On Spatial Solutions to Social Problems

One of the persistent themes in the history of housing policy in the United States is that we almost never do housing policy for its own sake; that is, to provide or ensure the provision of decent home and suitable living environment (to borrow the famous language from the 1949 Housing Act1) for all people. Instead, housing policy has long been about using housing for other purposes, such as limiting the potential for political unrest in the tenements during the Progressive Era's reforms or absorbing surplus labor in the Great Depression with the initiation of federal public housing in 1937. In this way, the current emphasis on mixed-income housing fits into a long and not particularly glorious history. Mixed-income housing also fits into another, and comparably inglorious, history: that of trying to solve social problems by way of spatial solutions. In this brief commentary, I will discuss the goals of poverty alleviation and socioeconomic interaction and the failures of mixed-income housing policy to achieve its nominal goals.Poverty AlleviationA central theme in, and justification for, mixed-income housing has always been its potential role in poverty alleviation. I am thus heartened by the recognition from Diane K. Levy, Zach McDade, and Kassie Bertumen that, after their exhaustive review of the evidence thus far, mixed-income housing does not reduce poverty. Furthermore, the useful (for reducing poverty and increasing incomes) things within mixed-income housing developments are services and interventions that could be implemented regardless of mixed-income housing. As Levy, McDade, and Bertumen put it, If poverty alleviation is the primary goal, however, the path to it does not appear to require mixed-income living. In this, the authors echo Joseph and his colleagues' conclusions of several years ago from their meta-analysis of the evidence (see Joseph, 2006; Joseph, Chaskin, and Webber, 2007). Simply, very little evidence indicates that living in mixed-income housing increases the earnings or wealth of poor people. This recognition does appear to be the emerging consensus, and I am glad to see the change in perception. To be clear, I am not glad to see that poor people living in mixed-income developments have failed to realize improvements in their incomes or wealth. I am, however, glad that this view is being acknowledged as the consensus, because it means that perhaps we can finally stop talking about mixed-income housing as a solution to poverty. Because doing so is fundamentally a distraction from addressing the real ways in which poverty reduction happens.Poverty is not a function of the spatial configuration of poverty. The concentration of poor people in geographic space may well make the experience of poverty worse for some, and it certainly seems to lead to unjust outcomes such as poor quality services, schools, and amenities. But the root causes of poverty within capitalism are the organization of labor markets (and the structure and character of the demand for labor in those markets) and the centrality of wage labor to household income and wealth. Efforts to deal with poverty must, on some level, deal with this reality. At the risk of being overly reductionist, we can thus reduce poverty in one of two primary ways: (1) by transforming the operation of labor markets such that being employed in waged labor means not being poor, or (2) by providing income support to those not in waged labor such that not being employed does not mean being poor.Theoretically, mixed-income housing does intervene in the operation of labor markets, in two different ways: (1) it potentially increases the social capital of poor people, and (2) it improves the soft skills that employers always say they want from their employees via middle-class role models (of course, employers often say that to mask their discriminatory hiring practices; see Moss and Tilly, 2003). These interventions are modest in the operations of labor markets, however-focusing solely on fairly limited supply-side components of labor markets2-and their effects are more apparent than real. …