May 7, 2015 | Issue Brief on Poverty and Inequality
Two economic research papers published this month show that where children live can have an impact on their prospects for success later in life. Parents already know that—and it is why houses in good neighborhoods often cost three or four times as much as houses in bad neighborhoods.
The new studies have garnered outsized attention, but the results are neither as clear nor as firm as reported. One paper finds both benefits and costs to relocating families with children. The other paper finds that children who moved had different outcomes in different places—but does not prove that their success or failure would be replicated by others who made the same moves.
The first study is straightforward and finds that the experimental Moving to Opportunity housing-voucher program in the 1990s had positive effects for young children and negative effects for teenagers. In their new neighborhoods, the younger kids benefited from growing up with safer streets, better schools, fewer gangs, more neighbors with intact families, more prosperity, harder-working adults, and the rest of the characteristics that define a “good neighborhood.” It confirms what families already know: Sometimes the path to opportunity involves a new location.
However, moving to a better neighborhood hurt teenagers’ future earnings, possibly because the move disrupted social networks. The disruption cost confirms the importance of place: Social networks are valuable even in very poor neighborhoods. The paper indicates that the disruption cost is large—equal to about five years of living in a better neighborhood—although it is statistically imprecise. The disruption cost affects young children as well as teens; the authors did not test whether it affects adults.
Families who participated in the Moving to Opportunity experiment understood the potential costs of moving: About half of those who were offered a subsidy turned down the chance to move. Presumably, the families who opted out were those for which the disruption costs would have been larger and the benefits smaller.
The paper makes a contribution by emphasizing that moving has up-front costs as well as long-term benefits for children. It implicitly cautions against housing policies that move families around and recommends targeting existing relocation opportunities to families with young children.
Moving to Opportunity has been widely studied. Previous studies found no effect on most income, health, and academic outcomes for both adults and children. Although the present study finds some positive income effects for young children, the experiment failed to achieve most of its objectives.
It would be unwise to turn the Moving to Opportunity experiment into a widespread feature of housing policy. An earlier generation of urban social planners also believed that geography caused poverty and moved thousands of poor families from their old tenements to brand-new government housing projects. The projects often became much worse neighborhoods and stand today as a symbol of governmental hubris and failure.
The second study is much harder to understand and economists are still debating what the results mean. Regrettably, the paper is ripe for misinterpretation. The authors call the study “The Impact of Neighborhoods on Intergenerational Mobility,” but the “neighborhoods” in question are counties or entire metropolitan and rural areas. Unlike the Moving to Opportunity study, the second study has almost nothing to say about neighborhoods in the sense of Plymouth-Exchange, Jamaica Plain, or Petworth.
The study primarily uses data gathered from families moving at least 100 miles, not families remaining within the same metro area. When a family moves, the authors compare the outcomes for younger children to the outcomes of older children. They do find that the effects of a move (positive or negative) have more influence on outcomes for siblings who are young at the time of the move than for those who are teenagers. They also find that geography matters more for boys than for girls.
However, unlike the first study, the second lacks experimental rigor. Families who move are dissimilar to families who stay put. The study also tries to remove the effects of parental income, so that a parent who earns $60,000 a year in Santa Clara County, California (median household income: $92,000), is directly compared to a parent who earns $60,000 a year in Colusa County, California (median household income: $52,000). But a parent capable of earning $60,000 a year in Colusa is likely more skilled or more motivated than one earning $60,000 in high-wage Santa Clara. Even more concerning is that the life events (such as job loss, divorce, or graduation) that spark a move to Santa Clara are likely to indicate a different trajectory in family fortunes than the life events that spark a move to rural Colusa.
The authors recognize this shortcoming, admitting that “the availability of jobs is another important factor in a location decision, [so] it is potentially misleading to consider the negative correlation with rent and house prices as an indication that it is cheaper on net to move to a [region] with a higher causal effect.”
The same caution can be extended to most of the paper. One of the primary benefits of living in Santa Clara County is the availability of well-paid work. So when The New York Times reports that Colusa County is “better than about 85 percent of [U.S.] counties”—including Santa Clara—for low-income children’s economic prospects, it is assuming that their parents make the same amount of money living in Colusa as they would in Santa Clara or anywhere else. For most prospective movers, ignoring the job market would be foolhardy.
