How We Should Really Think About Divorce: A Response to Cornwall and Coltrane

From my vantage point as a scholar of the family, the changing public discourse surrounding divorce is fascinating. Over the past 500 years, the divorce rate has almost always ticked upward, but it rose to unprecedented heights during the divorce boom that lasted from about 1965 to 1980. Since then, the rate has almost continually fallen, the longest sustained decline in recorded history. This has given scholars and journalists a lot to write about, but comparatively little attention has been paid to the intellectual history of divorce—in other words, how we think about divorce, and how our perceptions have evolved over time. So it was encouraging to see journalist Gail Cornwall and academic Scott Coltrane recently step into the breach with an article for Slate (“How Americans Became Convinced Divorce Is Bad for Kids.”) Unfortunately, they got a lot of things wrong (but a few things right).

First off: is divorce bad for kids? Cornwall and Coltrane deserve credit for casting a broad net in trying to answer this question, considering everything from the economics to the genetics of marital dissolution. But their writing is tendentious, stubbornly laboring to deny the possibility that the answer to this question could conceivably be yes.

A proper answer is indeed complicated. Let’s start with one of my favorite examples from a classic of the field, Sara McLanahan and Gary Sandefur’s Growing Up With a Single Parent: What Hurts, What Helps. The authors show that 31% of children from divorced families don’t graduate from high school on time, compared to 13% of kids from two-parent families. There are two ways of seeing these stats. If your cup is half empty, you point out that divorce more than doubles the chances of dropping out of high school. If your cup is half full, you’re equally correct to observe that more than two-thirds of children from divorced families manage to finish high school on time, and a small number of kids from intact families drop out anyway. Clearly many kids from divorced families do fine, but divorce is nonetheless correlated with an adverse outcome. Does correlation equal causation in this instance? I’ll return to this question in a bit.

The reasons divorce affects kids are numerous. Divorce reduces income. Divorce leads to less parental supervision. Divorce exposes kids to conflict. Divorce leads to residential mobility, often to a worse neighborhood. Divorce instills complicated lessons about romantic relationships in kids (the subject of my first book, Understanding the Divorce Cycle: The Children of Divorce in Their Own Marriages). Divorce even has a genetic component that’s correlated with some of its negative outcomes.

Cornwall and Coltrane acknowledge a lot of this but do so selectively. They’re correct in pointing out that divorce can be beneficial to kids when it removes them from a high-conflict marriage—but that’s not the end of the story. Sociologists Paul Amato and Alan Booth found that only a third of modern divorces follow on the heels of substantial conflict. Yet Amato also found that children from high-conflict divorced families were less likely to dissolve their own marriages than are the offspring of low-conflict divorced marriages. Conflict is broadly bad for kids, but it nonetheless conveys a lesson about sticking with a marriage (and if you were wondering, the offspring of married and divorced families themselves have equally happy marriages). What’s more, some of the negative consequences of parental divorce have abated over the years as divorce became more common. In other words, it’s complicated.

Does parental divorce have a causal negative impact on offspring well-being? We know now the answer to this question is yes, based on modern econometric methods and sophisticated data resources like sibling data.

In one of their most fatuous passages, Cornwall and Coltrane offer this strange ode to some of America’s enduring ills: “Most of the problems associated with being a child of divorce are instead related to sexism, racism, homophobia, shoddy recordkeeping, and insufficient government support.” Allow me to rephrase this: divorce is only bad for kids because it reflects the major injustices in American history (with the baffling addition of shoddy recordkeeping). 

Never mind that same-sex marriage has been the national law of the land for less than a decade, or that there is research showing that kids raised by same-sex parents do just as well as kids raised by heterosexual parents, or that gay and straight marriages break up at the same rate (although the science on this last claim is less settled). 

Never mind that a huge meta-analysis showed that the effects of parental divorce are consistently much stronger for white kids than for their African American peers.

