The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
w o r k i n g
p
a
p
e
r
14 29R
Do Tenant- and Place-Based Rental
Housing Programs Complement Each
Other? Evidence from Ohio
Brett Barkley, Amy Higgins, and
Francisca G.-C. Richter
FEDERAL RESERVE BANK OF CLEVELAND
Working papers of the Federal Reserve Bank of Cleveland are preliminary materials circulated to
stimulate discussion and critical comment on research in progress. They may not have been subject to the
formal editorial review accorded official Federal Reserve Bank of Cleveland publications. The views stated
herein are those of the authors and are not necessarily those of the Federal Reserve Bank of Cleveland or of
the Board of Governors of the Federal Reserve System.
Working papers are available on the Cleveland Fed’s website at:
www.clevelandfed.org/research.
Working Paper 14-29R
May 2015*
Do Tenant- and Place-Based Rental Housing Programs
Complement Each Other? Evidence from Ohio
Brett Barkley, Amy Higgins, and Francisca G.-C. Richter
We characterize Housing Choice Voucher (HCV) use in Low-Income Housing
Tax Credit (LIHTC) units and explore whether the subsidy overlap responds to
needs unmet by the HCV program alone. We present a subsidy allocation model
allowing for complementarity of tenant-based and place-based subsidies to guide
our analysis. Findings for Ohio in 2011 suggest vouchers tied to LIHTC units are
more likely allocated to very poor households with special needs. We also find
that HCV users who freely choose to redeem their voucher in a LIHTC unit make
up a somewhat larger share of LIHTC households in tighter or less affordable
markets. There is little evidence that using both programs in concert enables access to better neighborhoods: households across both programs live in neighborhoods that tend to have above 20% poverty rates, with HCVLIHTC users actually
living in higher poverty neighborhoods in the most urban Ohio counties when
compared to the HCV population as a whole.
Keywords: Affordable Housing, Housing Choice Vouchers, Low-Income Housing
Tax Credit, Housing Policy, Neighborhood Quality.
JEL Codes: R38, H71, R23.
Suggested citation: Barkley, Brett, Amy Higgins, and Francisca G.-C. Richter,
2015. “Do Tenant- and Place-Based Rental Housing Programs Complement Each
Other? Evidence from Ohio,” Federal Reserve Bank of Cleveland, working paper
no. 14-29R.
Brett Barkley and Amy Higgins are at the Federal Reserve Bank of Cleveland,
and Francisca G.-C. Richter is at Case Western Reserve University (francisca.
richter@case.edu). The authors thank Dr. Holly A. Holtzen and the staff of the
Ohio Housing Finance Agency for providing de-identified tenant-level data
and for helpful comments. The paper also benefited from insights of Stephanie
Moulton, Dionissi Aliprantis, and Gregory D. Johnson, CEO of the Cincinnati
Housing Authority.
*First version was November 2014.
1
Introduction
The longstanding debate over cost-effectiveness in the provision of rental housing subsidies is influenced by whether the housing goods offered by tenant-based and place-based
programs are considered substitute or complement products. Assuming that both programs
provide equal access to housing units and environments of similar quality, researchers
show that tenant-based programs are less costly than their place-based counterparts (Olsen
(2009); GAO (2002)). Thus, under the substitutability assumption, it is possible to house
a larger number of eligible households when allocating funds away from place-based and
into tenant-based programs. Alternatively, when place-based programs supply housing in
a tight rental market, deliver human or educational services, or support neighborhood revitalization efforts, they may reach subpopulations whose needs would be unmet under
a tenant-based program alone. Thus, some argue that the programs should be best seen
as complementary in the provision of housing for low-income households. These programs can also exhibit complementarity at the individual level. For instance, an elderly,
extremely low income individual may need the human services provided through a placebased program, and the deep rental subsidy provided by a housing voucher. Under the
complementarity assumption, the goal from a social planner’s perspective would be to optimally allocate resources across both programs in order to cover the heterogeneous needs
of eligible families.
While, to the extent of our knowledge, there is not an explicit coordination of programs at the federal level, the number of households using vouchers in units produced
by place-based programs is not insignificant. By 2011, 34% of Housing Choice Vouchers (HCVs) were being used to subsidize the rent of Low-Income Housing Tax Credit
(LIHTC) units in Ohio. Based on a sample from ten states, O’Regan and Horn (2013)
report that between 33.7% and 49.6% of low-income qualifying LIHTC tenants received
additional rental assistance – HCV or otherwise. Such overlap is not necessarily a recent
phenomenon; voucher use in LIHTC started to become more common in the 1990s as a
way to ensure a steady stream of tenants and alleviate bankruptcy problems of LIHTC
developments. By 1994, GAO (1997) and Buron et al. (2000) reported that 37%-39% of
2
LIHTC tenants received additional project-based or tenant-based rental assistance.
As a first step in understanding the extent to which these programs complement each
other, we lay out a simple allocation model that allows for complementarity of tenantand place-based subsidies, and we consider potential solutions under alternative optimization criteria. The model guides our subsequent analysis characterizing the subset of users
who access subsidies from both types of programs relative to single subsidy users. While
it is commonly understood that the overlap between programs evolved primarily to protect tax credit investors and developers, we find evidence in Ohio that tenants also benefit
likely due in part to local-level planning across agencies in which vouchers required to be
redeemed in a LIHTC unit, on average, are allocated to house the most vulnerable households – relative to LIHTC tenants without vouchers and the HCV population in general.
At the same time, it is not clear that voucher households in LIHTC see improvements in
neighborhood quality compared to the overall HCV population.
