![Sample our Built Environment journals, sign in here to start your access, Latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5926/TOC-Subject-Sample-2021-Built-Environment.jpg"})
Abstract
Federal housing subsidies are allocated without regard to spatial differences in the cost of living or quality of life. In this article, we calculate housing subsidy payments for participants in the Housing Choice Voucher (HCV) program and demonstrate that these subsidies are significantly related to metropolitan quality-of-life differentials. We then estimate amenity-adjusted subsidies and compare these estimates with data from the U.S. Department of Housing and Urban Development’s Location Affordability Portal. Our analysis yields three insights regarding the relationship between federal housing assistance payments (HAP), metropolitan quality-of-life differentials, and transportation cost burdens. First, HCV HAP show a strong inverse correlation with household transportation expenditures, and this is particularly pronounced for low-income households. Thus, HAP do not address location affordability because those living in high-transportation cost metropolitan areas receive the lowest housing subsidies. Second, we present evidence that HAP are positively related to metropolitan quality-of-life differentials. This suggests that high-amenity metropolitan areas also tend to be the most affordable from a transportation cost perspective. Third, our proposed amenity-adjusted HAP strongly reduce the inverse relationship between HAP and transportation cost burdens.
Acknowledgments
Several colleagues provided helpful comments and suggestions on this research, including J. Nicholas Dadson, Lan Deng, Jonathan Levine, Kirk McClure, and participants at seminar and conference presentations at the Annual Meetings of the Association of Collegiate Schools of Planning, the Regional Science Association International, the Urban Affairs Associations, and the Urban Economics Association. We also thank the special issue editor, John Renne, and several anonymous referees for valuable comments and advice. Bieri acknowledges financial support from the University of Michigan’s Graham Institute on Environmental Sustainability. The usual disclaimers apply.
Notes
1. The closest work in spirit to this aspect of our article is Fisher, Pollakowski, and Zabel (Citation2009) who propose an amenity-based housing affordability index for the Boston, Massachusetts, metro area. Our work differs in scope and focus in that we look at the interplay of amenities and housing subsidies at the national level.
2. Worst-case housing needs are experienced by unassisted very low-income renters who either pay more than one half of their monthly income for rent, or live in severely inadequate conditions, or both. HUD defines very low-income as below 50% of the local area median income, and extremely low-income as below 30% of area median income. Throughout this article, we focus on very low-income households in the sense of the HUD definition.
3. The Quality Housing and Work Responsibility Act of 1998, which consolidated the Section 8 certificate and voucher programs into the current HCV program, required that no less than 75% of any local public housing agency’s new HCV be awarded to families earning extremely low incomes (income at or below 30% of the area median income).
4. Because comprehensive microdata on HCV program applications are not available, we define participation rates as a share of HCV recipients relative to the theoretical base of all households that qualify, calculated using HUD income limits and Public Use Microdata Sample (PUMS) from the 2000 U.S. Census.
5. As we discuss below in more detail, we use data on metro-level transportation cost burdens for low-income households from the LAP to empirically evaluate the link between housing assistance payments and the notion of location affordability.
6. All features of the comparative rankings translate to the full sample of all 363 metro areas. If anything, the city size bias increases substantially as all of the 10 most affordable metro areas have a population of less than 350,000 inhabitants. For the full sample, the Spearman rank correlation between H and H+T is 0.32 (vs. −0.21 between H and T, and 0.83 between T and H+T).
7. If a household leases a unit with a gross rent above the payment standard for the family, the household share is plus any amount by which the gross rent exceeds the payment standard. Although HCV guidelines place great emphasis on communicating this concept to participating households while searching for housing, the household share cannot be calculated until a unit is selected.
8. See McClure (Citation2008, Citation2014) for more technical details on HCV program administration and program performance statistics.
9. Relative to the quality-of-life literature, the BKP estimates incorporate a number of important improvements, including (a) using PUMS data to develop a comprehensive national database on rents, wages, and spatially delineated amenities; (b) migration data to account for moving costs; (c) addressing spatial variation in the user cost of housing; (d) overcoming the conflation of spatial fixed effects with unobserved attributes of workers and houses; and (e) adapting a control function approach to address spatial Roy sorting.
10. Different levels of mobility among households are not inconsistent with the spatial equilibrium assumption per se as in this theoretical setting households equalize type-specific utility which will reflect the fact that moving costs of low-income households are high compared with those of average households. See Bieri and Dawkins (Citation2016) for more details.
11. In BKP, the main specification of the first-stage model is based on a semi-log parameterization, such that after removing the variation rents and wages from observable attributes of houses and workers, any remaining variation across locations will be absorbed by the fixed effects. However, because the fixed effects will conflate the implicit prices for amenities with the implicit prices for unobserved characteristics of houses and workers, additional steps are required to calculate implicit amenity expenditures (see Bieri et al., Citation2014, p. 19).
12. Recall from the preceding analysis that throughout this article we have been focusing on very low-income households in the sense of the HUD definition—that is, those households with incomes at 50% of AMI.
13. As an alternative to this scenario, a coverage-neutral scenario could level the playing field away from high-amenity metro areas, for example by adjusting housing subsidies relative to a regional baseline of low-income amenity expenditures. This would imply that high-amenity metro areas would see a reduction in subsidy levels, whereas less-attractive metro areas would see an increase in housing assistance payments. Such a scenario could be viewed as a middle-ground policy alternative that seeks to improve locational outcomes without substantial increases in overall rent burdens (calculations for such a scenario are available upon request).
14. HUD has identified three sources of errors contributing to improper rent subsidy payments: (a) incorrect subsidy determinations by program administrators; (b) unreported tenant income; and (c) incorrect billing. HUD has attempted to estimate the amounts of improper subsidies attributable to each source but has developed reliable estimates for only the first—and likely largest—source. HUD paid an estimated $1.4 billion in gross improper subsidies (consisting of $896 million in overpayments and $519 million in underpayments) in fiscal year 2003 as a result of program administrator errors—a 39% decline from HUD’s fiscal year 2000 (baseline) estimate.
15. This research also contributes to the Bureau of Economic Analysis’ recent green national accounting efforts of producing a credible approximation to implicit expenditures on nonmarket goods that complement the corresponding measures of personal consumption expenditures on private goods in the National Income and Product Accounts (National Research Council, Citation2005).
16. See, for example, the proposed but now defunct Transportation and Housing Affordability Transparency Act (THATA; H.R. 5824, 111th Congress, 2nd Session, June 2010).
17. We are grateful to one of the reviewers for this point.