Stanford study identifies another explanation for the ‘marriage premium’

Stanford sociologist Florencia Torche uncovers empirical evidence to show that the benefits of marriage to child development derive not just from individual characteristics of spouses and their circumstances. How society views marriage as an institution matters too.

Something big is happening with babies in Chile – and it’s at the heart of a unique study into what it means to be born to married versus unmarried parents.

A new study co-authored by Stanford sociologist Florencia Torche investigates a key question about marriage’s effects on child development and society’s role in promoting them. (Image credit: Getty Images)

In Chile, the proportion of births among women based on their marital status has shifted dramatically in the last 30 years. More unmarried women today give birth relative to married women. While this trend has been happening in most of the Western world, the drop has been especially sharp in Chile.

This has allowed Florencia Torche, a professor of sociology in the School of Humanities and Sciences at Stanford University, and Alejandra Abufhele of the Universidad Católica de Chile to investigate a key question about marriage’s effects on child development and society’s role in promoting them.

Social scientists have long known that the offspring of married couples have significant, lifelong advantages over children born to unwed mothers, including better mental and physical health, higher levels of education and higher incomes. Research into this “marriage premium” has identified the multitude of individual differences – in race, socioeconomic status, personality, among other characteristics – to account for the discrepancies between children of married and unmarried parents.

But what about society’s role? When it comes to the marriage premium and child development, society’s influence has always been part of the conversation, but assumptions about its impact have not been proven.

“The question of the institution of marriage – how normative or accepted it is within society – and how it might affect the marriage premium has been curiously missing as a focus of research,” said Torche, whose research focuses on inequality and well-being across generations, including the effects of early life exposures to shocks such as natural disasters, armed conflict and crackdowns on immigration.

Torche and Abufhele saw in Chile a “unique and exceptional opportunity” to tackle this question head on. In a paper published April 14 in the American Journal of Sociology, the pair detail striking evidence that societal perceptions of matrimony also contribute to the marriage premium.

“As marriage has lost its normative status in Chile, the marriage premium for children also declined to the point where it fully disappeared,” Torche said. “Our analysis of that decline shows that the status of marriage in society matters.”

The researchers say their findings do not discount the role that parents and their marital circumstances contribute to the marriage – it’s just not the whole story. “Both factors matter,” Torche said. “Individual characteristics matter and the extent to which marriage is a norm in a society also matter.”

A novel and striking insight

In Chile, marriage shifted from norm to exception within a generation – a swift reversal of fortune that was also comprehensive and measurable: The proportion of births among married women plunged from 66 percent in 1990 to 27 percent in 2016. What’s more, in the early 1990s, babies born to married mothers had a considerable advantage over newborns from outside of marriage in that they were less likely to be of low birth weight, premature or small for gestational age. But by the mid-2010s, this advantage was negligible in the case of low birth weight, had completely disappeared for preterm birth and had declined by about two-thirds for small-for-gestational-age birth.

Florencia Torche

Florencia Torche (Image credit: Steve Gladfelter)

“We found that this change was not due to demographic or socioeconomic differences between married and unmarried mothers or to the increase in cohabitation,” Torche said.

This conclusion was further supported by results from two additional analyses the researchers performed: First, they looked at births by marital status and infant health across regions of Chile. In the second, they studied health outcomes among siblings whose mother was unmarried when delivering one child and married when delivering another.

The three analyses were designed to complement one another. “There is always a risk that there are differences between women who marry before having children and those who do not that we cannot observe in the data,” Torche said. The birth certificate data that Torche and Abufhele relied on didn’t have information on, for example, differences in personality or health conditions that might explain why the marriage premium disappeared over time. By analyzing siblings of the same mother born under different marital statuses, the authors ruled out these and other unmeasured characteristics.

The researchers found that, on all three measures, the results were similar: As views on marriage in Chile changed, newborns of unmarried mothers were on average as healthy as those of married women.

“By triangulating evidence of marital fertility and infant health over time, across place and within siblings, we offer consistent evidence that the prevalence of marriage in society also factors into the marriage premium,” Torche said.

According to Torche, the overall finding – that society at large can reinforce the marriage premium – is important for policymaking. Any group that is considered to be outside the norm, such as single parents or non-heterosexuals, may face stigmatization or even discrimination from family members, co-workers, neighbors and institutions. For unwed mothers, it can lead to higher levels of stress, which is known to harm fetal development, or feelings of shame that prevent them from seeking support. When that happens, the marriage premium gets reinforced.

“When creating and implementing social policies, we need to be careful that the non-normative characteristics or statuses of the people they are intended to help are not portrayed as a problem,” said Torche. “In addition to harming the individuals, it limits what policies can achieve.”