Briefs on Affordable Housing Policy

This set of research briefs and related abstracts highlights the findings and policy implications of research on the topic of Affordable Housing Policy supported by the John D. and Catherine T. MacArthur Foundation, together with other related research.

Additional briefs and abstracts will be added to this page as they become available; please check back in the coming months for updates!

1. “Don’t Put it Here!” Do Subsidized Housing Developments Cause Nearby Property Values to Decline?

Related abstracts of MacArthur-funded reports referenced in this brief:

The Well-Being of Low- Income Children: Does Affordable Housing Matter?

Related abstracts of MacArthur-funded reports referenced in this brief:


Does Federally Subsidized Rental Housing Depress Neighborhood Property Values? (2005)

Authors: Ingrid Gould Ellen, Amy Ellen Schwartz, Ioan Voicu, and Michael H. Schill

Research Question: How does subsidized rental housing affect nearby property values?

Results: In this study, the researchers compared sales prices of New York City homes located near federally subsidized rental housing to sales prices of similar homes located in the same neighborhood, but farther from subsidized housing. Using New York City property sales and property tax assessment data, they found that on average, subsidized housing is associated with a small increase in neighboring property values. Benefits are larger in more distressed neighborhoods. In the case of city-assisted housing, the resulting increase in property tax revenue exceeded the cost of the subsidies provided by the city.

Broken down by program, LIHTC and Section 202 projects both affect nearby property values positively, with LIHTC developments conferring the larger benefit. Section 8 and public housing projects both affect nearby property values negatively, with Section 8 developments the more detrimental, but these effects diminish over time. As all types of subsidized housing developments increase in size, their positive effects decrease and their negative effects increase.

Data: The researchers used data on New York City property sales from 1974-2002; NYC Deparment of Finance property tax assessment data (RPAD file); NYC Department of Housing Preservation and Development administrative data on city-assisted housing from 1987-2000; and HUD and NYC Housing Authority data on housing units in the city from 1977 to 2000 that were subsidized through the federal LIHTC, Section 8, Section 202, and public housing programs.

Analysis: The study uses a multivariate analysis called “hedonic regression” that explains a home’s sales price as a function of its structural characteristics, such as building age and square footage, and its neighborhood surroundings. Sales prices of properties within 2,000 feet of subsidized housing are compared to sales prices of structurally similar properties located in the same neighborhood beyond the 2,000- foot radius.

Caveats: New York City’s housing market may be atypical because of its diversity, density, and large share of housing that is subsidized. The city purposely targeted subsidies to revitalize “distressed” properties that were negatively impacting surrounding neighborhoods, so some portion of the increase in nearby property values may be a function of what was removed rather than what was built. Results cannot necessarily be generalized to non-urban environments.

Find this study
: ffr.pdf

Back to top


Nonprofit Housing and Neighborhood Spillovers (2006)

Authors: Ingrid Gould Ellen & Ioan Voicu

Research Questions
: How do city-supported rental housing rehabilitation projects undertaken by nonprofit and for-profit developers affect nearby property values? How do effects differ between the two sectors?

: The researchers examined home price appreciation trends near city-supported rental housing developments in New York City that were rehabilitated by either for-profit or nonprofit developers. They found that rehabilitation of rental housing by both for-profit and nonprofit organizations raises surrounding property values. The “spillover” benefits of nonprofit housing last longer than those of for-profit housing. The magnitude of the benefits does not differ between sectors for large projects, but for-profit developers generate larger benefits than nonprofit developers when projects are small.

Data: The study uses New York City Department of Housing Preservation and Development (HDC) data on city-assisted rehabilitations of subsidized rental housing from 1976-2000; home sales price data from the NYC Department of Finance from 1980-1999; and NYC Department of Finance data on building characteristics from an administrative dataset used to assess property taxes (RPAD).

Analysis: The study uses a multivariate analysis called “hedonic regression” that explains a home’s sales price as a function of its structural characteristics, such as building age and square footage, and its neighborhood surroundings. Sales prices of properties within 1,000 feet of city-supported rental housing rehabilitation projects are compared to sales prices of structurally similar properties located in the same neighborhood beyond the 1,000-foot radius, both before and after completion of the rehabs.

: Because New York City represents a unique housing market, these findings may not apply in other locations.

Find this study: _090105_000.pdf

Back to top


Does Housing Matter for Poor Families? A Critical Summary of Research and Issues Still to be Resolved (2008)

Author: Sandra Newman

Research Questions: Does housing policy research from the last twenty-five years support the argument that housing provides benefits in addition to its economic value? How does housing impact the lives of residents, particularly children?

Results: Despite advances in the area of housing policy research over the last quarter century, the author calls for more research and rigor in this field of study. She cites several methodological issues in existing research on this topic –conceptualization, measurement, data, and analysis – that need to be resolved in order to establish a definitive answer about the value of housing and the impact that housing has on its occupants.

The literature review is divided into the following five categories:


Data: This literature review is primarily based on quantitative research studies from the last twenty-five years that contain data collected through systematic sample surveys but includes insights from a number of qualitative studies.

