«Moving to opportunity voluMe 14, nuMber 2 • 2012 U.S. Department of Housing and Urban Development | Office of Policy Development and Research ...»
General Equilibrium Effects Because of the possibility of general equilibrium effects, Sanbonmatsu et al. (2011) were reluctant to claim that the results would apply to a national policy of offering vouchers to all families with children living in public housing projects in census tracts with poverty rates exceeding 40 percent.
Some simple statistics suggest that these effects are likely to be minimal.
First, the results would surely have been very similar in the metropolitan areas involved had the experiment been expanded to all public housing in the metropolitan area rather than limited to public housing in the central cities of these areas. According to the U.S. Department of Housing and Urban Development’s (HUD’s) Picture of Subsidized Households (HUD, 2008), the central-city PHAs in these areas accounted for 88 percent of all public housing in their metropolitan areas.
They surely accounted for an even higher fraction of all public housing units in census tracts with poverty rates exceeding 40 percent. Therefore, the MTO results already reflect any general equilibrium effects of the policy initiative to a considerable extent.
Second, the reforms affected a part of the market that was too small to have any significant general equilibrium effects. According to HUD (2008), about 1.07 million households are living in public housing units and 37 percent are female headed with children. Data on MTO families indicate that 97.8 percent of the demonstration’s families with children are female headed. Under the assumption that this rate holds true for all families in public housing, about 37.8 percent of all families in public housing have children. Newman and Schnare (1997) estimated that about 36.5 percent of households in public housing live in census tracts with poverty rates in excess of 40 percent.
These numbers suggest that about 155,000 families with children live in public housing projects in these high-poverty neighborhoods. The MTO results suggest that 15 percent of these families would accept a traditional voucher to leave their public housing unit. Therefore, if the policy reform were limited to these families, about 23,000 of the 155,000 eligible households would use the offered vouchers.
This 23,000 figure, however, is an upper limit on the increase in net total demand for housing units in the private market. Under the MTO reform, households moved from the private market to public housing to replace the families who used the MTO vouchers. Thus, the net increase in total demand for units in the private market was about zero, albeit positive in the middle of the quality spectrum and negative for lower quality units. Voucher recipients tend to occupy rental units of about average quality, whereas families who enter public housing typically come from the worst units in the private market. The first alternative reform would have added as many units to the supply in the private market as to the demand (albeit not at each quality level) because the units MTO voucher recipients vacated would have been rented at market rents. They would have become a part of the unsubsidized supply. Only for the second alternative reform would the increase in net demand in the private market have been about 23,000 households. Under this reform, about 23,000 inhabitable public housing units would have been demolished.
This number of voucher recipients would easily be absorbed into the rental housing market with minimal effects on market rents or vacancy rates in any segment of the housing market. According to the American Community Survey, there were more than 44 million rental units and about 3.6 million vacant units available for rent in the U.S. in 2010. More than 220,000 of the vacant units had monthly contract rents in each $50 interval between $400 and $800, more than 350,000 units had asking rents between $800 and $899, more than 250,000 units had asking rents between $900 and $999, and more than 400,000 units had asking rents between $1,000 and $1,250 a month.
The evidence from the Experimental Housing Allowance Program leads to a similar conclusion. Its entitlement Housing Assistance Supply Experiment, which offered tenant-based housing assistance to the 15 to 20 percent of households with the lowest incomes in two metropolitan areas, had minimal effects on the market rents of units of any type (Lowry, 1983).
Conclusion MTO provides evidence on the effects of two important reforms of low-income housing policy on a subset of the people affected by those reforms. At this stage, it is not feasible to estimate the effects of these reforms on many others, namely, the additional people who received housing assistance on account of MTO. This article argues that the MTO evidence is applicable to two similar reforms that, unlike MTO, would have no effect on the number of families who receive housing assistance but would instead affect the nature of the housing assistance offered and the taxpayer cost of providing it. Supplementing the MTO results with estimates of the taxpayer cost of the reforms would provide a reasonably comprehensive analysis of their effects. This article suggests how to estimate the taxpayer costs of the alternative reforms. Doing so would significantly increase the value of MTO research for housing policy development. Existing evidence gives good reason to expect that one of the reforms would achieve the benefits to public housing tenants who accept the offered vouchers at a lower taxpayer cost than the current system.
