The real cost of divorce is inequality, not welfare

A report funded by a variety of "family values" groups made splashy headlines recently: "Single Parenthood Costs Taxpayers $112 Billion." The report lists two big reasons for these costs. One is increased welfare expenditures on poor families ($70 billion). The other is increased government spending to deal with the social problems caused by poor kids when they grow up ($42 billion).

The report is only partially correct; it vastly overstates the cost to society of single parenthood and fails to point out that social inequality is at the root of those costs. How do I know this? I did much of the research on which the report is based.


In the research by my colleague Guy Michaels and me, which is cited by the report, we isolated the causal effect of divorce on mothers' income. What we found surprised us: The average mom who divorces ends up with just as much income as an otherwise similar mom who stays married, and the overall effect of divorce on welfare spending is essentially nothing.

Why? Well, divorced moms are pretty resilient. They move in with relatives, they switch from part-time to full-time work and, perhaps most important, 70 percent remarry.


So that means there are two problems with this month's report. First, average income isn't affected, so there's roughly zero impact on tax revenues and welfare spending. Second, most of the mothers in our research end up remarried - so if the report is right about the other costs it identifies using our estimates, those costs aren't primarily caused by single parenthood.

Our research does find that divorce has a big impact on society. It's not an increase in welfare, however, but a rise in income inequality.

When the women we studied stayed in their first marriages, their families had a typical income of $48,108 in 2007. The worst-off 5 percent had incomes below $18,000, and 5 percent earned more than $83,000. By contrast, among mothers who divorced, the average income was about the same, but the bottom 5 percent had incomes below $4,000, while the top 5 percent had incomes above $171,000. Forty percent of moms whose first marriage ended had incomes outside the $18,000 to $83,000 range into which the vast majority (90 percent) of intact families fall.

In other words, divorce makes you more likely to become significantly richer or poorer. Some women get great jobs, marry successful men or both; others find out their skills aren't worth much on the labor market and marry men who are also struggling. The report got to $70 billion by wrongly counting only welfare spending on the losers, ignoring the higher taxes paid by the winners.

So does divorce cause economic problems? Not if you believe, as some do, that Americans' average income is all that matters.

But the report rightly points out that society experiences other costs from this income inequality, such as increases in criminal behavior. Cutting family income from $18,000 to $4,000 probably increases the odds that a kid grows up to be a criminal more than raising family income from $83,000 to $171,000 keeps that behavior from occurring.

In other words, if you take a group of middle-class kids and make some poor and others rich, you increase the crime rate. That means that even though the tax effects of divorce are a wash, the criminal justice and other behavioral costs from kids probably aren't.

This means that, measured correctly, the cost of divorce doesn't come from welfare payments but from crime and poverty linked to increasing American inequality. Of course, many other trends - uneven school funding and cuts to the estate tax in place of increases in the earned income tax credit, to name a couple - are also increasing inequality in our society.


Admitting that divorce costs taxpayers money and costs many children their futures means admitting that many other inequality-increasing policies have the same hidden price tag. I'm ready to admit that and get on with addressing all the costs and benefits, causes and consequences of rising inequality. I hope other groups interested in this problem are ready, too.

Elizabeth Oltmans Ananat is an assistant professor of public policy and economics at Duke University and a faculty research fellow at the National Bureau of Economic Research. Her e-mail is