Never mind that Cornwall and Coltrane’s claim about sexism is fundamentally impossible to establish or falsify. Our long history of maternal custody and sexism are inextricably related, but it’s intellectually lazy to ipso facto conclude that female custody is a reason why people think divorce is bad for kids. Cornwall and Coltrane ignore the fact that it was fathers who generally got custody of kids prior to the 19th century. Nor are they concerned with the fact that child outcomes don’t vary between single-mother and single-father families.

Never mind that Americans disapproved of divorce for decades subsequent to World War II, years in which American record keeping appears to my layman’s eye to be passably competent. And data for studying divorce has only improved with time. Sociologists generally use national panel data sets like the Fragile Families and Child Wellbeing Study or the National Longitudinal Surveys, not the government surveys from the Centers for Disease Control and Prevention or the National Center for Health Statistics. Surely Cornwall and Coltrane know this.

Cornwall and Coltrane do get some things right. I’m with them on the subject of government support. For instance, McLanahan and Sandefur showed that fully half of the gap in high school graduation rates between adolescents from married and divorced families can be explained by lower incomes following divorce. The poverty rate of single-mother families has remained stubbornly high for decades, and this is ultimately a public policy choice. Consider the projected benefits of the 2021 American Rescue Plan. A team of Columbia University researchers anticipated a 31% decline in the overall poverty rate, from 12.3% to 8.5%, an all-time low. Social Security and Medicare had transformational impacts on elderly poverty. 

Still, when it comes to the effects of divorce, money isn’t everything. As I showed in Understanding the Divorce Cycle, income plays a big part in some of parental divorce’s short-run consequences, like educational attainment, but it doesn’t seem to matter that much in divorce’s long-term effects, like the intergenerational transmission of divorce.

I also agree with Cornwall and Coltrane about the excessive influence that the late Judith Wallerstein’s genuinely shoddy research has had on how Americans think about divorce. Social scientists have long known that Wallerstein’s sample of kids from dysfunctional families proves little more than the fact, in the words of sociologist Andrew Cherlin, that “troubled parents often raise troubled children.”

The breakthrough public attention to Wallerstein’s research, first published in 1980, makes more sense in retrospect. By 1980, the divorce rate had quickly soared to unprecedented heights, to the point where one out of two American marriages dissolved. Divorce had become ubiquitous in America, but neither social science nor popular discourse had caught up. Based on comparably primitive research methods and—perhaps—ideological blinders, the prevailing academic assessment at the time was that divorce had no consequences for kids. Meanwhile, American popular culture had come to valorize divorce in a way that is jarring to modern sensibilities—I don’t mean more tolerant of divorce, but literally pro-divorce. The best picture winner at the 1980 Academy Awards was the divorce saga Kramer vs. Kramer, in which Meryl Streep’s character is praised for having the “courage” to leave her husband and abandon her child because she found the marriage unfulfilling. No one really says things like that anymore. 

In this context, Wallerstein’s research—arriving before the quality scholarship of Sara McLanahan, Paul Amato, Andrew Cherlin, Mavis Hetherington, Robert Emery, and others—might have seemed like a breath of fresh air. That doesn’t mean we should still be talking about it 40 years later.

And to get back to my earlier question, does parental divorce have a causal negative impact on offspring well-being? We know now the answer to this question is yes, based on modern econometric methods and sophisticated data resources like sibling data (see this, this, this, and one of my own studies). The effects vary and have many causes, but the evidence for them is overwhelming.

The premise of Cornwall and Coltrane’s intellectual history is fundamentally flawed. There is no logical reason why concern with enduring sexism, racism, and homophobia means we must reject decades of science on the relationship between family structure and offspring well-being. There’s just no reason at all why these are mutually exclusive positions. The world is a complicated place. A more honest depiction of divorce discourse might acknowledge that.

Nicholas H. Wolfinger is Professor of Family and Consumer Studies and Adjunct Professor of Sociology at the University of Utah. He is the author of Thanks for Nothing: The Economics of Single Motherhood since 1980, coauthored with Matthew McKeever, forthcoming from Oxford University Press. Follow him on Twitter at @NickWolfinger.

Editor's Note: The opinions expressed in this article are those of the author and do not necessarily reflect the official policy or views of the Institute for Family Studies.