Public housing in the United States has never sheltered a significant proportion of Americans, perhaps 3 percent at most, unlike in many western European countries where 10 to 40 percent of households, at various income levels, live in state-constructed buildings. But public housing has been a significant part of the debate over American government safety net programs, a significant factor in the history of large American cities over the last fifty years, and cruel disillusionment for social reformers (and many sociologists).

American public housing projects started in the New Deal, accelerated after the war, and then largely stopped in the 1970s, when they were widely described as abject failures. This verdict was hammered home by the well-publicized demolition in 1972 of the Pruitt–Igoe project in St. Louis. Federal support for housing since, skimpy as it is, has largely been in the form of “Section 8” vouchers and dispersed, low-density, mixed housing. The actual number of public housing units has shrunk in recent decades.

A new study in the Journal of Economic History, by Katharine L. Shester, fleshes out our understanding of what went wrong in this great social experiment. In some ways, large-scale public housing was doomed from the start; in other ways, perhaps different critical decisions could have made it work.

Building Up, Coming Down

The U.S. began building major projects to house needy families (which should be distinguished from projects to house the elderly) in the 1930s, but the program really took off in the 1950s, creating about 1 million units by 1973. It was a response to the post-war housing shortage and to many social scientists’ view at the time that poor housing itself – crowded, dilapidated quarters – contributed to social dysfunction.

But there were problems right from the start, including the projects’ very locations. The authorities put public housing projects, especially those in the large cities, disproportionately in poor black neighborhoods. As earlier studies showed [1], this was only in part because those neighborhoods had the worst housing and neediest people. In good measure, the politics of class and race decided location; those with the clout to resist low-income neighbors made sure that the construction took place where people who could not resist lived. Another problem at the start was government insistence on holding down construction costs, which would eventually produce problems with upkeep.

Placing large public housing structures in a neighborhood eventually contributed to further decline in the neighborhood. As shown in studies of Chicago [1] and Columbus, Ohio, in the 1960s and ‘70s, the presence of public housing accentuated local poverty, in part at least because better-off residents moved away. Shester’s study provides a new look at that process.

A Longer Look

Shester, in the new JEH article, examines the entire public housing experiment by looking at the whole country from 1933 to 1973. She shows that by 1970, even taking into account local conditions prior to their construction, public housing projects depressed counties’ social and economic levels. Critically, however, that was not true before 1970. Data from 1950 and 1960 suggest that public housing seemed to have positive local effects. Something changed in the 1960s. The source of change was not, in Shester’s analysis, the new projects built in the 1960s, but seems to have been some cumulative aspect of public housing generally.

Decisions and developments from 1950 to 1970, Shester argues, accelerated the physical deterioration of public housing and increased the concentration of troubled families living there. Limits on government maintenance funds, like the limits on the original construction costs, hampered the housing managers. And because of imposed rent ceilings, local housing agencies could not get the funds sufficient to keep up repairs by charging tenants. Physical dilapidation followed.

At the same time, tightening the requirements that housing be provided only to the neediest families meant that stable working-class families, once part of the mix, were gone. The renters became increasingly and exclusively the poorest and most troubled families. Their growing concentration in dense (and tense) settings compounded the problems of order. By 1970, public housing projects had gained their nightmarish image. Pruitt–Igoe (and others) came down.

Lessons Learned?

One lesson many policy analysts took away from the public housing experiment was that market solutions for housing shortages are better than government ones. Yet, the current Section 8 voucher system has its own problems. Another lesson is that mixed-income housing is preferable. A third is that low-density structures avoid some of the problems. The jury is still out on these ideas (as far as I know).

Yet another set of lessons are that the disasters that were Pruitt–Igoe, the Robert Taylor Homes, and others like them might perhaps have been avoided by better and, yes, more expensive planning—greater attention to location, more money up-front for construction and maintenance, and more social mixing. We hear an echo of this story in the current fallout from the Obamacare federal website debacle, that because of the political battles, the construction of the system was underfunded at the start. It seems that when the U.S. does public provisioning of basic needs, it does so halfheartedly, perhaps thereby promoting its failure.

[1] E.g., Massey and Kanaiaupuni, “Public Housing and the Concentration of Poverty,” Social Science Quarterly 74 (1993).