Acknowledgments The author thanks Jens Ludwig for his helpful comments.
Author Edgar O. Olsen is a professor in the Department of Economics and the Frank Batten School of Leadership and Public Policy at the University of Virginia.
References Dye, Richard F., and Daniel P. McMillen. 2007. “Teardowns and Land Values in the Chicago Metropolitan Area,” Journal of Urban Economics 61 (1): 45–63.
Holin, Mary Joel, et al. 2003. Interim Assessment of the HOPE VI Program: Cross-Site Report.
Bethesda, MD: Abt Associates Inc.
Leger, Mireille L., and Stephen D. Kennedy. 1990. Final Comprehensive Report of the Freestanding Housing Voucher Demonstration, Vols. 1 and 2. Cambridge, MA: Abt Associates Inc.
Lowry, Ira S., ed. 1983. Experimenting With Housing Allowances: The Final Report of the Housing Assistance Supply Experiment. Cambridge, MA: Oelgeschlager, Gunn & Hain.
Mayo, Stephen K., et al. 1980. Housing Allowances and Other Rental Assistance Programs— A Comparison Based on the Housing Allowance Demand Experiment, Part 2: Costs and Efficiency.
Cambridge, MA: Abt Associates Inc.
Newman, Sandra J., and Ann B. Schnare. 1997. “‘… And a Suitable Living Environment’: The Failure of Housing Programs To Deliver on Neighborhood Quality,” Housing Policy Debate 8: 703–741.
Olsen, Edgar O. 2009. The Cost-Effectiveness of Alternative Methods of Delivering Housing Subsidies. Unpublished manuscript. Charlottesville, VA: University of Virginia. http://www.virginia.edu/ economics/Workshops/papers/olsen/CESurvey2009.pdf (accessed June 2, 20120.
ORC Macro. 2001. Quality Control for Rental Assistance Subsidies Determination. Washington, DC:
U.S. Department of Housing and Urban Development, Office of Policy Development and Research.
Popkin, Susan J., et al. 2004. A Decade of HOPE VI: Research Findings and Policy Challenges.
Washington, DC: Urban Institute.
Sanbonmatsu, Lisa, Jens Ludwig, Larry F. Katz, Lisa A. Gennetian, Greg J. Duncan, Ronald C. Kessler, Emma Adam, Thomas W. McDade, and Stacy Tessler Lindau. 2011. Moving to Opportunity for Fair Housing Demonstration Program: Final Impacts Evaluation. Report prepared by the National Bureau of Economic Research for the U.S. Department of Housing and Urban Development, Office of Policy Development and Research. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research.
Turner, Margery Austin, et al. 2007. Estimating the Public Costs and Benefits of HOPE VI Investments:
Methodological Report. Washington, DC: Urban Institute.
U.S. Department of Housing and Urban Development (HUD). 2008. A Picture of Subsidized Households—2008. Available at http://www.huduser.org/portal/picture2008/index.html.
———. 1995. HUD Reinvention: From Blueprint to Action. Washington, DC: Government Printing Office.
U.S. General Accounting Office (GAO). 2001. Federal Housing Programs: What They Cost and What They Provide. GAO–01–901R. Washington, DC: U.S. General Accounting Office.
Wallace, James E., et al. 1981. Participation and Benefits in the Urban Section 8 Program: New Construction and Existing Housing, Vols. 1 and 2. Cambridge, MA: Abt Associates Inc.
Weber, Rachel, et al. 2006. “Tearing the City Down: Understanding Demolition Activity in Gentrifying Neighborhoods,” Journal of Urban Affairs 28 (1): 19–41.
