«Moving to opportunity voluMe 14, nuMber 2 • 2012 U.S. Department of Housing and Urban Development | Office of Policy Development and Research ...»
Economic Effects In contrast to finding effects on health, we generally do not detect any beneficial effects on employment and earnings, household poverty, or reliance on government assistance programs. Exhibit 5 shows that slightly more than one-half of adults in the control group were employed at the time they were interviewed. Compared with controls, adults in the experimental group were not significantly more likely to be employed. Adults in the Section 8 group were 7.7 percentage points less likely than adults in the control group to report that they were working (p.05), but we interpret this result with caution because adults in the Section 8 group were interviewed slightly later in calendar time than adults in the control group, during a period of rising unemployment.
Differences across groups may in part reflect different economic conditions at the time we surveyed the adults rather than effects on labor-market outcomes. (We think our mental and physical health measures are less susceptible to changes over short periods of time and that timing differences are less of a problem for those outcomes. One reason we think the Section 8 versus the control group difference in self-reported economic outcomes might be the result of the slight differences in when adults were surveyed comes from our analysis of administrative UI system data. The UI data enable us to compare employment for the Section 8 and control groups at the exact same point in time.
When examining employment in 2007 for all three groups, we find no statistically significant differences across them.
We also do not detect differences in household income or receipt of food stamps and TANF, except for a slight increase in the amount of food stamps received by the experimental group. The average household income of adults in the control group was $12,289 (in 2009 dollars),14 and the averages for adults in the experimental and Section 8 groups were similar. The proportion of households at or below the poverty threshold was roughly equal across the three randomized groups, with 59 percent of adults in the control group living in poor households. Of adults in the control group, 47 percent reported currently receiving food stamps and 16 percent reported receiving TANF benefits.
The corresponding rates for adults in the experimental and Section 8 groups were not significantly different. Using administrative records, we were also able to estimate the amount of food-stamp and We adjusted the responses to 2009 dollars using the Consumer Price Index for All Urban Consumers provided by the U.S. Bureau of Labor Statistics.
126 Moving to Opportunity The Long-Term Effects of Moving to Opportunity on Adult Health and Economic Self-Sufficiency
TANF benefits received during the 2-year period from July 2007 through June 2009. We observed no significant difference across the three randomized groups in the amount of TANF benefits received but did observe a higher level of food-stamp receipt (by $310 over 2 years, or an average of $13 per month) for adults in the experimental group compared with adults in the control group.
Discussion As Ludwig (2012) discussed, MTO generated large differences in neighborhood disadvantage and other conditions between the two treatment groups and the control group in the period immediately following random assignment. Over time, these differences narrowed, in large part because of improvements in the neighborhood conditions of the control group, but even with this narrowing the differences are sizable between groups over the entire 10- to 15-year period.
Our 10- to 15-year followup evaluation shows that MTO-induced changes in neighborhood conditions were associated with beneficial impacts on a number of key mental and physical health outcomes but with few, if any, effects on different economic self-sufficiency measures. Specifically, we found that adults in the experimental group had lower levels of psychological distress than did adults in the control group and that adults in the Section 8 group had lower levels of lifetime depression.
The results also suggest, however, an increase in substance dependence for the experimental group relative to the control group. On physical health, we detect beneficial effects on diabetes, severe obesity, and health limitations, but we do not detect effects on self-rated health, asthma, or hypertension. The voucher offer reduced the prevalence of diabetes by 5.2 percentage points for the experimental group.
