CityViews: Housing Programs Must Fight Segregation and Displacement with Different Weapons

Earlier this month, as part of our Building Justice series, author Craig Gurian criticized New York City’s longstanding practice of reserving a large portion of the apartments in each affordable-housing development for residents of the local community district, a policy that he believes reinforces segregation. Here, a well-known housing scholar writes with a different view:

The community preference analysis that Craig Gurian presents in “Building Justice: How Community Preferences Enforce Racial Segregation in New York City” raises important questions, but requires a more complex analysis than it’s given there. Greater clarity is not nit-picking, but has important political and policy consequences.

To begin with, the problem is deceptively stated, and uses the term “segregation” ambiguously, with misleading consequences. Gurian’s statement that “community districts are typically marked by high levels of segregation” opens the door to a frequent misunderstanding. Community districts are indeed marked by high levels of spatial clustering, clustering by various characteristics. But not all forms of clustering are alike, and not all constitute segregation, and the differences matter.

“Segregation,” as used throughout most of the Building Justice discussion, refers to a particular form of spatial clustering: generally, clustering by race and national origin, and/or color, which of course overlap with income. It implicitly means involuntary clustering, and clustering that is harmful to residents with particular characteristics. It is undesirable clustering that has been created and is part of present or past discrimination against and to the detriment of, oppressed groups.

Ingrid Gould Ellen, in the same series, uses more precise terminology:

“…ethnic concentration is not inherently harmful, but segregation can affect individual outcomes by constraining residential options and shaping the resources and services available in the neighborhoods in which different groups live.”

Segregation is then the result of the constraining of an ethnic concentration into a bounded space. Not all ethnic concentrations are constrained. Some are deeply desired, if for quite a variety of differing reasons. Even some segregated communities may morph into enclaves, felt and valued as homes by their residents.

New York’s 59 community districts are in effect legally defined by their differences from each other, with the formal intent of defining coherent neighborhoods—clusters of residents and activities with rough boundaries within which residents feel they have much in common with other residents. Their names are historically evolved, and run from Lower East Side to Upper East Side, to Harlem, to Stuyvesant Town, to Greenwich Village, to Battery Park City, to the Rockaways, to Bay Ridge to Crown Heights. They are by no means all segregated communities; in many, their separate identity is desired, not constrained—largely a legitimate matter of pride and desirable social solidarity.

Answers that treat all communities, including all ethnic communities, alike—and therefore might seek to ban or permit community preferences in all alike—do not make sense if the objective is to reduce segregation, properly understood.
It is one failing of the blanket opposition to neighborhood preferences in inclusionary housing that such opposition assumes implicitly that all residents of minority-majority clusters wish to leave them. Many do not.

Many who want to leave feel that way not because of a lack of affection for their own neighborhoods but despair about conditions there—poor schools, distrust of the criminal justice system, less effective garbage pickup than is needed, a preponderance of harassing landlords. Yes, they could be hurt if community preferences on affordable housing developments in better-off communities prevented them from accessing those new apartments. Equalize the quality of their community with others that are better off, however, and they might well prefer to stay put, and would want a community preference for new developments in their neck of the woods. It is the old dilemma: when to escape, and when to stand your ground (in a good sense!).

Public policy should not treat all communities alike. Certainly, it is desirable to treat communities that are segregated (in the definition used here and generally) as undesirable, and to treat them differently from those free of segregation. In the lawsuit challenging the neighborhood preference provisions of the city’s inclusionary housing programs, Black residents of truly segregated districts wanted to have the opportunity to move to Murray Hill, for example, and argued that providing a neighborhood preference in Murray Hill decreased their chances of moving here and thus reinforced segregation. And they are right. But would a community preference in, say, a gentrifying community also reinforce segregation? Or would it not rather give that community’s older residents the opportunity to benefit from improvements made in it? The statistics might reflect more racial clustering, but it would be voluntary, not the result of constraint.

