It's clear in America that family structure and poverty are intertwined: Nearly a third of households headed by single women live below the poverty line. And just six percent of families led by married couples are in the official ranks of the poor. Poverty, meanwhile, touches an astounding 45 percent of children who live without a father.

Recent research by Raj Chetty, Nathaniel Hendron, Patrick Kline, Emmanuel Saez and Nicholas Turner also found that intergenerational income mobility was lower in metropolitan areas with a larger share of single mothers, a bold-faced finding that touched off a new round of public debate over what this relationship means.

The researchers' findings on this front are summarized in this chart posted by Atif Mian and Amir Sufi on their House of Debt blog:

Atif Mian and Amir Sufi,

Each dot on that chart represents a single metro area, with local upward mobility clearly declining as the share of single mothers rises. "The statistical power of this one variable is stunning," Mian and Sufi write, in considering the link between family structure and income mobility (or inequality, as they frame it).

But they proceed with caution (bold emphasis below is mine):

Now, we are wading into some hugely controversial issues here. So it is crucial to be scientific about what this figure means. This is a correlation; it is not necessarily evidence of causation. It doesn’t mean that if we somehow paired up single mothers with a partner, then suddenly inequality would go down. Further, the channel through which inequality and family structure are related is complex. For example, the authors show that cities that have a large fraction of mothers that are single have less upward mobility even for families where both parents are in the home. So again, the channel is unlikely to be a direct effect of single mothers on inequality.

It's tempting to look at the tidy pattern above and conclude that the breakdown of marriage causes poverty, perpetuating inequality. As Mian and Sufi point out, your take on the causal direction here likely depends on your ideology. Conservatives have used this data to argue for the importance of marriage in lifting women and children out of poverty, as if the shortage of stable husbands were the source of the problem rather than one of its side effects.

Liberals argue, on the other hand, as sociologist William Julius Wilson has, that economic forces much larger than families have decimated urban communities and robbed many men of one of the prerequisites to a stable marriage: a job. By this thinking, the circumstances of poverty itself strain families. To suggest the opposite -- that frayed social values drive poverty -- is to point toward all of the wrong policy prescriptions, this argument says: Incentivize marriage. Teach family values. Just convince poor people to get married.

Mian and Sufi argue for more economic analysis of the "causal chain" here. But in the meantime, let's look at the explanation that, at the very least, the same underlying conditions may be creating single mothers and low social mobility in a city like Atlanta. That chart could easily be embraced by those convinced that "culture" is the problem here, that we should blame the prevalence of single mothers -- or the absence of fathers -- for the fact that poor children have little chance to move up in the world in some communities.

But that argument ignores what marriage might actually look like to a woman living in a neighborhood with high rates of poverty, unemployment and incarceration. It's true that marriage can bring stability and emotional benefits to the children of middle- and upper-class families. But that's not because the institution of marriage itself is universally beneficial. It's because certain kinds of marriages are beneficial, such as those between adults who don't have to worry about getting evicted, who can afford to pay their medical bills, who don't contend with the surrounding stresses of violence or joblessless or having to get to work without a car.

"The problem," Ohio State sociologist Kristi Williams recently told me, "is that there’s no evidence that the kind of marriages that poor, single parents enter into will have these same benefits."

Consider it that way, and a woman may not get married for reasons that have nothing to do with how much she values  the institution. A man may not get married for reasons that have nothing to do with whether he believes in a "culture of marriage." In isolating culture and values as the source of the problem, we can miss the other fundamental ways in which poor households without married parents differ from better-off households with married parents.