HUD Policy Changes to Improve Access to Low Poverty Neighborhoods
Katherine O’Regan, Assistant Secretary for Policy Development and Research.
Recent research by Harvard economist Raj Chetty and events in Baltimore and beyond draw a direct connection between place and upward economic mobility. Causal confirmation that low-income children have significantly higher earnings as adults and are more likely to go to college if raised in low-poverty neighborhoods has renewed interest in how to better accomplish the goal of providing good, healthy environments for families with children. Two impending changes to HUD policies aim to do exactly that: improve access to low poverty neighborhoods for assisted households.
1) Enabling Voucher Holders To Access Better Neighborhoods: Small Area Fair Market Rents
HUD’s Housing Choice Voucher (HCV) program has been criticized for not adequately helping families access low-poverty neighborhoods. One frequently cited reason is that rent payment standards, which are based on HUD’s Fair Market Rents (FMRs) and are currently determined for entire metropolitan areas, are insufficient. To calculate FMRs, HUD estimates the distribution of rents paid by recent movers in each metropolitan statistical area (MSA) and sets the FMR at the 40th percentile in that distribution — in other words, at a level that should be sufficient to rent 40 percent of the units in the MSA. Those units, however, may be located primarily in neighborhoods of high poverty. Within an MSA, public housing agencies (PHAs) are permitted to set payment standards between 90 and 110 percent of FMR, providing some flexibility to accommodate the higher rents of lower-poverty areas, but flexibility is generally limited.
In 2000, recognizing that in some markets the rent payment standard at the 40th percentile may not be enough to provide a broad set of housing opportunities, HUD established the use of 50th percentile FMRs for metropolitan areas with particularly high concentrations of voucher holders. (Baltimore is a 50th percentile MSA.) However, experience and research confirm that this approach has not been successful. Specifically, increasing the rent payment standard to the 50th percentile may still be insufficient to provide access to a significant share of lower-poverty neighborhoods while also increasing subsidies in high-poverty neighborhoods. Lifting maximum rents across the board may not provide the incentive or ability to differentially “tilt” HCV households toward better neighborhoods.
To that end, HUD has developed a methodology for setting FMRs at the ZIP Code level to create Small Area FMRs (SAFMRs), a more effective means of deconcentrating voucher holders. HUD mandated SAFMRs as part of a 2010 litigation settlement in Dallas and began a SAFMR demonstration in five PHAs in 2012.
Key SAFMR Research Findings
Recent research by former Office of Policy Development and Research colleague Rob Collinson (now at NYU) and his coauthor Peter Ganong (Harvard) provides evidence on the relative efficacy of SAFMRs and the 50th percentile.
Using multiple approaches to assess the effect of raising FMRs uniformly across a market (such as the 50th percentile policy), they find that such increases have a minimal effect on the housing and neighborhood quality of tenants but do result in rents increasing. They estimate that up to 89 percent of the increased subsidy benefits landlords in the form of higher rents. This last point confirms anecdotal evidence that in some low-rent neighborhoods, FMRs are above market rents for many units, and the concentration of voucher holders may be driving up rents.
In contrast, using data from the Dallas MSA, the authors find that under SAFMRs, HCV households have deconcentrated, moving to better neighborhoods. Specifically, prior to this policy, Dallas voucher holders who moved did not move to neighborhoods of higher quality; post-policy, voucher holders exited the lowest-quality neighborhoods in the inner city, moving further east and south to neighborhoods with significantly lower levels of crime and poverty. Importantly, this occurred at essentially the same total cost per voucher, and so did not come at the cost of serving fewer voucher households. It is worth emphasizing that these improved outcomes were at the same cost per voucher as the 40th percentile FMR; – so relative to the 50th percentile (our existing policy for deconcentration), the Dallas evidence suggests SAFMRs would be both more effective and less costly.
Given this evidence of a more effective policy, HUD published an Advanced Notice of Proposed Rulemaking in the Federal Register on June 2 to solicit public comments on replacing the 50th percentile standard with SAFMRs for the HCV program, along with changes to that program to improve its targeting to areas with the greatest need for and likelihood of benefiting from SAFMRs, and to eliminate areas cycling in and out of the program. We look forward to receiving public comments between now and July 2.
2) Better Leveraging LIHTC To Create Affordable Housing in Low-Poverty Areas Through Small DDAs
A second policy advance in which HUD plays a role impacts the Low-Income Housing Tax Credit (LIHTC) program, which is overseen by the Internal Revenue Service. HUD is responsible for the designation of Difficult Development Areas (DDAs), which, along with Qualified Census Tracts (QCTs), confer 30 percent more tax credits to properties located in these HUD-designated areas.
This increase in tax credits, often referred to as the basis boost, applies to both the competitive tax credits and bond-generated credits. Given the additional flexibility states have in granting basis boosts in the competitive program authorized in the Housing and Economic Recovery Act of 2008, federal designation of DDAs (and QCTs) are mainly relevant for bond-financed properties.
DDAs are meant to be areas that have relatively high construction, land, and utility costs, with a national cap that DDAs contain no more than 20 percent of the population living in metropolitan areas. HUD proxies for those costs using Fair Market Rents (FMRs), so DDAs are currently measured for entire metropolitan areas, meaning that all neighborhoods in the MSA receive this designation regardless of costs or rent.
In 2016, HUD will be moving to Small DDAs — designations (based on SAFMRs) to determine high-cost ZIP Codes. Using smaller geography will better target federal resources to higher-rent areas throughout the country rather than all neighborhoods in high-cost metropolitan areas. In 2015, 35 MSAs were designated as DDAs, but if SDDAs were in effect, all 233 MSAs would have at least one. (The Baltimore MSA, which is not a DDA, would have received 56 SDDAs). Based on 2015 numbers, the average poverty rate in DDAs would drop from 15.4 percent to 10.1 percent. This change should encourage more LIHTC project development (particularly bond-financed development) in higher-opportunity neighborhoods than the current policy does.
Given the national cap, the policy is not a win for all. The areas that currently benefit the most from DDA designations will lose some amount of subsidy. Furthermore, providing the basis boost doesn’t necessarily mean that development will occur. It is a start, and it may be a tool, but deals and local interest are necessary for the basis boost to result in housing.
These two polices each focus on ways in which housing assistance might further the goals of deconcentration of poverty. That is one arm of a two-prong approach — the other continues to be about improving the places where families live. It is critical that HUD — and other actors — pursue both: improve the ability of families to move to better places and make the places where families live better.
PD&R Edge Archives
Research & Publications
Housing Needs of Native Hawaiians: A Report From the Assessment of American Indian, Alaska Native, and Native Hawaiian Housing Needs
Energy Performance Contracting in HUD’s Public Housing Stock: A Brief Overview
Assessment of ARRA Green and Energy Retrofits in HUD-Subsidized Housing