• Research & Data
• Education
• Consumer Credit & Payments
• Bank Resources
• Community Development

• Newsroom
• Careers
• Publications

Thursday, May 23, 2013

[ – ] Text Size [ + ]  |  Print Page

Home > Community Development > Publications > Special Reports > Affordability and Availability of Rental Housing in Pennsylvania > Chapter 4: A Mid-Decade Update: Housing Conditions in 2005-06

Affordability and Availability of Rental Housing in Pennsylvania

By Erin Mierzwa, Federal Reserve Bank of Philadelphia, and Kathryn P. Nelson, Ph.D., Consultant, Affordable Housing Needs, with Harriet Newburger Ph.D., Federal Reserve Bank of Philadelphia

Chapter 4: A Mid-Decade Update: Housing Conditions in 2005-06

1 | 2 | 3

II. Findings at Regional Levels49

DCED Regions

To examine housing conditions within Pennsylvania, this chapter first compares 2000 and 2005-06 data in the six regions in Pennsylvania defined by DCED, which are shown in Map 3.50 The DCED regions are particularly relevant to rental housing policy because the Pennsylvania Housing Finance Agency (PHFA) has a regional set-aside for the allocation of low income housing tax credits (LIHTCs) based on these DCED regions.51 LIHTCs have been a major source of affordable rental housing over the last 15 years.

Because DCED regions are larger than the consolidated PUMAs analyzed in the next section, it is possible to estimate rental housing conditions at the DCED regional level more precisely. In turn, differences between 2000 and 2005-06 are more frequently statistically significant at the 90 percent confidence level. All commentary on changes in this chapter focuses on differences that were statistically significant at the 90 percent confidence level, unless otherwise noted.52

The data tables for DCED regions in Pennsylvania give additional detail for Region 1, the Philadelphia metropolitan division, because that region has the most renters and also the greatest shortage of affordable and available housing units. The central county, Philadelphia, is distinguished from its suburbs: Bucks, Chester, Delaware and Montgomery counties.

Rental Housing Conditions in 2005-06 by DCED Region

As noted above, statewide, the percentage of ELI renter households in Pennsylvania rose by 4 percentage points between 2000 and 2005-06. This represented an increase of approximately 50,000 ELI renter households, from 334,600 to 384,800. As Table 17 shows, each region experienced similar income shifts. The largest increases in ELI households occurred in Region 6, which includes Erie, and in the city of Philadelphia.

Region 1 had the highest share of ELI renter households (32 percent). Within this region, the city of Philadelphia had a much larger share of ELI renter households than its suburban counties did.53

Cost Burden. Cost burden pressures worsened in each DCED region between 2000 and 2005-06, particularly for ELI renters. But because part of the increase in cost burden shown by our data reflects procedural differences (as described in Appendix E), we focus on differences among regions in 2005-06.

As Map 4 illustrates, ELI renter households in the Philadelphia area faced the greatest cost burden pressure, with three-fourths having a severe cost burden. The incidence of severe cost burden was least common among ELI renters in the North Central, Southwest and Northeast regions.

As Table 18 details, within the Philadelphia metropolitan division, ELI renters were more likely to have cost burdens in the suburbs. There, four of five ELI renter households had severe cost burdens. VLI renters were also more likely to have cost burdens in Philadelphia’s suburbs than in other areas of the state, and almost a third had severe cost burdens there. In all regions of the state, LI renters very seldom had severe cost burdens.

Affordable Rental Housing Shortages. At the regional level as in the state, shortages of units affordable and available to ELI renters were greatest in every region. As Map 5 illustrates, Regions 1 and 3 had the greatest shortages of housing both affordable and available to ELI renter households. The shortage of affordable and available housing units per 100 ELI renter households was least pressing in Region 5, the Southwest, but there were still only 51 affordable and available units for every 100 ELI renter households there.

Between 2000 and 2005-06, the numbers of units affordable to ELI households fell sharply and significantly in each region, and the numbers of units affordable to renters with incomes between 0-50 percent of AMI also fell significantly54 (Table 19). Significant drops in units affordable and available to ELI renter households also occurred in Regions 6, 3, 2, and 1. (Within Region 1, the drop was larger and significant only in the Philadelphia suburbs.)

Region 1 remained the region with the greatest shortages of affordable and available housing for both ELI renters and renters with income below 50 percent of AMI. Within the region, the Philadelphia suburban counties had much less affordable and available housing than the city itself. The suburbs had only 25 affordable and available units per 100 ELI renters and 60 affordable and available units per 100 renters at 0-50 percent AMI.

Region 6, which experienced the greatest increase in the percentage of ELI renter households between 2000 and 2005-06, also experienced the largest reduction in the number of affordable and available housing units, from 58 to 43 units per 100 ELI renter households.

Looking more broadly at the experience of all lower-income renters by region in both 2000 and 2005-06, the total number of units affordable and available to renters with incomes at or below 80 percent of AMI slightly exceeded the number of such renters in all regions. Said differently, the regional supply and demand were roughly in balance for renters with incomes at or below 80 percent of AMI.

