Some Thoughts on Segregation, Discrimination, and Family and Community Ties

by Reihan Salam

Understanding segregation is essential to understanding the opportunity gaps that have such a profound effect on American life. A growing body of work, which we often discuss in this space, finds that social networks are the main conduits through which people acquire information about economic and educational opportunities, and through which cultural and social capital is transmitted. When we ban formal discrimination in the labor market or in housing, we don’t necessarily break patterns of in-group favoritism that tend to reinforce the dominant position of the dominant social group.

If we want to address opportunity gaps, we have to play a short game and a long game. The short game involves meeting the immediate needs of the poor and the marginalized. For example, how can we raise the quality of educational options available to children living in neighborhoods with high poverty concentrations, and how can our institutions better serve the interests of low-wage workers? The long game is about integration and inclusion.

In-group favoritism may well be a durable fact of life, but the definition of the in-group can change and grow more expansive over time. Yet this process works in complicated ways. The boundaries of the in-group are determined by the perceptions and the decisions of millions of individuals and families — e.g., who is and is not an appropriate choice for a neighbor, friend, or spouse? Perceptions of who is and is not an insider can change at a glacial pace, or they can move like a cascade. This process isn’t always or even often susceptible to policy intervention; indeed, policy interventions in this space often yield surprising, and sometimes quite perverse, consequences. Policy choices can, however, have an effect over very long periods of time. Laws and norms are not one and the same, but they do influence each other. And if the tendency towards assortative mating proves stronger than (say) our tendency to associate with people from similar cultural backgrounds, investing in the human capital of members of marginalized groups will help them gain access to the mainstream.

In the United States, the expansion of the dominant social group is most vividly illustrated by the evolution of who is and is not considered white, a subject that forms the basis of the academic subfield of “critical whiteness studies.” The banal point is that there was a time when various European-origin communities were considered separate and distinct from, and inferior to, a dominant Anglo-Protestant ethnocultural group, yet these communities came to be included in a broader category of American whites, due in part to high rates of intermarriage, the assimilation of Anglo-Protestant norms by immigrants and their descendants, and to some degree the embrace of cultural practices introduced by the relative newcomers.

Some argue that we are seeing a similar process as Americans of Latin American and Asian origin are incorporated into whiteness broadly (very broadly) understood, though I see this as an oversimplication. Ethnic attrition, in which members of minority groups cease to see themselves as members of these groups, is a real phenomenon, yet it is an uneven phenomenon. It is not relatively rare for people with only one Mexican-origin grandparent to identify as Latino while it is relatively rare for people with three Mexican-origin grandparents not to do so. Intermarriage rates, which contribute to ethnic attrition, are much higher for immigrants and second-generation Americans with levels of educational attainment that match or exceed the average for native-born Americans as a whole. And there is tentative evidence that some national-origin groups are more likely to disaffiliate themselves from Latino identity than others. It is also worth noting that some Asian American subgroups are less educated and affluent than others, and this also has bearing on the integration process across groups.

The story for African Americans is notably different from that of Latinos and Asian Americans. More than 8 percent of African Americans are foreign-born, and the second-generation share of the black population is larger still. It is still true, however, that a large majority of black Americans are the descendants of enslaved Africans brought to what is now the United States, which is to say this population is deeply-rooted in American life, yet it has been excluded from the dominant social group for a very long time. Integrating African Americans into the dominant social group is a profound challenge — some see it as an insurmountable challenge, particularly if we believe that the expansion of the dominant group through the incorporation of successive waves of immigrants is actually tied to the continued exclusion of blacks. Another way of approaching the issue is that the dominant social group is not best defined in racial terms, as there are marginalized whites as well as marginalized blacks, and that decades of upward mobility have meant that many African Americans are members of privileged yet diverse social networks. The problem is that this process hasn’t gone far enough.

One of the best books I’ve read in ages is Patrick Sharkey’s Stuck in Place, which I’ve referenced on several occasions over the past year. Sharkey, a sociologist at New York University who is very much a man of the left, emphasizes the importance of the intergenerational transmission of social outcomes. He argues that to understand the persistence of black poverty is closely tied to the fact that the adverse outcomes associated with poverty are reinforced from generation to generation. His most striking finding is that children raised in nonpoor neighborhoods raised by parents raised in poor neighborhoods fare roughly as well as children raised in poor neighborhoods raised by parents raised in nonpoor on cognitive tests, and that both groups of children far better than those raised in poor neighborhoods by parents raised in poor neighborhoods and far worse than children raised in nonpoor neighborhoods by parents raised in nonpoor neighborhoods.

Is is thus profoundly significant that African Americans are more likely to live in poor neighborhoods than non-blacks at every income level. That is, middle-income blacks and far more likely to live in poor neighborhoods than middle-income whites.