Previous economic and policy research focusing on LIHTC users has documented the
incidence of using a voucher in LIHTC but has not looked into whether this subsidy overlap provides housing services unmet by either program alone (Galvez (2002)). For that,
a comparison of HCV users in and out of LIHTC would be needed. Lacking the data
required for such approach, we resort to a comparison between the entire HCV population in Ohio and its subset of HCV holders living in LIHTC. The time period analyzed
spans 2011 and 2012. We use subsidized tenant level data in Ohio for 2011, provided
by the Ohio Housing Finance Agency (OHFA) and HUD’s Public Use Microdata Sample
(PUMS) for the year 2012. Tract level data for HCV users is from HUD’s A Picture of
Subsidized Households, 2011, and neighborhood quality variables are from the Census
Bureau1 , 2010.
In Section 2, we discuss relevant literature on the incidence of voucher use in LIHTC
units. Section 3 develops a simple housing subsidy allocation model with program complementarity, and Section 4 presents the comparative analysis of subsidy users. The last
1 2010
data in 2000 tract boundaries by Geolytics, Inc.
3
section provides some concluding thoughts.
2
The Incidence of Voucher Use in LIHTC Units
Of the main affordable housing programs in the US, the Low Income Housing Tax Credit
Program is the second largest and the primary source of subsidies for the construction of
rent restricted housing units (Table 2). LIHTC is designed so one of the following conditions apply: at least 40% of its units are affordable to households with incomes at 60%
of area median income (AMI), or at least 20% of units are affordable to households with
incomes at 50% of AMI2 . However, many states have implemented competitive selection preferences for units that have income restrictions as low as 30% of AMI (NCSHA
(2014)). For example, 7% of LIHTC units placed in service in Ohio during 2012 were
affordable for households with incomes less than 30% AMI whereas 63% were allocated
at the 60% AMI level and 30% at the 50% AMI level.3
The Housing Choice Voucher program is the main source of tenant-based subsidies.
As conceived, the program relies on the market supply of rental units at the Fair Market
Rent (FMR)4 and requires the voucher holder to pay 30% of their income on housing. The
remaining portion of the rent is covered by the voucher, conditional on the unit and tenant
meeting certain requirements specified by HUD. By design, housing should be affordable
and accessible to all HCV users, provided the market supplies enough units satisfying
HUD requirements to meet demand. Local public housing authorities (PHAs) must allocate at least 75% of their vouchers to extremely low-income tenants (households at or
below 30% of AMI) with the remainder of vouchers eligible for very low-income tenants
(households at or below 50% AMI).5 It is also important to note that PHAs can alter basic
targeting requirements (e.g., based on family size, age, disability status) to reduce wait
2 In
this context, a rental is affordable if it amounts to no more than 30% of household income.
the US, 10% of all LIHTC units placed in service during 2012 were allocated at the 30% AMI level.
Louisiana allocated the largest portion (48%) of LIHTC units at this threshold.
4 The Fair Market Rent is set at the 40th percentile of gross rent for typical, non-substandard units in a
given local housing market.
5 HUD defines low-income households at or below 80% AMI
3 In
4
times for such applicants and place them in suitable housing. This can extend to coordination with other housing programs, as well. For example, the Columbus Metropolitan
Housing Authority (CMHA), in their 5-year plan, commit to continued partnerships with
other housing and service providers to serve homeless, family, senior, and special needs
populations–which it says are typically served through the use of vouchers in connection
with LIHTC projects CMHA (2013).
HCVs may be used in a LIHTC unit via a place-based voucher (PBV) or a tenant-based
voucher (TBV). While PBV holders can only use their subsidy in the unit attached to the
voucher, TBV users are unrestricted, but choose to redeem their voucher in a LIHTC unit.
The allocation of PBVs from a local PHA to a specific project is regulated by HUD, but
planned locally. PHAs can use up to 20% of their previously allocated HCV vouchers for
this purpose by referring interested HCV applicants to property owners that have projectbased voucher units available. These are units for which a housing assistance payments
(HAP) contract between the property owner and PHA has already been approved. If these
qualifying units also receive additional federal subsidies, they will have also been subject
to a subsidy layering review before the establishment of the HAP contract (Federal Register 2014).6 Importantly, HUD expedites the HAP and subsidy layering review process for
LIHTC projects in order to encourage the use of PBVs in LIHTC units Roberts (2008). It
is also worth noting that LIHTC landlords cannot refuse to rent to voucher holders solely
because of their Section 8 status, which in practicality translates into additional units added
to the supply of HUD certified rentals.7 In effect, these factors serve to expand the market
of units available to HCV holders.
6 In addition to LIHTC, subsidies may include Rural Housing Services (RHS) Section 515, Federal Hous-
ing Administration (FHA) insurance, tax-exempt bond financing, FHA risk-sharing, Project-Based Section
8, McKinney-Vento Homeless Assistance, Historic Rehab, Community Development Block Grant (CDBG),
HOME investment partnership program, HOPE VI, Native-American Housing Assistance (NAHASDA),
and the Neighborhood Stabilization Program (NSP). Any or all of these funds may overlap. In Ohio, according to HUD’s National LIHTC Database (1987-2012), the most common federal subsidy programs to be
used in concert with LIHTC are tax-exempt bond financing (8.96%), RHS Section 515 (11.3%), and HOME
(17.56%)
7 Internal Revenue Code §42(h)(6)(B)(iv)
5
Given program design, most LIHTC units will not be affordable to households that
qualify for HCVs. Thus, O’Regan and Horn (2013) argue that the HCV program has a
significant role in allowing households with extremely low income to access LIHTC developments. Their data consists of 480,000 LIHTC units from fifteen state housing finance
agencies (HFAs) for 2009 and 2010, representing approximately 30% of the national LIHTC housing stock. The analysis finds that states with higher rates of HCV use in LIHTC
also have higher rates of extremely low-income households served by LIHTC, suggesting more focused resources for the most disadvantaged populations. They also point out
that LIHTC households without additional rental assistance experience higher levels of
instability (i.e. occupancy turnover) than LIHTC households using a HCV.
