance both at the individual and macroeconomic level. That, along with the evidence that children from married, two-parent families have better outcomes on a variety of measures than those from single-parent or blended families, makes me think that there probably is a causal effect.
But as everyone on the panel agreed, we don’t actually have to choose. It can be true that marriage promotes economic growth, and also that economic growth promotes marriage.
Of course, that leaves us with the muddled question of what we should do about the decline in both economic opportunity and family stability that we’ve seen over the last few decades. There’s not much evidence that the government can successfully promote marriage — at least, not without undesirable measures like curtailing legal divorce.
On the other hand, the liberal arguments often end up sounding something like this: “Fixing marital decline is hard, so we need to do something easier, like ending poverty and reversing labor market polarization!” I was airily informed on Twitter that we knew what to do about the economic problems, so let’s at least do that. In fact, we don’t know what to do about the economic problems; if it was truly so obvious, we’d already have done it.
The decline of middle-class jobs results from a complex tangle of factors — in declining order of importance, technological change, trade and possibly immigration — that would not be easy to attack and would have substantial costs if we did. Cash transfers remain unproven, have costs of their own, and will run up against other budget priorities, like the massive cash transfers we’re scheduled to make to our nation’s old via Social Security and pension programs.
The problem the panel spent the most time discussing was establishing causation. But even if we solve that problem, the harder and longer discussions are yet to come.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.