A paper published by the Urban Institute, “Spatial Mismatch and Federally Supported Rental Housing,” found that assisted households had substantially lower access to jobs within a standard commuting distance than did extremely low-income unassisted rental households. The study included households living in public housing, those receiving project-based rental assistance, and those using Housing Choice Vouchers. These federally assisted rental households had, on average, 6,032 more job seekers than available jobs within a 6.3-mile radius. In contrast, unassisted extremely low-income households had 3,056 more job seekers than available jobs within a 6.3-mile radius.
The authors analyzed 16 Metropolitan Statistical Areas (MSAs) for spatial mismatch between the availability of lower-wage jobs and the location of subsidized housing in 2015. The authors used HUD’s Picture of Subsidized Households (POSH) for 2015 to determine the number of assisted households within each census tract, including households in public housing, project-based Section 8 rental assistance, moderate rehabilitation, supportive-housing programs, below-market interest-rate programs, and those receiving Housing Choice Vouchers. They used the 2013-2017 American Community Survey (ACS) five-year estimates to calculate the number of extremely low-income households eligible for assistance in neighborhoods within those MSAs. They counted a household as eligible for assistance if its income was less than 30% of the area median income for a household of four people. The difference between the ACS estimate of extremely low-income households and the POSH counts of assisted households provided an estimate of the number of unassisted extremely low-income households. Finally, Snagajob, the largest online marketplace for hourly jobs, provided the authors with data about job listings and the number of job applicants in 2015 at the zip code level. Most of those job listings were for full-time positions and many were minimum-wage positions. As of 2017, Snagajob listings accounted for 13% of all new hires in the 16 MSAs.
In order to assess spatial mismatch, the authors counted the number of jobs available within a 6.3 mile radius of each zip code and compared those job listings with estimates of the number of assisted and unassisted extremely low-income households in each area. (They chose 6.3 miles as a reasonable commuting distance because it was the average distance between job applicants’ homes and the workplaces to which they applied on Snagajob.) The authors found that the average spatial mismatch for assisted households was significantly worse than the mismatch for unassisted extremely low-income households. The average assisted household lived in a neighborhood with 6,032 more job seekers than job postings within 6.3 miles while the average unassisted household with an extremely low income lived in a neighborhood with 3,056 more job seekers than jobs within 6.3 miles. Among residents of public housing, the ratio is even worse—they lived in neighborhoods with over 8,000 more job seekers than jobs within 6.3 miles.
The authors also describe variation among the 16 MSAs. In Boston, Minneapolis, Nashville, and Seattle, assisted households lived in close proximity to more jobs than unassisted households with extremely low incomes. Atlanta, Chicago, Los Angeles, and Washington, DC and have a much larger spatial mismatch for assisted households. The report provides closer analyses of the spatial mismatch in Atlanta, Chicago, Seattle, and Washington, DC.
The authors conclude that many federally assisted low-income households may not have the flexibility to live in or move to close-proximity to jobs. The report’s findings could be read alongside another recent study, “Does Jobs Proximity Matter in the Housing Choice Voucher Program?” (see Memo, 11/18/19), which found that employed recipients of vouchers were not more likely to reside closer to jobs, and that a greater density of jobs in close proximity was not correlated with greater earnings for households. Taken together, these results emphasize a greater need for research into potential causal relationships between subsidized housing and access to jobs.
The Urban Institute report is at: https://urbn.is/32NE40U