Williamson et al. (2009) analyze 2004 tenant level data from Florida to determine the
extent of the HCV-LIHTC overlap in general. They are able to identify HCV holders
in LIHTC units but cannot discriminate between PBVs and TBVs. The study finds that
a significant portion of HCV-LIHTC users (30%) are housed in LIHTC units located in
Qualified Census Tracts (QCTs), usually low-income neighborhoods where developers
receive 1.3 times the tax credit that would be allocated outside of QCTs. In a subsequent
study, Williamson (2011) finds that LIHTC tenants also receiving HCV assistance are
much less likely to be cost burdened than other LIHTC tenants. Her analysis shows that
65% of HCV-LIHTC users do not experience a housing cost burden compared to only 9%
of LIHTC households without a voucher.
The aforementioned studies characterize the HCV-LIHTC population in relation to
other LIHTC tenants. However, such analyses are unable to inform on whether the subsidy
overlap responds to needs unmet by the HCV program alone. The relevant counterfactual
analysis for such test would imply comparing the housing conditions of HCV households
within a locality, with and without the availability of LIHTC rentals. As Williamson et al.
(2009) suggest, it is possible that the use of HCV holders in LIHTC units is a response to
the scarcity of HUD-certified affordable housing units in the private rental market. It may
also signal household preferences for newer, higher quality units than what are typically
available to HCV users – all LIHTC units have been built since 1987 Galvez (2002). HCV
6
users may also choose LIHTC units in search for better neighborhoods or the provision of
special services within the living environment.
Focusing on the potential issue that HCV use in LIHTC may respond to the undersupply of certified FMR units in the local market, Eriksen (2009) suggests that LIHTC
developments may actually crowd out unsubsidized rental housing that would have otherwise been built. But the extent of such an effect is not entirely clear. Baum-Snow and
Marion (2009) find that LIHTC projects crowd out new rental construction in areas where
property values are rising but do not have the same effect in stable or declining areas. One
of the largest experiments conducted by HUD in 1973 provides valuable insight on private rental market responses to an expansion of tenant-based housing subsidies (Frieden
(1980)). The Experimental Housing Allowance Program (EHAP) consisted of three components: a demand-side experiment (how families would make use of their allowances),
a supply-side experiment (local housing market responses to increased demand), and an
administrative experiment (to test effectiveness of different administrative arrangements).
Despite the fear that discriminatory practices would affect take-up rates, the experiment
showed that minority and ‘’hard to house” families were able to find housing of acceptable quality. Female headed-households and families on welfare assistance were the main
participants, but participation by the elderly was low. Still, only about half of families
eligible for allowances participated and the program had small effects on mobility and
quality of housing. Partly due to the low take-up rate, there was virtually no increase in
rents following the open enrollment policy.
While the use of HCV in LIHTC units might be a response to tight low-income rental
markets, desirability of better quality, newer units, or need for special services, it could
also represent an opportunity for HCV households to access better neighborhoods in close
proximity to their social networks. Neighborhood environment influences can have a profound impact on the lives of individuals. The program design of both HCV and LIHTC
consider the importance of neighborhood quality. HCVs were originally designed to allow for movement out of public housing units by providing neighborhood choice to poor
households. As for LIHTC, many states preference their credits to projects that are part of
7
a broader community revitalization plan in an effort to improve neighborhoods.
Unfortunately, most studies of neighborhood quality access by housing subsidy users
consistently reach disappointing conclusions. Deng (2007) studies HCV and LIHTC programs in six Metropolitan Statistical Areas (MSAs), San Jose, New York, Boston, Cleveland, Atlanta, and Miami. She finds that both HCV and LIHTC units were predominantly
located in areas with low school quality. On the other hand, Newman and Schnare (1997)
find slightly better prospects for HCV than LIHTC households. Compared to public housing, LIHTC units provide poor households access to better quality neighborhoods; but
when compared to HCVs, LIHTC unit dwellers experience lower neighborhood quality.
Lens et al. (2011) investigate whether HCVs assist tenants in moving to neighborhoods
with less crime. Their analysis makes use of two datasets, 2000 census tract level crime
data for 91 cities and census tract data from 1998-2008 for Austin, Chicago, Cleveland,
Denver, Indianapolis, Philadelphia, and Seattle. The authors find that voucher households
occupied lower crime neighborhoods compared to tenants occupying LIHTC and public
housing units. Residents living in LIHTC and public housing units experienced crime rates
comparable to poor renters in general. Richter et al. (2013) consider neighborhood quality
of HCV and LIHTC households in 2000 and the 2006-2009 period for Cuyahoga, Franklin,
and Hamilton counties in Ohio, as well as Allegheny County in Pennsylvania (quality is
a function of poverty, education, and employment indicators). They find that Allegheny
County neighborhoods with LIHTC projects placed in service by 2000 improved in quality
relative to other poor neighborhoods without LIHTC projects. And the level of quality experienced by LIHTC users in 2006-2009 was comparable to that of HCV users. However,
these results did not hold for any of the Ohio counties studied.
3
An Allocation Model of Housing Subsidies
We present a simple model of housing subsidy allocation that comports to the complementarity assumption of housing programs. Assume there is a fixed population of low-income
8
households H and that subsidies to rent and construction are available via tenant- and
place-based programs, respectively. The tenant-based or voucher program subsidizes the
rent of certified market units with a subsidy inversely related to the income of the renter.
The place-based program subsidizes the construction of project units designed under specific guidelines with the condition that rent be restricted to a certain level in order to be
affordable to a targeted low-income population. The housing subsidy cost is denoted by
Cv for vouchers and C p for a project unit, and we assume that Cv < C p .