Weinberg, Daniel H. 1982. “Housing Benefits From the Section 8 Program,” Evaluation Review 6:
Cityscape 205 206 Moving to Opportunity Moving Neighborhoods
Versus Reforming Schools:
A Canadian’s Perspective Philip Oreopoulos University of Toronto The Moving to Opportunity (MTO) for Fair Housing demonstration provided a definitive opportunity to consider the often-wondered question of what would result from helping relocate lowincome families away from some of the worst living conditions in the United States. I commend all those involved for the study’s extremely clever design, well-executed operation, and thorough data analysis. I hope the work helps further inspire government and academic collaboration to create demonstration projects to answer equally important policy questions.
The public housing projects in which MTO families initially resided were often characterized by poorly maintained facilities, high levels of crime, extreme ethnic and racial segregation, and isolation from other neighborhoods. Ethnographic studies documenting life in these projects offer persuasive examples of how these conditions adversely affected every aspect of tenants’ day-to-day existence (for example, Venkatesh, 2000). Given this setting, it is perhaps not surprising that, 5 years after being offered housing vouchers and assistance to move, participants felt substantially safer at night and more satisfied with their environment. As the final impacts evaluation (Sanbonmatsu et al., 2011) suggests, these improvements in living conditions may very well relate to the observed subsequent improvements in self-reported mental health (feeling calm and less depressed).
The MTO experiment, however, appears to have had no effect, or possibly negative effects, on children’s educational outcomes. This article focuses on these provoking results. Sanbonmatsu et al. (2011) showed that MTO had no detectable long-term effects on math or reading achievement test scores, whether for youth aged 10 to 12, who spent little of their lives in housing projects, or for the combined youth sample aged 13 to 20. High school graduation and postsecondary attendance rates for older youth were actually lower for the experimental group than for the control group, and the difference was statistically significant in some cases at the 5- and 10-percent levels.
Student attitudes and expectations about school were about the same, although male youth in the experimental group were more likely to have been suspended or expelled.
These findings are consistent with those from Jacob (2004) and some of my earlier research (Oreopoulos, 2003). Jacob (2004) examined families who were offered housing vouchers that enabled them to move from buildings in Chicago housing projects set for demolition because of irreparable conditions. Average census-tract poverty rates for families who received vouchers fell significantly Cityscape 207 Cityscape: A Journal of Policy Development and Research • Volume 14, Number 2 • 2012 U.S. Department of Housing and Urban Development • Office of Policy Development and Research Oreopoulos after 5 years, but children’s math test scores, attendance, retention, and dropout rates were no different compared with those of families from the same projects whose units were not scheduled for demolition. In a previous paper, I used administrative data to track children who grew up in Toronto public housing projects—some in high-density areas, others in smaller buildings in more residential and middle-income areas. Applicants had virtually no control over which project was offered to them at the time they were at the top of the waiting list, so project assignment was effectively random. Although living conditions and exposure to crime varied substantially, I found no differences in eventual earnings, unemployment likelihood, and welfare receipt between residents of the largest and smallest projects. I concluded that, although social interactions may affect social and economic well-being, neighbor interactions may not be important enough to significantly influence behavior, at least on average. Individuals have greater choice over whom to interact with in a neighborhood setting than in a classroom or college dormitory setting. Perhaps for this reason, evidence of group effects is more convincing using within-school variation of classmates or roommates than across-school or across-neighborhood variation (for example, Carrell, Fullerton, and West, 2009; Friesen and Krauth, 2011; Hoxby, 2000; Lavy and Schlosser, 2011).
At the time I completed my study, I wondered whether the final impacts evaluation would make my study obsolete, given its more compelling experimental design and its more interesting sample population of individuals from some of the most impoverished neighborhoods in the United States. Arguably, if MTO showed no detectable effects from moving these individuals away from extremely poor housing conditions, we would not expect to detect effects from anywhere else. On the other hand, if MTO showed significant effects, sample and neighborhood differences would plausibly reconcile those findings with my results. Although large public housing projects in Toronto are unattractive, they do not exhibit nearly the same degree of crime and racial segregation that occur in high-poverty neighborhoods in the United States (for example, Oreopoulos, 2008).