Devolution, Deconcentration and the Non-profit Sector: Rental Housing Subsidies in the Greater Boston Area

Abstract

In an effort to promote deconcentration and flexibility, the non-profit sector has been called to play a greater role in the production and management of federally subsidised rental housing in the United States. This paper explores the interjurisdictional distribution of subsidised units in the Greater Boston area and investigates how it is shaped by housing non-profits, controlling for demographic, fiscal, economic and regulatory characteristics. Findings suggest that housing non-profits contribute positively to the proportion of subsidised units in a given municipality. Yet, the geographical concentration of such agencies in poorer and heavily subsidised jurisdictions, due in part to restrictive zoning regulations, prevents them from promoting geographical mobility among assisted households.

Notes

Categorical grants provide federal funds to sub-national governments as entitlements for specific purposes, while block grants provide general revenue to local government without specifying particular uses. The latter give states and local government greater discretion in fund allocation and often take resources away from social programmes based on needs. For example, specific public housing programmes have been replaced by Community Development Block Grants, which can be used to build or maintain affordable housing units but are often used to promote local economic development projects that only indirectly, if at all, benefit low-income residents.

Between 1980 and 2000, federal expenditure for housing programmes fell from almost $70 billion (in 2000 dollars) to $24.8 billion (Weismann, 2002, p. 153).

The 2004 US Census Boston–Cambridge–Quincy New England City and Town Area (NECTA) definition is used and only Massachusetts county sub-divisions are included in the analysis. NECTA are defined using commuting patterns and therefore correspond roughly to local labour markets. This differs from the Metropolitan Statistical Areas that are based on county boundaries which may not properly reflect metropolitan economic activity.

Massachusetts' Chapter 40B law, approved in 1969, effectively allows developers to appeal and override local zoning regulations in cases where less than 10 per cent of housing units in a given municipality are considered affordable (CHAPA, 2009).

Other programmes include older HUD and FHA programmes such as Section 8 Moderate rehabilitation, Section 8 New and Substantial Rehabilitation and Section 236 of the National Housing Act which provide subsidies to private owners of units rented to low-income tenants. Data on the number of units and their location are very reliable; they are based on HUD's administrative reports and are rechecked internally against multiple sources of records.

These data have been successfully used in other studies of non-profit locations (Bielefeld, 2000; Peck, 2008; Joassart-Marcelli and Wolch, 2004). Joassart-Marcelli and Wolch (2004) address the validity of using these data for geographical analyses.

These are organisations in the ‘L’ category of the National Taxonomy of Exempt Entities. The data exclude organisations with annual revenue below $25 000 and religious organisations not required to file form 990. In the case of housing, it is unlikely that many effective organisations with a budget below $25 000 exist. Moreover, while religious organisations are involved in the production of housing, they usually concentrate their effort on homeless shelters which are excluded from this analysis. There are no theoretical reasons to believe that the geographical distribution of religious housing organisations would differ significantly from that of other non-profits.

The Massachusetts Nonprofit Housing Association consists of nine regional agencies covering every town and city in the state. In the Greater Boston area, these agencies are the Metropolitan Boston Housing Partnership located in Boston, Community Teamwork in Lowell, the South Middlesex Opportunity Council in Framingham and the South Shore Housing Development Corporation in Kingston. Although these agencies cover designated areas, their projects are often restricted to a few cities or towns. For instance, the South Shore Housing Development Corporation manages 1181 affordable rental housing units of which 523 are located in Kingston.

To adjust for the high cost of living in the Greater Boston area, 150 per cent of the official poverty threshold is used to define poverty. For example, in 2000, this amounted to $20 811 per year for a family of one adult and two children.

Many of these programmes were put in place in the 1970s and 1980s for a period of 30 years. As we reach the end of the affordability requirement periods, owners are no longer legally required to rent to low-income households and/or charge limited rents.

Note that the dot density layer that shows the distribution of federally subsidised units is based on census-tract data and thus reflects the distribution of units within municipal boundaries in greater detail than would the municipal aggregations.

See note 10.

Data on employment are from the Census ZIP code business patterns for 1994 and 2000. ZIP codes can be allocated to towns and cities. In cases where a ZIP code did not fall completely within a town's boundaries, it was allocated to the town with the greatest share of its area.

To compute this figure, I used the 2006 MassGIS trains and MBTA rapid transit layers (MassGIS, 2004, 2006) to select tracts with their centroid within one mile of public transit stations. I then summed the number of subsidised rental units located within these tracts.

Local fiscal capacity is estimated by applying the representative regional tax rate to each town or city's aggregate household income. The representative tax rate is obtained by dividing the region's local revenue (i.e. total revenue minus intergovernmental revenue) by the region's aggregate household income. This measure captures how much a town would be able to get if it were to raise local revenues at a rate similar to the region's average. Aggregate household income data come from the 2000 Census of Population and Housing. Data on municipal revenue, by source, come from the Massachusetts Department of Revenue (2000).

Minimum lot sizes are computed using MassGIS data on land use zoning (2002) and 2000 Census of Population and Housing data on housing units.

The Citizens' Housing and Planning Association (CHAPA, 2008) provides many examples of such legal battles that have resulted in long delays and ultimately reductions in the number of affordable units built. According to their report, abutter and municipal lawsuits in Massachusetts have delayed the development of almost 10 000 units since 2000 by an average of 2 years. For example, in 2004, the non-profit organisation Jewish Community Housing for the Elderly (JCHE), requested a permit to build a 150-unit mixed-income elderly housing development with 90 affordable units in Framigham with partial funding from the US Department of Housing and Urban Development. Following opposition from local residents, the project has been put on hold by legal battles in the Land Court and the Appeals Court. No units have been built despite available funding.

Additional information

Notes on contributors

Pascale Joassart-Marcelli

The author received a US Department of Housing and Urban Developments Urban Scholar Fellowship in 2004–05 that allowed her to begin this research project. During that time, she was a Visiting Fellow at Harvard University's Joint Center for Housing Studies. She would like to thank Jennifer Raymond for research assistance and Nic Retsinas and Enrico Marcelli for useful comments and suggestions.