Let households in H be characterized by two binary variables. Variable y classifies
the population into poor (y = 0) and very poor (y = 1) such that poor–yet not very poor–
households can marginally afford to rent in the open market and construction subsidized
units with around 30% of their income8 . Variable z captures non-income factors that classify them as “hard to house,”9 , considerably reducing the value of a certified market unit
relative to a planned project subsidized unit. These factors may include the household’s
own characteristics and those of the housing market she faces. In this stylized model, the
threshold that defines z is such that it takes the value of 1 if the household requires a project
unit to be housed, either because it provides an additional unit to a tight market or because
of the additional housing services it offers.
Thus, households are classified into four types T (y, z) as follows: T1 = T (0, 0), T2 =
T (1, 0), T3 = T (0, 1), T4 = T (1, 1), with the last category encompassing the hard to house,
very poor households. Given this classification and denoting the number of households in
each type by n j , we can specify H = {(i, j)| i = 1, . . . , n j , and j = 1, . . . , 4}, where (i, j)
8 Say
a poor household has income at 60% AMI or below and a very poor household has income at 50%
AMI or below. For a family of four, respectively, this would equate to approximately $991 and $782 per
month in the Cleveland MSA. The 40th percentile or fair market rent for a 2 bedroom unit in the Cleveland
MSA is $727 per month.
9 Cunningham et al. (2005) refer to the “hard to house” as high-need households, such as grandparents
caring for grandchildren, persons with disabilities, very large households, and families coping with an array of difficult problems and barriers to housing access. Here, we extend this concept to include market
characteristics that, interacted with household conditions, narrow choices in the low-income rental market.
9
represents the ith household of the jth type. We define subsidy binary variables as
v = 1{voucher assigned},
p = 1{project unit assigned},
so (v(i, j), p(i, j)) denotes a subsidy assignment to household (i, j) by the social planner,
according to some optimization criteria. Subsidy assignments take values (1, 1), (1, 0),
(0, 1), or (0, 0) if both subsidies, voucher only, project unit only, or no subsidies are accessible to (i, j), respectively.
The housing outcome function h(v(i, j), p(i, j)) defined in Table 1 takes the value of 1
if household (i, j) can be housed given the subsidy allocation (v(i, j), p(i, j)) and 0 otherwise.
Given Table 1, we can express the housing outcome function for household (i, j) with
the help of the logical and operator, ∧, as:
h(v(i, j), p(i, j)) =
1,
v(i, j),
p(i, j),
v(i, j) ∧ p(i, j),
if j = 1
if j = 2
(1)
if j = 3
if j = 4
So type T1 households can be housed without a subsidy and type T2 households can
be housed if assigned a voucher but not a project unit since, by definition, they could not
afford it. Type T3 households are hard to house in the local rental market but can afford
and be housed with a project unit. Finally, type T4 households are hard to house and very
poor so they can only be housed in a project unit and with the added rental subsidy of a
voucher. We assign weights w j reflecting the social planner’s value of housing type T j
relative to other household types.
10
Table 1: Tabulation of the housing outcome
function h(v(i, j), p(i, j))
Type ( j) y
1
0
2
1
3
0
4
1
z
0
v
1
1
0
0
0 1
1
0
0
1 1
1
0
0
1 1
1
0
0
p
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
11
h(v(., j), p(., j))
1
1
1
1
1
1
0
0
1
0
1
0
1
0
0
0
The total budget allocation to housing programs is B and we assume that B is not large
enough to entirely house any of the three types in need of a subsidy.
B ≤ min{n2Cv , n3C p , n4 (Cv +C p )}
(2)
We now proceed to consider alternative optimization goals. Suppose the objective is to
allocate subsidies (v(i, j), p(i, j)) among households in H − T1 in order to house the largest
number of households. Then, w j = 1/3 for all j = 2, 3, 4, and we have:
4
maximize
qv ,q p
nj
∑ ∑ w j h(v(i, j), p(i, j))
j=2 i=1
subject to qvCv + q pC p ≤ B
4
qv =
nj
∑ ∑ v(i, j)
(3)
j=2 i=1
4 nj
qp =
∑ ∑ p(i, j)
j=2 i=1
Given that Cv < C p and the restriction imposed in 2, the problem stated in 3 is solved when
all resources are devoted to house bB/Cv c households of type T2 with vouchers, the least
costly alternative, as illustrated in Figure 1, point A10 . With this setup, it is best to invest
all resources in the type for which the lower cost subsidy delivers the needed housing.
This implies taking all resources away from the hard to house groups, which is clearly an
undesirable outcome.
If on the other hand, the goal is to house the most disadvantaged (those with both,
economic and non-economic barriers to housing), then w4 = 1 and w j = 0 for all j 6= 4.
In this case, bB/(Cv +C p )c households in T4 will be housed (see point D in Figure 1). In
the figure, F is a solution satisfying this condition, such that subsidies are provided for q2
households of type T2 , q3 households of type T3 , and q4 households of type T4 . Alternatively, the objective could require benefiting households from all types. Each allocation is
10 The
floor function bxc gives the largest integer that is smaller than x.
12
efficient as long as T2 households use a voucher and not a construction-subsidized unit.
qp
bB/(Cv +C p )c
D
F
q3 of T3
A
q4 of T4
qv
bB/Cv c
q2 of T2
Figure 1: Housing subsidy allocations resulting from various optimization setups
Now, let’s step away from the idealized scenario in which the social planner can identify types and assign subsidies accordingly. Much like in actuality, both subsides are available, with income restrictions applying for voucher qualification and rent restrictions for
construction subsidies. In this case, T2 households cannot be prevented from redeeming
their vouchers in construction-subsidized units. Likewise, T1 households may also choose
to rent in a construction-subsidized unit, which, according to the above objective, would
constitute an inefficient allocation of resources. In reality, however, it is difficult to devise
a strict classification of households into types. What we consider T1 households may be
on the margin of becoming T2 households if their income fluctuates and barely surpasses
the affordability threshold. T1 and T2 households may also be close to falling into the T3
and T4 groups, respectively, if the supply of affordable market units decreases in quantity
or quality.
