Can Affordable Housing Policies Reduce Health Disparities?

This article was originally published November 2, 2023 on the Housing Perspectives blog of the Harvard Joint Center for Housing Studies.

Despite substantial government investment aimed at reducing health disparities, large differences in health persist across geographic and racial lines in the United States. One prominent theory is that these health disparities are driven in part by disparities in the neighborhood environments where people live. If this is the case, then housing policies that desegregate, and therefore reduce disparities in neighborhood environments, may be an effective means for reducing disparities in health.

However, relatively little is known about the health effects of housing policies that explicitly seek to reverse neighborhood disparities. In "Early-Life Health Impacts of Affordable Housing in Higher-Income Areas: Evidence from Massachusetts Chapter 40B," a paper that won the Center’s Best Paper on Housing Prize earlier this year, I address this gap by studying the impacts of one such policy on the health of pregnant parents and their newborn infants. I focus specifically on Massachusetts Chapter 40B, a fair-share housing law enacted in 1969 that seeks to facilitate moves for low- and moderate-income households to higher-income areas by increasing the supply of affordable housing in those areas. Since it was enacted in 1969, 40B has facilitated the construction of more than 18,000 affordable rental and ownership units largely in the wealthy suburbs around Boston.

To study how moving to 40B housing impacts health, I use a novel dataset on 40B building addresses linked to birth records and highly detailed longitudinal address data. These data allow me to compare the health outcomes of pregnant people and their newborn infants in families that moved to 40B units to the health of people who moved at the same time from similar neighborhoods, but to less affluent areas.

I find that moving to 40B rental housing produces large, clinically relevant improvements in newborn infants' health and some small gains in birthing parents’ health. In contrast, I find no evidence of any health effects for people who bought units in a project approved via the 40B process. This may be due to the fact that compared to owners, 40B renters tended to be less educated, be in worse health prior to moving to 40B housing, and/or moved to 40B homes from lower income neighborhoods.

What is it about moving to 40B rental housing that might improve the health of newborn infants? One potential explanation is that there is something about moving to subsidized housing that is driving better birth outcomes. To test this, I compare the health effects of moving to 40B housing with the health effects of moving to housing built under the other major source of new affordable housing in Massachusetts: the federal Low-Income Housing Tax Credit (LIHTC) program. Unlike 40B, many LIHTC-funded projects are located in neighborhoods that have similar (or worse) conditions than their residents’ previous neighborhoods. LIHTC tenants likely have slightly lower incomes than 40B tenants. However using data from birth records, I find no evidence that moving to LIHTC housing affects newborn infant health, though it does produce similar small improvements to 40B in some birthing parents' health outcomes. This implies that in contrast to infant health, parental health may be improved by a general feature of subsidized housing, such as the additional income or stability provided by the subsidy.

Another possibility – which serves as the central motivation for my research – is that 40B facilitates moves that change neighborhood environments which, in turn, improve infant health. By changing the environments where people live, a new neighborhood could impact infant health both directly and indirectly. For example, neighborhoods may have different levels of pollution which has been shown to directly impact birth outcomes, such as whether an infant is born at a healthy weight. Neighborhood settings could also affect infants’ health by fostering changes to the health or health behaviors of the birthing parent.

My analyses support this neighborhood relocation hypothesis. First, moving to 40B housing facilitates larger changes in neighborhood environments than might be possible when moving to other types of housing. That is, households that move into 40B rental units relocate to neighborhoods with higher incomes, lower poverty and incarceration rates, and less exposure to pollutants than the households who move elsewhere. This stands in contrast to my sample of the LIHTC program, which tended to facilitate moves to lower-income, higher-poverty neighborhoods than would otherwise be possible.

Moreover, improvements in birth outcomes are larger for beneficiaries who experienced the greatest changes in neighborhood environments. For instance, health gains were greatest among infants born to Black birthing parents who also experienced the largest changes in median household income, male incarceration rates, and pollution exposure. Similarly, children born to people who moved to 40B rentals from higher poverty neighborhoods and neighborhoods with a greater proportion of non-white or Hispanic residents had larger gains in birth weight and infant gestation than children born to those who moved from lower poverty and whiter neighborhoods. These results, combined with the lack of evidence that moving to LIHTC housing affects newborn infant health, suggest that infant health improvements are partially explained by changes to neighborhood environments.

A large and growing body of research indicates that overly restrictive zoning laws have fostered racial segregation, inflated house prices, and led to less equity in political processes. This research shows that housing policies which aim to overcome restrictive zoning may also address important health disparities as well.