There are two problems with a blanket ban on community preferences. The first is that it would undercut current residents’ struggles for improvement, and endanger their abilities to benefit from improvements that they have fought for.

Further, a blanket ban detracts from solidarity in really attempting to solve the problem of segregation. It pits residents constrained by segregation in one community against those similarly restrained or even worse off in another. Remember, community preferences only deal with who among all those needing affordable housing will get it; they don’t expand the supply of affordable housing, only juggle who among those needing it will get it. With community preferences in better off communities poor people in those communities will have an advantage over poor people in less well-off communities. Without such preferences, poor people in worse-off communities will face competition for affordable housing from others in even more neglected communities.

Solving the problem of segregation will require more than simply saying yes or no to community preferences in housing policies. Such preference policies may indeed accentuate segregation in some communities, as the lawsuit properly argues. But in other communities, preferences are needed for those fighting the ill effects of segregation.
The fight against segregation is an important one. The evidence is clear that segregation is grossly unfair in its impact on living conditions in segregated areas; Gould Ellen’s cited article spells out the differences. But while desegregation is likely over time by itself to produce improvement in living conditions in segregated areas; that result is not automatic; it must be accompanied by policies in the desegregating neighborhoods for housing improvements, school improvements, criminal justice improvements, educational improvements, environmental improvements, and so on. The goal is not simply desegregating, but reducing inequality, in all the conditions of life, in all communities for their present residents adversely affected by that inequality, wherever they live. It must thus further provide for at least minimum level of support for all the disadvantaged residents of all parts of the city, as a right.

To reach that broad goal of desegregation, the handling of community preference issues in the inclusionary zoning and affordable housing programs must move to a higher level of planning, dealing with housing and community development problems in all parts of the city in a comprehensive fashion. In segregated areas, policies must involve comprehensive improvements both in capital investments and in services.

A fundamental solution requires legislation that will not easily be adopted by the relevant governmental bodies. It must start with the acknowledgement that different neighborhoods in New York City, certainly at the community district level if not below, require differential treatment.

There is a model, if at the federal level, of exactly what is required: the Voting Rights Act of 1965. It separates voting districts into those with and without “significant histories of voter discrimination,” and prescribed remedies specifically for those with such histories. It used a statistical measure of “significant,” a parallel to which might be easily conceived for purposes of measuring segregation. The model of a social impact statement (itself modeled on environmental impact statements, with their frequent requirements of public hearings and other steps) might be adopted as one further way of identifying districts with a significant history of segregation and evaluating policies of public action affecting those districts, including not only investment and budgeting, but also regulations and discretionary actions, like zoning decisions or educational policies.

A Housing Rights Act might be exactly what is needed. And its principles might be the guide for a whole range of local, regional, and state actions. Where the litmus test under the Voting Rights Act is: “does the policy facilitate or limit the right to vote of those historically constrained in its exercise,” the litmus test under a Housing Rights Act might be, “does the policy facilitate or limit the ability to exercise choice among adequate alternatives for housing and neighborhood?”

Blanket prohibition of the use of community preferences in affordable housing programs would deny many residents, particularly those in neglected communities striving to improve their communities, of the choice of whether to stay and improve or escape from such communities. Donald Trump has been severely criticized for his implicit assumption that all residents of majority-minority neighborhoods are living in undesirable neighborhoods and wish to leave them, when in fact the establishment of a “right to stay put” is becoming a rallying cry in many communities, particularly in those threatened by displacement from gentrification. The generally accepted desirability of community self-determination would likewise weigh against flat prohibition of neighborhood preferences in all communities.

The details are indeed complex, and pose some tricky questions, and the nuanced handling of community preferences in affordable housing are among them. But the principles should be clear. Maximizing differentiated, democratic and just community self-determination, in a thoughtfully considered city wide context, should be among them.

Peter Marcuse is a professor of urban planning at Columbia University.