13
In Section 4 we characterize users of LIHTC by type of subsidy: PBV, TBV, and
LIHTC-only. Insofar as housing administrators allocate the larger subsidies to the hardest
to house, poorer households, we would expect to see a concentration of T4 households in
PBV units and T3 households in LIHTC-only units. At the same time, T2 households may
choose to live in PBV or TBV units, while T1 households may be found in LIHTC-only
units. In tight housing markets, assuming that landlords in the open market would be less
likely to rent to HCV holders, we would expect to see an increased share of households
choosing to redeem their vouchers in LIHTC units. Some of these households would be
classified as T4 ’s even when in less tight markets they could fall into the T2 group.
4
Characterization of the HCV and LIHTC Populations
in Ohio
HCV is the largest subsidized housing program in Ohio, with 33% of all subsidized households falling under this program in 2011. LIHTC comprises the third largest component
of subsidized housing in Ohio, at 23% in 2011 (Table 2). While the number of participants
across both programs adds to about 154,000, the number of unique households covered is
less than 124,000 due to the fact that about 30,000 households use a voucher in a LIHTC
unit. In Ohio, during 2011, 34% of HCVs were being used in a LIHTC unit, with 23%
of them being PBVs and 11% TBVs (Table 2). Relative to all LIHTC households, PBV
holders represented 34% of that pool, while 15% of LIHTC households were TBV users.
51% of LIHTC users were not subsidized under the HCV program.
In order to include local housing market conditions in the analysis, we calculate an
index of tightness or unaffordability in the low-income housing market. Counties in Ohio
are ranked according to the share of renter households who pay more than 30% of their
income in rent11 . This is similar to a reverse-ranking based on a regional affordability
measure(Richter et al., 2013). The larger the share, the ’tighter’ the low income market.
In tighter markets we would expect TBV use to be more common. To give a sense of
11 Income
and rent data are from the American Community Survey 2009-2013
14
Table 2: Subsidized Housing in Ohio and
the US, 2011
Ohio
US
LIHTC
63,992 1,832,564
PBV
21,883 –
TBV
9,788 –
HCV
91,313 2,100,000
Public
43,546 1,100,000
Section 8 PB 72,376 1,200,000
the geographic distribution of subsidized housing in Ohio, Figure 2 shows the ratio of
program users to poor population by county and by tightness in the low-income rental
market. Perhaps not surprisingly, the ratio of HCV users to poor is larger in urban counties,
whereas LIHTC is slightly more evenly distributed across both urban and rural areas. In
most counties, the HCV population is much larger than the LIHTC population. However,
it is interesting to observe that in the tightest markets the added shares of subsidies tend
to be larger, which could suggest a greater need for low income and certified units in such
markets.
We now compare PBV, TBV, LIHTC-only, and all HCV users (includes PBV, TBV, and
HCV only) by the characteristics of individuals they house. HCV data is from PUMS, for
the year 2012, while PBV and TBV data is from OHFA for the year 2011.12 . According
to the model, we would expect that the PBV and to some extent, the TBV populations
encompass the hard to house, very poor group of eligible households. Our comparisons
are limited to those characteristics collected in both data sets. Table 3 shows a higher
incidence of elderly (62 and older) among LIHTC users as compared to the entire HCV
population. However, non-elderly head of households with a disability are clearly more
represented in the HCV program than in any of the LIHTC categories. Comparing TBV
12 The PUMS for subsidized households is not available in 2011.
We do not attempt to calculate the relative
size of the overlap at the county level since the data for voucher and LIHTC use are collected at different
points in time, which coupled with high levels of mobility of subsidy users makes counts inconsistent at
lower aggregate levels.
15
holders with the entire HCV population, it is possible to say that the elderly are more likely
to use their voucher in a LIHTC unit, while the reverse is true for the non-elderly, disabled
population of HCV holders.
Table 3: Share of Household Heads by Assistance Type 2011-2012
62 or Older
Persons w/ disabilities, under 62
Black
White
Hispanic
HCV
LIHTC-Only
13.44
31.10
61.41
36.99
1.53
29.69
3.01
41.59
54.5
41.59
PBV
TBV
28.03 22.56
14.12 7.92
54.60 61.37
44.34 37.16
54.6 61.37
The income distribution of subsidized households across programs shows that TBV
holders are very similar to the overall population of HCV users with a median income of
about $10,000 (Figure 3). On the other hand, PBV holders tend to be on the lower end
of the distribution, with a median income under $8,000 and about 10% of its population
receiving no income at all. Other characteristics of the PBV population suggest this group
may lack family support. Over fifty percent of PBV users have a household size of 1,
compared to a larger household size, on average, for all HCV users. Accordingly, PBV
users also tend to live in much smaller units–nearly fifty percent are one or zero bedroom
apartments. As expected, LIHTC dwellers with no additional housing subsidies are less
needy. In Ohio, the median income for this group was about $20,000 in 2011.
We also look at the distribution of households by race under each subsidy combination
(Table 3). Issues such as discrimination in the private rental market could lead to a higher
share of African-American households using their HCVs in LIHTC units when compared
to the HCV holder population as a whole. However, the data only tells us that the shares of
African-American head of households in the HCV population and the population of TBV
holders is comparable, at about 60%.
16
4.1
Characterizing voucher use within LIHTC
While we are limited in our comparison of HCV users in and out of LIHTC units, data
does allow for a richer comparison of HCV users within LIHTC. Based on the stylized
model described in Section 3, we would expect to see a concentration of T4 households
in PBV and TBV units, within the constraints given by the number of HCVs available.
