by Alexandra Killewald
Through the second half of the 20th century, American women participated in the labor force at increasing rates. Divorce rates also rose rapidly in the 1960s and 1970s, raising concerns that the trends were related: perhaps marriages became less stable because women were no longer dependent on men for their financial well-being.
But women’s economic independence from their husbands isn’t the only way that money, work, and divorce could be associated. Another possibility is that couples who have few financial resources are more likely to divorce, maybe because they argue more about how to spend those limited resources.
Yet another possibility is that divorce is more likely when spouses don’t fulfill what’s socially expected of husbands or wives. Those expectations might include that the husband is employed full-time or that the wife takes primary responsibility for housework. In this case, it isn’t money itself that affects the risk of divorce, but the work spouses do, both in the labor force and at home.
In a recently published study, I tested these theories and whether the predictors of divorce had changed over time for American couples. In particular, I wondered whether American couples had become more accepting of wives’ employment and of couples sharing responsibility for housework, so that these behaviors no longer increased the risk of divorce.
National data on marriage, divorce and separation
To answer this question, I used data on married couple from the Panel Study of Income Dynamics (PSID). Because the PSID follows households over time, I was able to examine how couples’ work and economic circumstances were associated with the risk that they would soon divorce.
I used samples of divorced and separated women from the decennial U.S. censuses and the American Community Survey to model the factors that make previously-married women better or worse off financially. I then used the results of these models to estimate the likely economic consequences of divorce for women who were currently married.
Work patterns and norms, not money, play a key role in divorces
Couples’ risk of divorce didn’t change much based on the household’s total income or how much the wife depended on marriage for her economic well-being. In other words, the money didn’t matter.
Work patterns, however, did matter, although the results depended on when the couple was married.
For couples married before 1975, marriages were more stable when the wife did more of the housework, suggesting that a more traditional division of labor was expected for these couples.
For the couples married in 1975 or later, however, whether the wife was employed full-time or not and how the couple divided housework weren’t significant predictors of the risk of divorce. For these couples marred since 1975, what did matter was whether the husband was employed full-time. For an otherwise typical couple, the risk of divorce in the next year was 2.5% when the husband was employed full-time, compared to 3.3% when he was not.
What can’t the study tell us?
I can’t estimate the consequences of husbands’ less than full-time employment separately for husbands who are involuntarily unemployed versus those who are at home by choice. It’s possible that involuntary unemployment is particularly disruptive for couples and that couples who deliberately choose a division of labor in which the husband takes responsibility at home instead of full-time employment do not experience heightened risk of divorce.
I also don’t have information to determine the process by which husbands’ less than full-time employment translates into heightened risk of divorce. It could be that unemployed husbands experience depression that negatively affects their marriages, or it could be that wives are frustrated when husbands don’t fulfill the norm of full-time employment, or perhaps couples with a husband not employed full-time receive less social support for their marriage from family and friends. I can’t adjudicate among these possibilities with my data.
Finally, the analysis includes only different-sex couples, because there are too few same-sex couples in the PSID to generate meaningful estimates for this group. Whether the risk of divorce for same-sex couples follows similar patterns as for different-sex couples is a topic for future research.
What does the study tell us about changes in the American family and the expectations of husbands and wives?
The gender revolution and the feminist movement have allowed women to take on activities that were previously closed to them. Women have surged ahead of men in college completion rates, and they make up about a third of doctors and lawyers. Of course, men have changed their behaviors some, too — taking on more responsibility at home — but the change has been more modest.
My results suggest that how much money a family has and how well a woman can support herself in the event of divorce are not key determinants of divorce. For couples married since the mid-1970s, divorce risks also don’t change much with the wife’s employment or the couple’s division of labor.
Instead, what has endured is the norm of male breadwinning; while wives can balance paid and unpaid work in a variety of ways without threatening marital stability, anything other than full-time employment for husbands is associated with increased risk of divorce.
Alexandra Killewald is Professor of Sociology at Harvard University. This article summarizes findings from “Money, Work, and Marital Stability: Assessing Change in the Gendered Determinants of Divorce” in American Sociological Review.
Image: Fabrice Lambert via Wikimedia Commons.