Home Owners Loan Corporation's "redlining" map of Chicago. (Courtesy of Urban Oasis)

Nowadays it is increasingly rare to encounter studies of residential segregation and racial inequality in the United States that do not reckon with the history of discrimination. Still in many scholarly settings and popular venues, our debates concerning urban change and opportunity are distorted by a powerful myth about the places that Americans call home: namely, that patterns of residential development have been driven, above all else, by the preferences of individual housing consumers, or even of entire generations of such consumers. This myth has consequences because it obscures the powerful institutions that have shaped metropolitan landscapes and created opportunity for some while systematically denying it to others. Of course preferences matter for understanding U.S. history, but individual preferences alone did not draw our urban and suburban maps. Nor did they alone determine winners and losers in the markets for residence and community resources.

I was reminded of this myth’s endurance by a recent piece in Wonkblog discussing new economics research about, in the words of one of the study’s authors, the “emergence of segregation” in northern cities. Allison Shertzer and Randall P. Walsh have done remarkable work compiling and digitally mapping census information to produce “the first systematic analysis” of white households’ “relocation decisions” between 1900 and 1930. The authors seek to gauge the “relative influence” of two variables that contributed to segregation: “white flight” from racially integrated neighborhoods, which they also call “population sorting”; and “institutional barriers constructed by whites” or “collective white action” (such as vigilantism or adoption of race restrictive covenants) to prevent blacks from settling there in the first place. The data shows that whites left integrated neighborhoods during this era at an increasing rate, leading the authors to conclude that “flight” was so statistically significant that institutional racism was probably not, in the final analysis, decisive. “Segregation would likely have arisen even without the presence of discriminatory institutions,” they write, and then suggest the policy implications. “White flight from black neighborhoods is an individual behavior that cannot be limited by local or federal agencies.”

Here is the problem: This bold claim is challenged by an enormous body of historical evidence showing that Americans’ decisions about residence have rarely if ever been “individual behaviors” that are separable from a host of social and institutional contexts. Why, then, do arguments about the power of individual housing “preferences” continue to carry such weight?

Part of the answer lies in our political culture’s celebration of the free market for homeownership, despite the fact that American property markets have never been wholly “free.”  Of course most people are aware that racial separation and discrimination are longtime features of American life, and they are familiar with notorious (usually Southern) actors in this story, such as George Wallace or Bull Connor. Yet most people are surprised to learn the mundane details of residential exclusion, because it is a history not simply of racist mobs and deed restrictions but also powerful institutions and public policy.

Throughout much of the 20th century, discrimination by race was integral to the design, development, marketing and even financing of American cities and suburbs. Discrimination was sanctioned and aggressively promoted by real estate boards, neighborhood associations, municipal governments, state and federal courts, mortgage lenders, and a host of federal housing and development programs.  Together they helped to draw sharp neighborhood boundaries, deny equal access to markets and places, and produce obscene disparities in wealth, opportunity and basic quality of life.  Our contemporary urban and suburban landscapes continue to reflect that history and seldom a day passes when we are not reminded of its legacies.

There are some signs that the “free market” myth of American residential segregation is showing its age, in part because activism and journalists’ interest has introduced more people to the nuances and geographic scope of discrimination’s institutional history. For example, more people are now familiar with things like the "redlining” maps  commissioned in the 1930s by a U.S. government agency called the Home Owners Loan Corporation (HOLC). The maps promoted residential development by indicating which neighborhoods and potential borrowers were eligible for the generous mortgage assistance programs of the Federal Housing Administration and the Veterans Administration. And by carefully recording neighborhoods’ ethnic and racial makeup, they provided literal road maps for the realtors, lenders, and public officials who oversaw and financed decades of racial exclusion in the American housing market. Federal mortgage programs made homeownership affordable for the majority of households, for the first time in the nation’s history. But because the programs refused to support lending in racially integrated neighborhoods and in most minority neighborhoods, the lion’s share of benefits after World War II went to whites, who became the primary homeowning class in the nation’s fast-growing suburbs.

Significantly, the HOLC popularized institutional practices that had been widely embraced prior to the Great Depression, in the decades covered by Shertzer and Walsh’s new study. The federal government did not formally get out of the discrimination business until passage of the Fair Housing Act in 1968 and, since then, enforcement has been uneven at best.

But “free market” mythologies aside, another question remains. Why does a paper in economics suggest that the institutional story is separable from that of whites’ “individual” preferences?  The answer lies partly in social scientific method. When economists describe markets, for example, they treat causal factors (including human motivations) as variables that can be captured with metrics. In Shertzer and Walsh’s study, a white person’s choice to leave an integrated neighborhood — the “relocation decision” — counts as evidence of one thing: personal racism. The fact of moving, alone, is a “gauge of racial distaste” that is measured for its frequency, yet divorced from other forces that might have influenced these decisions. Because the methodology requires that “institutional” racism and “personal” racism are tracked separately, the latter — here recorded as “changes in white animus” — registers as a separate force that is statistically significant for explaining racial segregation.

In this approach, only quantifiable variables count. And access to new data and digital mapping tools, we are told, provide the authors with unprecedented “capacity to rigorously investigate the mechanisms responsible for the emergence of segregation.”  The lesson here is that creative data-mining and a proven social scientific method provide new insight into the past that can help us to solve an historical puzzle. They ask readers to accept the premise that scholars can fully disaggregate and scientifically weigh the forces that shaped individuals’ market behaviors.

My concern is that such research encourages readers to question the “rigor” of work that rejects this premise and thus tells a very different story about housing and racial segregation. I fear that readers might dismiss the work of scholars who contend that data can often be a blunt instrument for capturing the range of  ideasopportunities and constraints that influence people's housing choices in the United States.

Just consider some of the forces shaping whites’ “relocation decisions” in the early 20th century that escape measure by census data. Census data does not measure for the 1924 requirement by the National Association of Real Estate Boards that members refrain from introducing into a neighborhood a member of “any race or nationality” that would “clearly be detrimental to property values.” Nor does it measure the influence of “blockbusting” by realtors, or local neighborhood publications warning whites that integration would defy the “laws of nature” or state and federal court rulings that upheld the right to exclude on racial grounds. These are only a few of the examples suggesting the limits of reading intent or motive purely from independent and quantifiable variables.  Even the most sophisticated quantitative methods and digital mapping projects need to fully engage the story that we already know.

David M. P. Freund, associate professor of history at the University of Maryland at College Park, is the author of "Colored Property: State Policy and White Racial Politics in Suburban America" and "The Modern American Metropolis: A Documentary Reader."