Given that only about a quarter of all income-eligible households receive rental housing
assistance, we expect there will be T4 households in LIHTC-only units as well. According
to our model, T3 households would occupy LIHTC-only units, and the low-income rental
market would accommodate the needs of T2 voucher users and T1 households. However,
T2 households may decide to use their vouchers in LIHTC units and T1 households may
also freely rent in a LIHTC-only unit. Given the higher degree of coordination involved
in allocating PBVs, we expect to see a higher proportion of T4 ’s among PBVs than among
TBVs. In addition, more TBVs are likely to be present in tight markets compared to
markets where affordable rental housing is more accessible.
We approximate the model classification of household types with the data of LIHTC
tenants which includes income, age, household size, number of children, and indicators
for disability and having previously been homeless. Tenants are classified by income into
‘poor’ and ‘very poor’ if their incomes, adjusted for household size, are above and below
50% AMI in their county of residence, respectively. They are classified as ‘hard to house’
if they meet any of the following criteria: are 62 year of age and older; are disabled;
were previously homeless; are in the top tercile relative to the number of children in the
household. Table 4 provides a cross tabulation of households by type and subsidy use
in LIHTC. We see that Type 1’s constitute only 5% of the population and most of these
households (92%) are not PBV or TBV users. Type 3 households, those that are hard
to house but are not very poor, make up 13% of the population and most of this group
(93%) is not housed with a voucher. Technically, Type 1 and Type 3 households should
not receive vouchers according to program rules, as their household income is greater than
50% of AMI. The fact that some show up as voucher users in the data could be due to a
change in income status since receiving the voucher. Type 2’s and Type 4’s constitute the
17
other 30% and 52% of the population, respectively and their distribution in LIHTC is quite
similar, with a little under half of them being in units with a PBV attached.
Table 4: Cross Tabulation of Households by Type and
Subsidy use in LIHTC, Ohio 2011. Top number is
row percentage, bottom number is column percentage.
Typology
LIHTC-only
PBV
TBV
Total
Type 1
92.40
10.51
3.56
0.49
4.04
1.33
100
5.24
Type 2
36.43
23.41
45.17
35.12
18.40
34.38
100
29.61
Type 3
92.70
26.71
3.84
1.34
3.46
2.90
100
13.28
Type 4
34.97
39.37
46.28
63.05
18.75
61.38
100
51.87
Total
46.07
100
38.08 15.85 100
100
100
100
Table 5 reports households paying greater than 30% (rent burdened) and 50% (severe
rent burden) of their monthly household income in rent. The voucher program by definition only requires households to pay 30% of their adjusted gross income in rent, but a
voucher user still may be rent burdened if the cost of the selected unit exceeds the payment standard set by HUD, generally the 40th percentile market rent. Focusing on Type 2
and Type 4 households, we observe that over 80% of LIHTC-only subsidy users in each
sub-sample are either rent burdened or severely rent burdened. While a lower portion of
PBV and TBV users are rent burdened, still as many as 40% of Type 2 PBV users and
over 50% of Type 2 TBV users experience either rent burden or severe rent burden. Type
4 PBV users experience among the lowest rates of rent burden. This could be due to more
30% AMI units being allocated for PBV use; 43% of 6,133 extremely low-income units
are occupied by PBV users while only 10% are occupied by TBV users. For sake of com18
Table 5: Percent of income paid to rent inclusive of utilities by Type and Subsidy use in
LIHTC, Ohio 2011. Tenants who did not report income are excluded. Top number is count
of households in each category, bottom number is percentage.
Type 1
LIHTC-only
≤ 30%
1,847
67.9
> 30% 871
32.02
> 50% 2
0.07
Total
2,720
100
Type 3
LIHTC-only
≤ 30%
1,967
65.05
> 30% 1,056
34.92
> 50% 1
0.03
Total
3,024
100
All Types
LIHTC-only
≤ 30%
6,380
29.93
> 30% 10,870
51
> 50% 4,063
19.06
Total
21,313
100
PBV
TBV
Total
Type 2
LIHTC-only
50
90.91
5
9.09
102
91.07
10
8.93
55
100
112
100
1,999
69.24
886
30.69
2
0.07
2,887
100
939
15.64
3,434
57.2
1,631
27.17
6,004
100
3,446
57.96
1,347
22.66
1,152
19.38
5,945
100
1,410
49.51
1,045
36.69
393
13.8
2,848
100
5,795
39.16
5,826
39.37
3,176
21.46
14,797
100
PBV
TBV
Total
9,176
81.83
1,157
10.32
880
7.85
11,213
100
2,930
56.38
1,699
32.69
568
10.93
5,197
100
13,733
52.87
8,365
32.2
3,877
14.93
25,975
100
PBV
TBV
Total
Type 4
LIHTC-only
145
96.03
6
3.97
121
93.8
8
6.2
151
100
129
100
2,233
67.58
1,070
32.38
1
0.03
3,304
100
1,627
17.01
5,509
57.6
2,429
25.39
9,565
100
PBV
TBV
Total
12,817
73.81
2,515
14.48
2,032
11.7
17,364
100
4,563
55.07
2,762
33.33
961
11.6
8,286
100
23,760
50.59
16,147
34.38
7,056
15.02
46,963
100
19
PBV
TBV
Total
parison, McClure (2005) found that 38% of all HCV users were rent burdened in 2002.
Similarly, as mentioned in Section 2, Williamson (2011) found that approximately 35% of
HCV users in LIHTC units were rent burdened.
In order to further characterize the set of households in Ohio who use HCV within
LIHTC, we specify a multinomial logit model for the probability that LIHTC households
belong to one of the three categories of users in question: PBV holders, TBV holders and
other LIHTC users.