Effects on some key outcomes are of a clinically relevant magnitude. For example, MTO generated very large reductions in diabetes that might be comparable to those found in studies or programs designed to prevent diabetes. In comparing MTO to medical interventions, keep in mind that most clinical trials in medicine usually enroll study samples that are more socioeconomically advantaged than the low-income families who enrolled in MTO, often enroll individuals at high risk for a particular health problem, and may measure the outcome differently. In addition, reproductive-age women are disproportionately underrepresented in these studies. With those qualifications in mind, MTO’s effects on diabetes appear to be similar in magnitude to those found in the Diabetes Prevention Program (DPP) randomized trial of lifestyle interventions designed to prevent diabetes that took place in clinical centers across the United States (Knowler et al., 2009). In DPP, a case manager met with participants for 16 initial sessions (and monthly afterwards) to help them modify their diet and exercise habits, with the goal of reducing body weight by 7 percent and engaging in
2.5 hours of moderate physical activity per week. Over a 10-year period, the lifestyle intervention reduced the incidence of new cases of diabetes by about 34 percent relative to the incidence in the placebo group, an effect of similar magnitude to the experimental group treatment.15 MTO diabetes Under the assumption that about 5 percent of MTO adults may have had diabetes at the start of the program, we estimate that the incidence of new cases among adults in the control group was about 1.22 per 100 person years (where 1.22 = [0.204 final prevalence – 0.05 assumed baseline prevalence] x [1/12.67 years] x 100 years) and was about 0.81 per 100 person years (where 0.81 = [0.204 final control prevalence – 0.052 effect – 0.05 assumed baseline prevalence] x [1/12.67 years] x 100 years) for the MTO treatment group. Thus, we estimate about a 34-percent reduction in the incidence of new cases (0.34 = [0.81 – 1.22]/1.22).
128 Moving to Opportunity The Long-Term Effects of Moving to Opportunity on Adult Health and Economic Self-Sufficiency effects are also noteworthy because of the costs associated with the disease; Trogdon and Hylands (2008) estimate that the annual medical expenditures of people with diabetes are 239 percent greater than those of people without diabetes (after adjusting for demographic differences).
The results we report here from our long-term (10- to 15-year) survey of MTO adults for the final impacts evaluation suggest that the lack of MTO effect on economic outcomes found in the followup (4- to 7-year) survey for the interim impacts evaluation was not simply because of the disruptive effects of moving itself or of the fact that families may not have been in their new neighborhoods long enough to fully adapt and take advantage of any new opportunities in those areas. Given the previous nonexperimental research literature suggesting that neighborhood environments affect labor market outcomes, what might explain why we do not observe beneficial effects on adult selfsufficiency? That MTO may have had only modest effects on the mechanisms that affect employment and earnings is one explanation. For example, the areas to which families moved through MTO may not have offered more plentiful job opportunities. At 4 to 7 years after baseline, the interim impacts evaluation showed little effect of moves on local job availability as measured by employment growth by residential ZIP Code (Kling, Liebman, and Katz, 2007). In addition, although MTO moves appear to have increased the likelihood that adults in the experimental group had a college-educated friend, qualitative work with MTO families suggests that new neighbors may not have known about the types of job openings that MTO adults were seeking (Turney et al., 2006).
Our findings suggest that housing mobility programs similar to MTO are unlikely, by themselves, to be panaceas for the labor-market difficulties of disadvantaged families living in public housing projects in inner-city neighborhoods. Investing in high-quality training and employment services may be a more promising way to improve labor-market outcomes for very disadvantaged adult samples of the sort that enrolled in MTO. For example, the Jobs-Plus demonstration produced sustained (7-year) earnings gains for adult public housing residents without disabilities through employment and training services, changes in rent rules to increase work incentives, and neighbor-to-neighbor outreach centering on work (Riccio, 2010). Several training programs that prepare underskilled workers for skilled positions in a specific industry and connect them to employers with job openings have also produced substantial earnings gains for disadvantaged adults in large U.S. cities, and these gains applied to women and African-American adults in the study (Maguire et al., 2010).
For very disadvantaged adults like those who participated in MTO, policies that directly increase skills, help with the acquisition of work experience, assist with job searches, and provide workplace supports may be necessary to improve economic self-sufficiency. The MTO findings suggest that housing mobility programs could help improve the physical health and mental well-being of disadvantaged adults. Our hope is that future work helps illuminate the mechanisms through which community environments influence health outcomes to help guide the design of community-level interventions that can improve health without having to rely on relocating families to new areas.
Cityscape 129 Sanbonmatsu, Marvakov, Potter, Yang, Adam, Congdon, Duncan, Gennetian, Katz, Kling, Kessler, Lindau, Ludwig, and McDade Authors Lisa Sanbonmatsu is a senior researcher at the National Bureau of Economic Research.