Caveats: The author identifies a number of shortcomings in existing housing policy research. First, the relationship between housing and outcomes for individuals and families is still a relatively new area of research and there is not yet a systematic approach to collect and measure data. Second, most of this research focuses on correlation, not causation. Therefore, we do not know whether housing is the key factor that causes positive outcomes and benefits beyond housing, or if unmeasured factors contribute to positive outcomes as well. The author recommends that a set of key housing measures be established, tested and widely disseminated to gather data for future housing policy research.

Back to top


Geographic Differences in Housing Prices and the Well-Being of Children and Parents (2008)

Authors: Sandra Newman, Joseph Harkness, and Scott Holupka

Research Questions: Does the well-being of parents and children differ based on whether they live in a high priced (“unaffordable”) or low priced (“affordable”) housing market? Is the debate about geographic adjustments in the poverty line informed by evidence about children’s outcomes in high-priced versus low-priced housing markets?
Results: The authors find little evidence of associations between higher housing prices and negative cognitive, health, and behavioral outcomes among low-income children. Parents and children in high-priced housing markets appear to fare no worse than those in lower-priced markets. In fact, some children living in areas with higher Fair Market Rents (FMRs) experience slightly more favorable outcomes compared to those living in areas with lower FMRs.

There are the two main hypotheses that explain the impact of housing prices and children’s well-being. The first hypothesis suggests that high-priced markets are “bad” because they may jeopardize children’s health and well-being. The second hypothesis suggests that high-priced markets are “good” because they give children access to better schools, lower crime rates, and the like. The results of this study are more consistent with the second hypothesis than the first. However, it is not clear to what extent the results reflect the impact on children’s well-being of unmeasured attributes of parents who spend more of their income on housing or neighborhood characteristics (better schools, health facilities, economic opportunities, lower crime).

Data: The Panel Study of Income Dynamics (PSID)1, together with its 1997 and 2002 Child Development Supplements (CDS)2 is the main data source for this study. The U.S. Department of Housing and Urban Development’s FMRs are used as the housing price measure. It represents the 40th percentile of rents for each housing market in the nation.

Statistical modeling is used to analyze the relationship between housing costs and child and parent well-being. The analysis is conducted on two groups of children; poor and near poor. The “poor” subsample is characterized as children in families earning less than the federal poverty line. The “near poor” subsample is characterized as families earning between one and two times the federal poverty line. Covariates include race, age, tenure, migration history, among others, to assess whether these variables impact the relationship between affordability and children’s well being.

Outcomes for children and parents are measured in 1997 and 2002. Three main analyses are conducted: one using the 1997 outcomes; the second using the 2002 outcomes; and the third looking at change between 1997 and 2002. Children's outcomes include cognitive test results, health assessment, and behavior problems, while parental outcomes include depression, self-esteem, aggravation and economic stress).

: This is the first analysis of the topic using a single database. There is need for “stress testing” using other databases, which Newman and Holupka are now doing.

1. PSID is an ongoing longitudinal study of American households conducted by the University of Michigan Survey Research Center. Learn more at
2. The CDS measures children’s health, cognitive ability, neighborhood conditions, and parental mental health.

Back to top


Something’s Gotta Give: Working Families and the Cost of Housing (2005)

Author: Barbara J. Lipman

Research Question: How are individual families affected by high housing costs? Do families make tradeoffs in spending on other necessities to pay for housing?

Results: This report summarizes findings from two studies that look at how living in affordable housing affects the quality of life of working families. The first study, conducted by the Economic Policy Institute, examined the Department of Labor’s Consumer Expenditure Survey to determine whether expenditures on food, clothing, healthcare and other necessities differed among families in affordable housing versus those in unaffordable housing. That is, to what extend does spending an excessive portion of household income on housing cause material hardship for family members?

The study found that households paying more than half of their income on housing costs commonly spend less on other essential expenses such as food, clothing, and health care. As the size of the family increases, these hardships tend to be exacerbated. Moreover, those households most likely to experience material hardships tend to be renters (as compared to homeowners) and members of traditionally disadvantaged social and economic groups, including low-income, poor, and single-parent families, racial and ethnic minorities, immigrants and parents with low levels of educational attainment.

A companion study by the research group Child Trends looked at the 2002 National Survey of America’s Families to determine the effects of paying more than half of income for housing on the well-being of children and parents. That study found that factors such as parents’ income and education, as well as neighborhood quality, appeared to affect the well-being of children in complex ways. Children in the poorest working families with high housing cost burdens are more likely than those in more affordable housing to report having fair or poor health. But just the opposite is true for children in the highest tier of working families (earning 100 to 120 percent of area median income) – those with high housing cost burdens are less likely to experience these problems, possibly because their income and neighborhood offer greater access to medical care.

: This report summarizes findings from other studies that use data from the federal government’s Consumer Expenditure Survey and the 2002 National Survey of America’s Families.

: The report includes analysis of findings from the 2002 National Survey of America’s Families, which are preliminary and not based on longitudinal data.

Find this study:

Back to top



The Center gratefully acknowledges the support of the John D. and Catherine T. MacArthur Foundation for these materials. The findings and conclusions they present, however, are those of the authors, and do not necessarily reflect the opinions of the Foundation.