Why do richer and better-educated Americans tend to get and stay married while poorer and less educated ones do not? There are doubtless many factors in play, from economic ones like the decline of stable blue-collar jobs to cultural ones like the view of marriage as a capstone available only after attaining great financial security, as well as fatalistic attitudes and declining social trust among working-class Americans.

Economists Shelly Lundberg and Robert Pollak propose one more factor in a recent article: different patterns in childrearing. They argue that varying childrearing strategies, “rather than being an unintended byproduct” of class gaps in marriage and parenthood, actually drive people’s decisions about marriage. Here’s their one-paragraph summary of childrearing differences:

Parents with more education and income spend substantially more time with their children and also spend more money on them. These differences in investments of time and money mirror socioeconomic differences in parenting practices and attitudes that have been documented by psychologists and sociologists. Middle and upper-class parents are typically very involved in their children’s activities, and engage in what sociologist Annette Lareau characterises as “concerted cultivation” of their children. Working class and poor parents, in contrast, allow for the “natural growth” of their children. Their restricted resources lead poor parents to value “survival over achievement.”

That these differences exist and affect marriage and parenthood decisions seems indisputable. But might the arrow of causation between childrearing strategies and relationship choices go both ways? In addition to childrearing methods affecting the decision to get married, that is, deciding whether to marry could affect childrearing methods.

It seems intuitively obvious that a mom with a decent husband or supportive partner is better off than a single mom: for a mother, a partner means not just a second income but also emotional support and parenting help. When a mom becomes tired or impatient, her partner can take over the kids. With such support, a mom may be able to establish better relationships with her children and (ideally) spend her time with them engaging in “concerted cultivation” rather than, say, spanking them out of anger and frustration.

Empirical data provide some support for this theory. To take just one relevant area, let’s look at maternal depression. As analysts at Child Trends have written, researchers have found “again and again” that “low-income parents, especially single mothers, have higher rates of depression and depressive symptoms than their higher-income counterparts.” A 2003 study that controlled for mothers’ income, education, and age showed that single moms’ higher levels of stress and lower social support played a major role in their overall depression rate (which was more than twice that of married mothers). After controlling for various measures of stress and social support in addition to the factors mentioned before, the researchers found that single mothers were still almost 50% more likely than married mothers to have experienced an episode of depression in the past year.

How is all of this relevant? Depression, unsurprisingly, has negative effects on parenting. To quote Child Trends analysts again:

Parents who suffer from depressive symptoms are less likely to have feelings of self-efficacy and engage in positive parenting behaviors. Data from the Fragile Families and Child Well Being Study shows that mothers with persistent depression invest less time with their children on positive activities such as reading, outings, trips to the park, and indoor play.

That last list of activities is what “concerted cultivation” (active, high-investment parenting) looks like, at least in early childhood. And there’s more:

Multiple studies have shown that children with depressed mothers are more likely to have behavior problems, poor academic performance, and delays in cognitive and social development. Not only have studies shown links between parents struggling with depression and increased child injuries and visits to the emergency room, but a recent study conducted in the Bronx suggests that mothers with depressive symptoms are two and a half times more likely to have an overweight or obese child.

In short, single moms face a greater risk of depression, and depression typically makes it harder to be a good parent. Some children suffer worse outcomes as a result. In this case, it’s not just low-investment parenting strategies causing women to have children outside of marriage; it’s the stresses of single motherhood contributing to depression, which in turn leads, on average, to less successful parenting.

In the relationship between parenting investment and marriage decisions, then, causation seems to run both ways. Getting married wouldn’t be a solution to all single mothers’ problems, but at the very least, it looks like getting married prior to having kids protects mothers against depression and thus helps them be better parents. Policymakers should keep these things in mind as they seek to close the parenting gap between high- and low-income Americans.