Jordan Marvakov is an economist at Eastern Research Group, Inc.
Nicholas A. Potter is a research analyst at the National Bureau of Economic Research.
Fanghua Yang is a research analyst at the National Bureau of Economic Research.
Emma Adam is a professor of human development and social policy and a faculty fellow at the Institute for Policy Research at Northwestern University.
William J. Congdon is a research director at The Brookings Institution.
Greg J. Duncan is a distinguished professor in the School of Education at the University of California, Irvine.
Lisa A. Gennetian is senior research director of economic studies at The Brookings Institution.
Lawrence F. Katz is the Elisabeth Allison Professor of Economics at Harvard University and a research associate at the National Bureau of Economic Research.
Jeffrey R. Kling is associate director for economic analysis at the Congressional Budget Office and a faculty research fellow at the National Bureau of Economic Research.
Ronald C. Kessler is the McNeil Family Professor of Health Care Policy at Harvard Medical School.
Stacy Tessler Lindau is associate professor of obstetrics/gynecology and medicine-geriatrics at the University of Chicago.
Jens Ludwig is the McCormick Foundation Professor of Social Service Administration, Law, and Public Policy at the University of Chicago and a research associate at the National Bureau of Economic Research.
Thomas W. McDade is professor of anthropology, faculty fellow at the Institute for Policy Research and Director of Cells to Society (C2S): Center on Social Disparities and Health at Northwestern University.
References Allard, Scott W. 2004. “Access to Social Services: The Changing Urban Geography of Poverty and Service Provision.” Survey Series. Washington, DC: The Brookings Institution.
American Diabetes Association. 2010. “Standards of Medical Care in Diabetes—2010,” Diabetes Care 33 (1): S11–S61.
Angrist, Joshua D., Guido W. Imbens, and Donald B. Rubin. 1996. “Identification of Causal Effects Using Instrumental Variables,” Journal of the American Statistical Association 91 (434): 444–455.
Anil, Bulent, David L. Sjoquist, and Sally Wallace. 2010. “The Effect of a Program-Based Housing Move on Employment: HOPE VI in Atlanta,” Southern Economic Journal 77 (1): 138–160.
Bagley, Christopher, Solomon Jacobson, and Clare Palmer. 1973. “Social Structure and the Ecological Distribution of Mental Illness, Suicide, and Delinquency,” Psychological Medicine 3 (2): 177–187.
130 Moving to Opportunity The Long-Term Effects of Moving to Opportunity on Adult Health and Economic Self-Sufficiency
Bayer, Patrick, Stephen L. Ross, and Giorgio Topa. 2008. “Place of Work and Place of Residence:
Informal Hiring Networks and Labor Market Outcomes,” Journal of Political Economy 116 (6):
Bennett, Gary G., Lorna H. McNeill, Kathleen Y. Wolin, Dustin T. Duncan, Elaine Puleo, and Karen M. Emmons. 2007. “Safe To Walk? Neighborhood Safety and Physical Activity Among Public Housing Residents,” PLoS Medicine 4 (10): 1599–1606.
Berkman, Lisa F., and Thomas Glass. 2000. “Social Integration, Social Networks, Social Support,
and Health.” In Social Epidemiology, edited by Lisa F. Berkman and Ichiro Kawachi. New York:
Oxford University Press: 137–173.
Bloom, Howard S. 1984. “Accounting for No-Shows in Experimental Evaluation Designs,” Evaluation Review 8 (2): 225–246.
Bloomberg, Gordon R., and Edith Chen. 2005. “The Relationship of Psychologic Stress With Childhood Asthma,” Immunology and Allergy Clinics of North America 25 (1): 83–105.
Bryant-Stephens, Tyra. 2009. “Asthma Disparities in Urban Environments,” The Journal of Allergy and Clinical Immunology 123 (6): 1199–1206.
Chobanian, Aram V., George L. Bakris, Henry R. Black, William C. Cushman, Lee A. Green, Joseph L. Izzo, Daniel W. Jones, et al. 2003. “Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure,” Hypertension 42 (6): 1206–1252.