According to NYT reporter Jason DeParle: “there are suggestions that the absence of a father in the house makes it harder for children to climb the economic ladder.” To support this proposition, he cites an “unpublished analysis” of the effects of single parenthood on intergenerational income mobility that only follows children until their mid-20s.

This seems an odd choice to say the least. In a blog post, DeParle says that “there is not much that examines long-term effects [of being raised by a single parent] on a child’s chances of moving up the income ladder.” At the end of the post, he claims that the unpublished analysis he relies on is “broadly consistent with those found in a study by the Pew Mobility Project, which used a different data set.” 

Ok, so why not use data from the Pew study? Among other things, it has the benefit of being published (not in a peer-reviewed journal, but it was reviewed by peers) and available on the Internet, and written by two experienced researchers at major universities who have published in peer-reviewed journals. It also follows children beyond their mid-20. 

Perhaps because that study tells a different and more nuanced story than DeParle wants to tell. Here are some of the main findings:

  • looking at “absolute” mobility, 88 percent of children born to unmarried mothers have greater family income than their parents compared with 82 percent of children of always-married mothers (not a statistically significant difference) and 74 percent of children of ever-divorced mothers (statistically significant), this same finding holds when looking only at parents in the bottom third of the income distribution.
  • looking at “relative” mobility, among children who start in the bottom third of the income distribution, 42 percent of those born to unmarried mothers move up to the middle or top third of the income distribution as adults, compared with 50 percent of children with continuously married parents (authors don’t say if statistically significant) and 26 percent with ever-divorced parents.  

So, contra DeParle, the suggestion in the published research is that:

  • using the absolute mobility measure, children born to unmarried mothers have no harder time climbing the economic ladder than children born to married mothers;
  • using the relative mobility measure, there may be a modest difference, but what is most striking here is how many children in both groups don’t move out of the bottom third; 
  • children born to unmarried mothers have an easier time climbing the economic ladder than children born to mothers who have been married but divorce. Note that this is broadly consistent with research I cited in a previous post finding that poverty rates for unmarried mothers are actually lower than for mothers who marry but then divorce.

One response might be that if these mothers hadn’t divorced their children would have done better than the children of unmarried mothers. But this seems unlikely. Parental conflict among married parents is associated with a number of poor outcomes. As Kelly Musick of Cornell and Ann Meier of Minnesota have found in a paper recently published in Social Science Research:

…. children from high conflict married-parent families do more poorly in the domains of schooling and substance use, and are at greater risk of early family formation and dissolution, relative to children from low conflict married-parent families. In half of our outcomes, high conflict, stepfather, and single-mother families are statistically indistinguishable in their associations with young adult well- being. These findings hold once account is taken of key mechanisms posited to link family type and child outcomes. They are consistent with recent research on marriage and the well-being of adults, showing that although marriage confers benefits to adults on average, those in poor quality marriages are no better off than the single and, indeed, may fare worse on some measures.

Finally, there is other relevant published research that would have been worth citing. In a 2004 paper published in Demography, “Family Structure, Intergenerational Mobility, and the Reproduction of Poverty: Evidence for Increasing Polarization?”, Kelly Musick of Cornell University and Robert Mare of UCLA show that “current rates of intergenerational inheritance [of poverty and family structure] have little effect on population change over time. They account for only a small share of the recent historical change in poverty and family structure and play no role in exacerbating existing economic disparities by family structure.” Mare and Musick conclude:

….intergenerational associations between poverty and family structure at the micro level do not necessarily translate into important population-level changes. Individual-level studies form the basis of our understanding of the inheritance of poverty and family structure. Intergenerational associations between poverty and family structure have been found consistently and lead to questions about their consequences for future generations. But these associations alone cannot reveal how they affect aggregate population trends. Although intergenerational inheritance may contribute in a small way, changes in the relative number of persons in different family and socioeconomic statuses must be understood in terms of broader social, economic, and cultural developments, such as shifts in the relative economic positions of men and women, ideational changes, and changes in the material aspirations of younger cohorts. They must be understood, that is, in terms largely outside the micro-level effects of family structure on subsequent generations.a

Similarly, in "Recent Trends in the Inheritance of Poverty and Family Structure", published in 2006 in Social Science Review, they note that "poverty and single parenthood are tied together in dicussions of the underclass and in perceptions of the public" and argue that "instead of assuming that the problems are closely linked to one another, we need to treat their interrelationships as a matter for empirical investigation." Using statistical techniques to untangle the associations between poverty and family structure, they conclude that:

Net of the correlation between poverty and family structure within a generation, the intergenerational transmission of poverty is significantly stronger than the intergenerational transmission of family structure, and neither childhood poverty nor family structure affects the other in adulthood. Finally, despite important changes in the distributions of poverty and family structure, we find no evidence of change in the processes of intergenerational inheritance over time.

If DeParle had done less tangling and more untangling, his piece would have been much more useful and informative.