Let me start with a confession: during the decade or so I’ve been studying family breakdown, I’ve sometimes understated the impact of neighborhoods on children’s chances in life. That’s a mistake. The point should have been obvious: the most conscientious, Good Night Moon-reading, homework-supervising, PTA-joining, stably married mom and dad living in, say, East New York or Detroit are still going to have a harder time guiding their kids through the complex demands of contemporary childhood and adolescence than a married couple living twenty or thirty miles away in Great Neck or Grosse Point. Neighborhoods matter. They matter a lot.

That truth has a tragic racial dimension. Among the most persistent legacies of America’s racial history is residential segregation. To this day, African Americans are far more likely to grow up in high-poverty neighborhoods than white Americans. That means that as a group blacks not only go through life with less money, but also have more limited political influence, poorer public services, more crime, and fewer schools offering a path to a better life.

These facts are sobering enough on their own, but now a number of scholars who study “neighborhood effects” are warning that individual urban communities have more staying power than the familiar image of a mobile, ever-modernizing America would seem to suggest. In Great American City, University of Chicago sociologist Robert Sampson shows that even as residents arrive and leave, are born and die, neighborhoods tend to maintain an essential identity. For better and worse, local institutions persist; lousy schools stay lousy schools, better schools remain successful, active community organizations become institutionalized for the long run, and residents demonstrate “collective efficacy”—or not—over generations.

Neighborhoods tend to maintain an essential identity—and even imprint themselves on residents who leave.

Due in part to lingering discrimination and in part to a human preference for community cohesion and rootedness, that persistence has been especially notable in black areas. African Americans in particular are “stuck in place,” as Patrick Sharkey puts it in a recent book of that title. Black poverty and inequality are inter-generational, notes Sharkey, largely because they are rooted in black neighborhoods, neighborhoods that are generally poor and lacking in public investment as well as social cohesion. In fact, poor neighborhoods tend to imprint themselves on those residents who leave: African Americans who grew up in such areas have a smaller chance of producing upwardly mobile children regardless of their individual level of education and even if they have moved to a better area. The “racial mobility gap”—the fact that black children are much less likely than white children to move into a higher income quintile than they grew up in—is inextricably tied up with the prevalence of black neighborhood disadvantage.

So, yes, I stand corrected. But the literature about “neighborhood effects,” at least as represented here by Sampson and Sharkey, stumbles on its own priorities. By using the neighborhood, rather than the individual or family, as the window into disadvantage, it minimizes the significance of children’s early, largely home-centered experience, not to mention the depth of their psychological ties to their mothers and fathers. In Sharkey’s work in particular, families are portrayed as just another environmental characteristic like housing stock or crime rates. That understates the family’s importance not just to an individual child but to the community as a whole.

Because families are the primary locus of children’s socialization, they remain a significant generator of a community’s behavior patterns. Everything we know about child outcomes supports the conclusion that single-parent households are at greater risk than those of married couples of producing children who will make less-than-ideal students and neighbors. You can make a case that the 80% and more of children in poor, black communities being raised apart from their fathers and amid frequent domestic disruptions caused by the arrival and departure of stepfathers and siblings is a consequence of living in impoverished, segregated neighborhoods. But it’s hard to deny that fractured families also cause poverty to endure in the next generation.

Neighborhood effects are inextricably tied to family effects. And vice versa.

That families define neighborhoods is one take-away from another recent study by Raj Chetty and his co-authors at the Equality of Opportunity Project, which W. Bradford Wilcox wrote about earlier this week. The researchers compared social mobility in various geographical regions. They looked at several characteristics of the residents of the areas under investigation: racial makeup, education levels, incomes, and so forth. But of all the characteristics they considered, a region’s dominant family structure had the strongest correlation with local children’s chances of moving up the income ladder—stronger than a region’s racial composition and stronger than its education levels. In fact, areas with high proportions of single-parent families had less mobility even for kids whose parents were married. The reverse was also true: areas with a high proportion of married-couple families improve the lot of married couples’ own kids as well as the kids of their single-parent neighbors.

In other words, neighborhood effects are inextricably tied to family effects. And vice versa.