As Harvard economist Edward Glaeser and Duke economist Jacob Vigdor have observed, racial segregation declined in the first decade of this century. They note that while in 1960, half of all black Americans lived in neighborhoods with an African-American share above 80 percent, the same was true of 20 percent of blacks as of 2010. (I’d be curious to see what would happen if we separated foreign-born and second-generation African-Americans from the rest of the black population.) They also found that segregation declines sharply as levels of educational attainment rise among African Americans, but of course this is fully compatible with Sharkey’s analysis.

So why would middle-income and even upper-middle-income black Americans prefer to live in high-poverty neighborhoods. In a recent discussion of Sharkey’s work, Jamelle Bouie of Slate offered a partial hypothesis:

Simply put, the persistence of poor neighborhoods is a fact of life for the large majority of blacks; it’s been transmitted from one generation to the next, and shows little sign of changing. All of which raises an obvious question: Why do blacks have a hard time leaving impoverished neighborhoods?

“When white families advance in economic status,” writes Sharkey, “they are able to translate this economic advantage into spatial advantage by buying into communities that provide quality schools and healthy environments for children.” The same isn’t true for black Americans, and some of the answer has to include present and ongoing housing discrimination. For example, in one study—conducted by the Department of Housing and the Urban Institute—black renters learned about fewer rental units and fewer homes than their white counterparts.

Once you grasp the staggering differences between black and white neighborhoods, it becomes much easier to explain a whole realm of phenomena. Take the achievement gap between middle-class black students and their white peers. It’s easy to look at this and jump to cultural explanations—that this is a function of black culture and not income or wealth. But, when we say middle-class black kids are more likely to live in poor neighborhoods, what we’re also saying is that they’re less likely to have social networks with professionals, and more likely to be exposed to violence and crime. [Emphasis added]

While I don’t disagree that discrimination might play a role, and perhaps even a significant role, I’m not sure that this is the most important part of the story. Two thoughts immediately come to mind. The first is that there is some work, from Patrick Bayer, Hanming Fang, and Robert McMillan, that residential segregation might actually increase as income differences across groups grow more equal:

This paper introduces a mechanism that, contrary to standard reasoning, may lead segregation in U.S. cities to increase as racial inequality narrows. Specifically, when the proportion of highly educated blacks rises holding white education fixed, new middle-class black neighborhoods can emerge, and these are attractive to blacks, resulting in increases in segregation as households re-sort. To examine the importance of this ‘neighborhood formation’ mechanism in practice, we propose a new two-part research design that yields distinctive cross-sectional and time-series predictions. In cross section, if our mechanism is important, inequality and segregation should be negatively related for older blacks, as we find using both the 1990 and 2000 Censuses. In time series, a negative relationship should also be apparent, particularly for older blacks. Controlling for white education, we show that increased black educational attainment in a city between 1990 and 2000 leads to a signicant rise in segregation, especially for older blacks, and to a marked increase in the number of middle-class black communities, consistent with neighborhood formation. Of broader relevance, our findings point to a negative feedback loop likely to inhibit reductions in segregation and racial inequality over time.

(I intend to write more about Bayer et al.)

You would think that highly educated blacks would be less likely to face housing discrimination, yet it seems that a not inconsiderable number of black Americans prefer living in heavily-black communities when other options are available to them. As the number of affluent black households in a metropolitan area increases, it is possible to live in a black neighborhood that is not disproportionately poor, and so living in an integrated or heavily-nonblack neighborhood is not the only way to avoid some of the downsides of living in a poor neighborhood. This phenomenon has bearing on the diversity of social networks, which has bearing on the workings of in-group favoritism. Yet it is not obvious to me that this is a tendency that we should necessarily seek to counteract through policy intervention. Might at least some middle-income blacks who live in poor neighborhoods actually be choosing to do so? And if this will damage the educational and labor market outcomes of their children, should we have a public education campaign encouraging them not to do so? Once we introduce the possibility that housing discrimination is not the whole story, things suddenly get a lot more interesting and difficult.

And finally, consider one possible reason middle-income African Americans might prefer to live in heavily-black neighborhoods even if the neighborhoods in question are quite poor. Upward social mobility can be more difficult to achieve in some cultural communities in others, e.g., communities in which it is expected that successful family members will provide assistance to less-successful family members will also tend to be communities in which it is harder for successful individuals to accumulate assets. Sharing what you have can make it hard to save. At the same time, those who achieve upward mobility by severing social ties to loved ones might find that when they experience a crisis, whether economic or interpersonal, etc., they don’t have people who will be willing to lend them a hand. Different families manage this tradeoff in different ways, and it seems plausible that at least some black Americans are choosing to remain rooted in black communities not because of housing discrimination, but because they are mindful of the importance of maintaining strong family and community ties in the face of uncertainty. This story resonates with me, as I have many loved ones who’ve been plagued by mental illness.