The Johns Hopkins Gazette: April 22, 2002
April 22, 2002
VOL. 31, NO. 31


Public Housing Can Be Positive for Children

Teens, preteens fare better later than poor kids whose housing wasn't subsidized

By Lavinia Edmunds
Institute for Policy Studies

Johns Hopkins Gazette Online Edition

Public housing projects are commonly regarded as breeding grounds for drugs and crime, not as positive places in which to grow up. But a new study by researchers at the Johns Hopkins Institute for Policy Studies shows that poor children who spent at least some of their preteen or teenage years in public housing were better off than if they had not lived there.

The study by Sandra Newman and Joseph Harkness, published in a recent issue of the Journal of Policy Analysis and Management, compared public housing residents with eligible nonresidents, using 25 years' worth of records. To gain a clearer idea of the impact of public housing, the researchers took into account factors such as family income and education levels of parents. Among the key findings:

Every year that someone lived in public housing between the ages of 10 and 17 increased by 7 percentage points the chances of that person working between the ages of 25 and 27.

People who lived in public housing later spent less time on welfare. For every year of public housing residence, welfare use was reduced by nearly three-quarters of a year between the ages of 20 and 27.

Every year of residence in public housing increased annual earnings by $1,860 for those between the ages of 25 and 27.

The study showed that children who grew up in public housing still faced obstacles of poverty and had worse outcomes on most measures than children of similar income levels who did not live in public housing, but when public housing was isolated from other factors, those outcomes were shown to be due to more disadvantaged family backgrounds, not to public housing itself.

According to the authors, public housing may have improved the life chances of children because it offered improved living conditions and less transience. Also, they said, housing subsidies may have freed up income that families could spend on items to benefit children's development. In addition, public housing could have relieved financial pressure on parents, reducing their levels of stress and depression.

Newman and Harkness' work, The Long-Term Effect of Public Housing on Self-Sufficiency, is the first study of long-term effects of public housing on children. Most research in housing has addressed "bricks and mortar" issues, such as the number of dwelling units meeting housing codes. But as the goals of assisted housing have shifted from shelter to self-sufficiency, debate has increasingly focused on the need for improving outcomes, such as labor force participation.

Housing policy has changed, too, as the government has attempted to reduce the concentration of the very poor in assisted housing developments. Assisted housing has begun to serve more mixed-income residents, making it similar in composition to that of assisted housing during the 1970s, the period examined in the study.

To measure the long-term effects, Harkness and Newman used the Panel Study of Income Dynamics, a database at the University of Michigan that is an ongoing survey of U.S. households since 1968. The PSID follows members of its sample over time. One of the challenges of research on subsidized housing lies in the development of an accurate sample. Studies in the past have relied on self-reports by individuals in sample surveys, but these are subject to the inaccuracies of recall and confusion over the definition of "public housing."

In order to identify public housing residents accurately, Newman matched the PSID data with addresses of assisted (or subsidized) housing units obtained from the U.S. Department of Housing and Urban Development, state housing agencies and other sources throughout the United States. This method makes the sample a unique and valuable resource for measuring the impact of public housing.

The study sample includes both those who lived in public housing at some point between the ages of 10 and 16 and those who were eligible but were unassisted between the ages of 10 and 16 during the period from 1968 to 1982. Adult outcomes from ages 20 to 27 cover the period from 1978 through 1993.

The paper is available online at