In absolute terms in 2005-06, Region 1 had the greatest shortage of affordable and available housing units for ELI renter households (over 90,000 units, 41 percent of the state’s total) and also for renters with incomes between 0-50 percent of AMI (48 percent of state’s total). Shortages were also substantial in the Southwest and Northeast (Regions 5 and 2) (Table 20).

Within Region 1, the city of Philadelphia had a much larger shortage of affordable units available to ELI renter households than its four suburban counties. Notably, however, the suburban shortage of affordable units available to renters with incomes between 0-50 percent of AMI was more than double that of the city (32,800 vs. 14,950). The difference between the two locations suggests that, in the suburbs, the number of renters with incomes between 30-50 percent of AMI roughly equaled the number of units with rents affordable to that income range. Philadelphia city, by contrast, apparently had many more units affordable to incomes between 30-50 percent of AMI than renters in that income range.

Rental Vacancy Rates by Unit Affordability to Lower-Income Households. Vacancy rates are often used as indicators of housing supply, but they can be difficult to interpret, particularly when drawn from small samples such as the ACS. For example, a high vacancy rate could signal an adequate supply of rental housing, but it also could reflect too many units of poor quality or units in locations with declining demand. But this indicator can help distinguish tight markets with growing demand from loose markets with less demand. Furthermore, having sufficient vacancies among units with below fair market rents (FMRs) is important to the successful use of vouchers.55 Table 21 shows how vacancy rates differ across regions and how they have changed since 2000. The fact that several of the recent changes are statistically significant suggests that the ACS sample size is sufficient to provide meaningful data on vacancy rates at this level of geographic aggregation.

Overall, the statewide vacancy rates of 10 percent and above and the significant increases since 2000 among units affordable to most income ranges imply that most rental markets in Pennsylvania are relatively loose and loosening further. But there is considerable variation in the vacancy rates among DCED regions. Region 5, which includes Pittsburgh, had the highest vacancy rates among DCED regions for each affordability range.

The Philadelphia region also had high and increasing vacancy rates in each affordability range, and it experienced the greatest increase in the vacancy rates for units affordable to ELI renters from 2000 to 2005-06. The city of Philadelphia had particularly high vacancy rates among units affordable to ELI, VLI, and LI households, and vacancy rates increased for each range between 2000 and 2005-06. The high vacancy rates in Philadelphia are not surprising, since Philadelphia has struggled with its vacant housing stock, both because of the quality of the stock and also because of the city’s declining population.56 Among units affordable to ELI renter households, the suburban counties had markedly lower vacancy rates than either Philadelphia city or the state average for Pennsylvania.

Low vacancy rates can reflect needs for additional affordable rental housing. Region 2, the Northeast, had the lowest vacancy rate for ELI-affordable units in 2005-06 and also the lowest rate overall. Furthermore, it was the only region where vacancy rates dropped significantly, both overall and in the ELI income range. Region 2 contains areas such as Monroe and Pike counties that have some of Pennsylvania’s greatest shortages of units affordable and available to ELI and VLI renters.

1 | 2 | 3

• 49See Appendix G for the detailed tables that support the commentary in this section.
• 50See Appendix F for discussion of changes between 1990 and 2000 by DCED region.
• 51See “Pennsylvania Housing Finance Agency Amended Allocation Plan for Year 2009,” p. 4.
• 52This study uses the term significant to refer to changes that are statistically significant.
• 53The city of Philadelphia and the county of Philadelphia constitute the same area.
• 54The one exception is Region 4, in which the decrease in units affordable to renters with incomes between 0-50 percent of AMI is not significant.
• 55The FMR for any market area, determined annually by HUD, is typically the 40th percentile rent for nonluxury units that were recently rented to a new tenant, although adjustments to this level are sometimes made. See HUD User for additional information. The FMR helps determine the subsidy that a household using a voucher receives. Specifically, the household pays 30 percent of its income in gross rent and HUD provides a rental subsidy to the household for the difference between the tenant payment and the FMR or gross rent, whichever is less. If this subsidy is not sufficient to cover the full rent of the unit, the household may pay the additional amount out of its own pocket, in which case the household will pay more than 30 percent of its income in rent, incurring a cost burden. Vacancies among below FMR units are important for voucher success because in order to use a newly issued voucher, a potential user must search in the private market to find a unit that passes HUD’s housing quality standards within 120 days of receipt (or must already live in such a unit). The lower the vacancy rate among below-FMR units, the harder it will be for the household to find an acceptable affordable unit when it is not already living in a unit that meets these standards. In this case, some households may not find a unit in the permitted time and will lose their vouchers, while others may rent a unit whose rent is above the FMR and will have at least some cost burden. See the Glossary for the definitions of FMR, rental subsidy, and voucher. In addition, Appendix D provides county-level FMRs and discusses the implications of their affordability to different income ranges.
• 56See Appendix A, Table 7 for population changes.

• Last updated: Friday, June 11, 2010