Technically speaking, the effects that the authors estimate are analogous to the “treatment on the treated.” That is not the same as estimating the marginal treatment effect, which would be of vastly more interest to parents and policymakers.
The paper is so easily misinterpreted because it answers a question that no one is asking while sounding similar to questions that many people do ask: “Where should I move to maximize my children’s opportunities? Should I move at all?” The study does not answer those questions.
Research on neighborhood effects may tempt policymakers to play real estate agent, subsidizing moves from one place to another. A better approach is to use the research as a reminder of the importance of local policies.
Municipalities can improve children’s outcomes by promoting public safety and using competition and parental choice to offer better and more diverse schools. Counties with high wages and low unemployment, like Santa Clara, should permit denser residential construction, allowing more families to afford access to their job markets.
Many, perhaps most, American families will move to a better neighborhood or more prosperous city at some point in their lives. But American families need new opportunities, not just a shuffling around of existing ones. The response to failing schools, shuttered factories, and gang-haunted playgrounds can—and should—be much more innovative than a moving truck.—Salim Furth, PhD, is Research Fellow in Macroeconomics in the Center for Data Analysis, of the Institute for Economic Freedom and Opportunity, at The Heritage Foundation.
 County-level estimates provide a natural link to local news. See, for example, Jerome Stockfisch, “Growing Up Poor in Hillsborough? It’s Going to Cost You,” Tampa Tribune, May 4, 2015, http://tbo.com/news/education/growing-up-poor-in-hillsborough-its-going-to-cost-you-20150504/ (accessed May 5, 2015), and Adam Carlson, “Most Metro Counties Are Actually Bad for Your Child’s Economic Future,” Atlanta Journal-Constitution, May 5, 2015, http://www.ajc.com/news/business/most-metro-counties-are-actually-bad-your-childs-e/nk858/ (accessed May 5, 2015).
 Raj Chetty, Nathaniel Hendren, and Lawrence F. Katz, “The Effects of Exposure to Better Neighborhoods on Children: New Evidence from the Moving to Opportunity Experiment,” Working Paper, May 2015, http://www.equality-of-opportunity.org/images/mto_paper.pdf (accessed May 5, 2015).
 Ibid., p. 27.
 David Muhlhausen, Do Federal Social Programs Work? (Santa Barbara, CA: Praeger, 2013), pp. 190–211.
 Raj Chetty and Nathaniel Hendren, “The Impacts of Neighborhoods on Intergenerational Mobility: Childhood Exposure Effects and County-Level Estimates,” Working Paper, May 2015, http://www.equality-of-opportunity.org/images/nbhds_paper.pdf (accessed May 5, 2015).
 Neighborhoods in which the author has lived. They are in Rochester, NY, Boston, and Washington, DC, respectively.
 U.S. Census Bureau, “State and County QuickFacts: Santa Clara County, California,” data for 2009–2013, http://quickfacts.census.gov/qfd/states/06/06085.html (accessed May 5, 2015).
 U.S. Census Bureau, “State and County QuickFacts: Colusa County, California,” data for 2009–2013, http://quickfacts.census.gov/qfd/states/06/06011.html (accessed May 5, 2015).
 Chetty and Hendren, “The Impacts of Neighborhoods on Intergenerational Mobility,” p. 77.
 Gregor Aisch et al., “The Best and Worst Places to Grow Up: How Your Area Compares,” The New York Times interactive feature, http://www.nytimes.com/interactive/2015/05/03/upshot/the-best-and-worst-places-to-grow-up-how-your-area-compares.html (accessed May 5, 2015).
 Treatment effects are an area of study in economics. See, for example, James Heckman and Edward Vytlacil, “Chapter 71: Econometric Evaluation of Social Programs, Part II: Using the Marginal Treatment Effect to Organize Alternative Econometric Estimators to Evaluate Social Programs, and to Forecast their Effects in New Environments,” in Handbook of Econometrics, James Heckman and Edward Leamer, eds., Vol. 6, Part B, pp. 4875–5143, 2007, http://www.sciencedirect.com/science/article/pii/S1573441207060710 (accessed May 6, 2015).
 The paper answers the question: “Among parents who moved and made a certain level of income, which locations were associated with the best outcomes for their children?”