P(yi = j) =
exp(xi0 j β j )
0 β ) + exp(x0 β )
1 + exp(xi2
2
i3 3
, j = 1, 2, 3
(4)
The multinomial logit is derived from the underlying assumption that to each LIHTC
household corresponds a random variable taking one of three possible user-category values, and these random variables are independent from each other. That is, each individual
i has probabilities P(yi = 1), P(yi = 2), P(yi = 3) of belonging to one of the three user
categories: PBV, TBV, LIHTC-only, respectively. The three categories are mutually exclusive and exhaustive. The log odds ratio for category j relative to the baseline category
(LIHTC-only) is set to be a linear function of households characteristics (xi j vector), and
this relation leads to equation413 .
We provide two specifications. Model 1 estimates user category probabilities given
the following household characteristics: age, race, gender of household head, number of
children, homelessness status, disability status, and an income from employment dummy.
Finally, a dummy for living in a county above the 80th percentile of the low income housing tightness index is also included. Results are presented in the form of marginal effects.
For each covariate and category combination, the marginal effect estimates the change in
probability of belonging to the specific category due to a unit change in the covariate.
The second model estimates user category probabilities given previously derived indi13 Note
that the purpose of this model is to describe user categorization based on household covariates.
We are not estimating a conditional logit model of consumer choice.
20
Table 6: Marginal Effects of Select Characteristics on
User Type Probabilities. Standard errors in parentheses.
PBV
TBV
Model 1
Elderly
-0.118*** -0.050***
(0.005)
(0.005)
Female household head
0.058*** 0.066***
(0.005)
(0.004)
Household with children
0.021*** 0.075***
(0.005)
(0.005)
Head or spouse w/ disability
0.144*** -0.025***
(0.007)
(0.005)
No income from employment 0.314*** 0.065***
(0.005)
(0.004)
Homeless
0.099*** -0.004
(0.012)
(0.010)
Tighter market
0.034*** 0.034***
(0.005)
(0.004)
Model 2
Very poor
Hard to house
Tighter market
Tighter market
at ‘very poor’=1
Hard to house
at ‘very poor’=1
21
0.421**
(0.003)
0.010**
(0.004)
-0.005
(0.004)
0.150**
(0.003)
0.001
(0.003)
0.037**
(0.004)
-0.001
(0.005)
0.011**
(0.005)
0.050**
(0.004)
0.003
(0.004)
cator variables for ‘very poor’ and ‘hard-to-house’ that define household types, as well as
the indicator for tighter rental market. The model includes interactions among these three
variables as these interactions define the household types. Note that, unlike with linear
models, marginal effects in nonlinear models are non-constant even in the absence of interaction terms(Karaca-Mandic et al., 2012). However, our inclusion of interaction terms
in the model allows for a more flexible form of marginal effects.
Table 6 presents the average marginal effects (AME) estimates derived from the multinomial logit of equation 1, with LIHTC-only as the base outcome. The AME is the average
change in probability of being a PBV or TBV user -given LIHTC residence- due to a one
unit increase in the explanatory variable. Estimates from Model 1 confirm that PBV holders are clearly a more vulnerable population relative to other LIHTC dwellers. All else
equal, having no income from employment increases the probability of being a PBV user
by 31 percentage points, and having a disability increases this probability by 14 percentage
points. The extent of the latter effect is limited in that less than 20% of LIHTC projects
in Ohio provide supportive services for the elderly and less than 10% for persons with
disabilities (Table 7). If supportive services in LIHTC were expanded, we might see this
effect increase.
Being elderly decreases the likelihood that the LIHTC user is a PBV or TBV holder.
However, a few caveats should be kept in mind. The marginal effects calculation reflects
that the largest share of elderly LIHTC users are LIHTC-only users (in line with what we
would consider Type 3s and possibly Type 1s). However, relative to the entire HCV population, the share of elderly tenants is much higher among PBVs and TBVs (Table 3). It is
also worth noting that there is a higher share of elderly and persons with disabilities in the
most recently built LIHTC projects; this is especially true among PBV users (Figure 4 and
5). Distinct from the provision of supportive services, the fact that newer buildings house
a higher share of elderly and persons with disabilities seems to suggest a growing trend to
match the hardest to house HCV applicants with units most suitable to their needs. Since
2007, OHFA has required all new construction units to meet certain Universal Design elements, including a no-step entrance into the unit, a width of 32 inches for all doors, and a
22
Table 7: LIHTC Projects Providing Supportive Services. Top row is count,
bottom row is percentage.
Ohio
US
One or more services
Yes
477
27.93
No
1,231
72.00
Total 1,708
100
16,308
43.4
21,271
56.60
37,579
100
Services for the elderly
Yes
313
18.33
No
1,395
82.00
Total 1,708
100
5,055
13.45
32,524
86.55
37,579
100
Ohio
Services for families
52
3.00
1,656
96.96
1,708
100
15
0.88
No
1,693
99.12
Total 1,708
100
10,250
27.28
27,329
72.72
37,579
100
Services for persons with disabilities
57
3.00
1,651
96.66
1,708
100
Services for the homeless
Yes
US
1,000
2.66
36,579
97.34
37,579
100
23
3,028
8.06%
34,551
91.94
37,579
100
bathroom on the main floor with clear floor space of 30 inches by 48 inches OHFA (2006).
It is also observed that the probability of voucher use in LIHTC units increases by
about three and one-half percent if tenants are located in tight housing markets. This
result is only indicative of housing market conditions in Ohio in 2011. Focusing on Model
2, we see that when the market becomes tighter, ‘very poor’ households have 5% higher
chances of being TBV holders and lower chances of being LIHTC-only users. If a ‘very
poor’ household becomes ‘hard to house,’ their probability of being in LIHTC as a PBV
holder slightly increases.
Altogether, data suggests a pattern of allocation somewhat consistent with our stylized
model presented in Section 3. PHAs appear most likely to refer households for PBV
placement in LIHTC so as to house those with the lowest incomes or with disabilities.
And in tighter markets, the use of vouchers in LIHTC seems to respond to a lower supply
of units that meet FMR standards in the private rental market. Access to HCV tenantlevel data would allow a more robust analysis and provide an opportunity to confirm these
results.
