Exploring the Spatial Distribution of Low Income Housing Tax Credit Properties

Publication (Report)

Beginning in 1972 with the creation of the Section 8 (Housing Choice Voucher) rental housing assistance program, the U.S. Department of Housing and Urban Development (HUD) has placed increasing emphasis on dispersing the geographic pattern of housing assistance. Responding in part to decades of research pointing to the links between concentrated poverty, racial segregation, and barriers to economic opportunity (see Ellen and Turner (1997) for a review), HUD programs now seek to enhance low income households’ access to a wider variety of neighborhoods with more desirable public services and amenities.

 

This emphasis on de-concentrating affordable housing assistance was expanded in the 1990s with the creation of the HOPE VI program, which sought to replace severely-distressed public housing with new mixed income housing developments, and the Moving to Opportunity demonstration initiative, which provided rental housing assistance to those seeking to move out of high-poverty neighborhoods. HUD’s FY 2011 budget proposes an expansion of these efforts with its “Transforming Rental Assistance” initiative, which seeks to expand housing choice and increase the share of HUD-assisted households living in low-poverty communities.

 

While expanding housing choice through the dispersal of housing assistance has become a primary objective of HUD, it is not clear whether this objective is being met in the nation’s largest affordable housing subsidy program, the Low Income Housing Tax Credit (LIHTC). The LIHTC program was created as part of the Tax Reform Act of 1986 and is jointly-administered by the U.S. Department of Treasury’s Internal Revenue Service (IRS) (26 U.S.C. § 42) and local and state housing finance agencies. Unlike affordable housing programs administered by HUD, geographic dispersal of LIHTC subsidies is not an explicit intent of the program. Since the IRS is primarily concerned with administration of the tax code, it may not have an interest or capacity to aggressively monitor the geographic distribution of LIHTC assistance to ensure consistency with HUD’s housing assistance objectives. Furthermore, LIHTC provisions enacted in 1989 which increase a developer’s credit eligibility if they propose developing units in high-poverty census tracts may actually encourage the concentration of LIHTC housing assistance within high-poverty areas (Freeman 2004). Despite these apparent policy inconsistencies, little research has been done to evaluate thegeographic pattern of LIHTC properties to determine if the spatial distribution of properties exhibits higher than expected patterns of clustering. A limited number of studies have examined the aggregate characteristics of census tracts where LIHTC units have been placed, along with the spatial relationships among LIHTC census tracts, but no study to date has taken advantage of HUD’s publicly-available LIHTC database to examine the geographic pattern of the LIHTC properties themselves.

 

This study fills this gap. For the nation’s 10 largest metropolitan areas, the geographic pattern of LIHTC properties is examined to determine if the pattern of clustering is higher than what would be expected under conditions of “complete spatial randomness.” Local patterns of clustering within each metropolitan area are also examined, followed by a descriptive analysis of the average characteristics (e.g. number of units per property, percent new construction, location in central cities, census tract poverty rates, census tract racial composition) of properties located within local clusters. The analysis concludes with an examination of patterns of spatial diffusion to examine whether neighboring LIHTC properties are constructed within a similar time frame, in years. Together, each of these analyses paints a picture of the geographic distribution of LIHTC properties and suggests possible policy responses to ensure that LIHTC policies are aligned with national poverty deconcentration objectives.

TitleExploring the Spatial Distribution of Low Income Housing Tax Credit Properties
Publication TypeReport
Year of Publication2011
AuthorsDawkins, Casey
Pagination47
Date Published02/2011
InstitutionU.S. Department of Housing and Urban Development
TypeAssisted Housing Research Cadre Report
Abstract

Beginning in 1972 with the creation of the Section 8 (Housing Choice Voucher) rental housing assistance program, the U.S. Department of Housing and Urban Development (HUD) has placed increasing emphasis on dispersing the geographic pattern of housing assistance. Responding in part to decades of research pointing to the links between concentrated poverty, racial segregation, and barriers to economic opportunity (see Ellen and Turner (1997) for a review), HUD programs now seek to enhance low income households’ access to a wider variety of neighborhoods with more desirable public services and amenities.

 

This emphasis on de-concentrating affordable housing assistance was expanded in the 1990s with the creation of the HOPE VI program, which sought to replace severely-distressed public housing with new mixed income housing developments, and the Moving to Opportunity demonstration initiative, which provided rental housing assistance to those seeking to move out of high-poverty neighborhoods. HUD’s FY 2011 budget proposes an expansion of these efforts with its “Transforming Rental Assistance” initiative, which seeks to expand housing choice and increase the share of HUD-assisted households living in low-poverty communities.

 

While expanding housing choice through the dispersal of housing assistance has become a primary objective of HUD, it is not clear whether this objective is being met in the nation’s largest affordable housing subsidy program, the Low Income Housing Tax Credit (LIHTC). The LIHTC program was created as part of the Tax Reform Act of 1986 and is jointly-administered by the U.S. Department of Treasury’s Internal Revenue Service (IRS) (26 U.S.C. § 42) and local and state housing finance agencies. Unlike affordable housing programs administered by HUD, geographic dispersal of LIHTC subsidies is not an explicit intent of the program. Since the IRS is primarily concerned with administration of the tax code, it may not have an interest or capacity to aggressively monitor the geographic distribution of LIHTC assistance to ensure consistency with HUD’s housing assistance objectives. Furthermore, LIHTC provisions enacted in 1989 which increase a developer’s credit eligibility if they propose developing units in high-poverty census tracts may actually encourage the concentration of LIHTC housing assistance within high-poverty areas (Freeman 2004). Despite these apparent policy inconsistencies, little research has been done to evaluate thegeographic pattern of LIHTC properties to determine if the spatial distribution of properties exhibits higher than expected patterns of clustering. A limited number of studies have examined the aggregate characteristics of census tracts where LIHTC units have been placed, along with the spatial relationships among LIHTC census tracts, but no study to date has taken advantage of HUD’s publicly-available LIHTC database to examine the geographic pattern of the LIHTC properties themselves.

 

This study fills this gap. For the nation’s 10 largest metropolitan areas, the geographic pattern of LIHTC properties is examined to determine if the pattern of clustering is higher than what would be expected under conditions of “complete spatial randomness.” Local patterns of clustering within each metropolitan area are also examined, followed by a descriptive analysis of the average characteristics (e.g. number of units per property, percent new construction, location in central cities, census tract poverty rates, census tract racial composition) of properties located within local clusters. The analysis concludes with an examination of patterns of spatial diffusion to examine whether neighboring LIHTC properties are constructed within a similar time frame, in years. Together, each of these analyses paints a picture of the geographic distribution of LIHTC properties and suggests possible policy responses to ensure that LIHTC policies are aligned with national poverty deconcentration objectives.