4.2
Neighborhood Quality for HCV users in LIHTC units relative to
all HCV users
Previous work comparing neighborhood quality under place- and tenant-based programs
has not identified neighborhood quality for the group of households subsidized under both
programs. There is no way of uncovering the complex decision making process by which
individuals decide where to locate with a voucher. However, we can see whether those
selecting into a LIHTC unit – via a TBV or PBV – experience better neighborhood quality
as compared to the entire population of HCV users. We know that the HCV population
is made of individuals living in and out of LIHTC units and that those in LIHTC units
represented about 34% of the HCV population in Ohio, 2011.
HCVpop = T BVpop + PBVpop + HCV nonLIHTC pop .
24
Unable to identify HCV users not living in LIHTC units (HCV nonLIHTC pop ), we compare average neighborhood quality for the voucher population in LIHTC (T BVpop +
PBVpop ) with that of the HCV population as a whole (HCVpop ).
Figure 6 shows the average neighborhood quality for voucher users in select Ohio
counties versus that for the subset of TBV and PBV holders, jointly. Figure 7 compares
neighborhood quality for TBV and PBV holders. Neighborhood quality is measured at
the census tract level and it is relative to all tracts in the United States in 2010. We use
quantiles of the first principal component estimated from the following variables: percent
poverty, percent in the labor force, percent with a high school diploma, percent with a
bachelor’s degree, and percent employed.14 Thus, the highest quality neighborhood in
the US would have an index of 100 and the median quality neighborhood according to
our measure would have an index of 50. The average neighborhood quality is calculated
for all tracts within a county, weighted by the number of HCV, TBV or PBV holders
per tract, accordingly. Counts of HCV holders within each tract are from the Picture of
Subsidized Households, 2011 and counts for TBV and PBV holders are from 2011 LIHTC
data provided by OHFA.15 Similarly, average poverty rates experienced by HCV, TBV and
PBV users at the county level are presented in Figures 8 and 9. The size of the bubble in
these graphs reflects the relative share of TBV and PBV users in the HCV populations.
The larger the bubble, the more TBV and PBV holders there are in the county relative to
the total number of HCV holders.
By focusing on the bottom left quadrant of Figure 6, we see that on average, HCV
holders in Ohio counties live in neighborhoods of below median quality; and for those
voucher holders living in a LIHTC unit (TBV+PBV holders), neighborhood quality tends
14 The
first component explains 67% of the variation in the 2010 data.
these two data sets are collected at different points in time, and low-income populations are
highly mobile, we cannot expect alignment of counts at the census tract level. However, the data does give
us a snapshot of the distribution of voucher holders across neighborhoods and thus, we can compare neighborhood quality for all three groups (HCV, PBV, TBV) aggregated to the county level. Counties included
in our study had a population greater than 50,000 to primarily capture urban areas, per US Census Bureau
definition, and greater than 50 PBV & 50 TBV holders to ensure low voucher populations aren’t given undue
weight.
15 Because
25
to remain below the median. Figure 7 shows that in some counties PBV holders experience higher average neighborhood quality than TBV holders, yet in these counties, both
groups remain below the median of quality (bottom left quadrant). There are, however, a
few counties in which subsidized households experience above the median neighborhood
quality (right side quadrants). In these counties, it is most likely that TBV holders are
located in better neighborhoods than their PBV counterparts.
To complement the rank-based analysis of neighborhood quality we provide graphs using actual neighborhood poverty rate averages experienced by the subsidized populations
within counties. Here, we divide the plots into four quadrants to mark average poverty
rates below and above 0.2 or 20%. It is worth noting that the average neighborhood poverty
rate for the HCV population is above 10% for all counties studied and HCV-LIHTC households do not benefit from living in less poor neighborhoods relative to the overall HCV
population (Figure 8). Figure 9 shows that compared to PBV users, TBV users – those
who chose to live in LIHTC unit – tend to live in less poor neighborhoods, but only in a
handful of counties the average neighborhood poverty for TBV users is below 20%.
5
Conclusions
Evidence from Ohio suggests that HCV and LIHTC programs exhibit some degree of
complementarity, particularly, when serving very poor households and those considered
hard to house. The overlap of programs is not at all insignificant: In Ohio, during 2011,
34% of LIHTC households received place-based voucher assistance and another 15% of
LIHTC dwellers were using their tenant-based voucher to live in a tax-credit unit. We
develop a subsidy allocation model to guide our analysis of voucher use in LIHTC. Though
coordination between the two rental housing programs is limited at the state and federal
levels, more intentional planning among housing administrators seems to occur at the local
level in order to allocate vouchers for use in LIHTC to the most vulnerable households.
This is important in that the deep subsidy of the HCV provides access to LIHTC units
that otherwise might not be affordable and which could convey certain benefits – e.g.
26
supportive services, newer housing that incorporates universal design principles, etc. –
compared to the rest of the affordable rental housing stock. We also find that the population
of TBV users, who freely choose to live in LIHTC units, is somewhat larger in tighter
rental markets, which we interpret as a response to a lower supply of private market rental
units that meet FMR standards. Otherwise, TBV users are similar to the overall HCV
population in terms of incomes and ethnicity, but older on average.
While our analysis of HCV users in LIHTC points to benefits of the overlap for certain
populations under certain market conditions, there is very little evidence to support that
PBV or TBV usage enables access to better quality neighborhoods relative to other HCV
users. In the Ohio counties studied, the average HCV user and the average voucher holder
living in LIHTC units, occupy neighborhoods which are below the 50% median neighborhood quality index, with poverty rates of above 20%. There are only a few Ohio counties –
Clermont (Cincinnati MSA) and Delaware (Columbus MSA) – where PBV or TBV users,
on average, reside in better neighborhoods than the HCV population as a whole, and at
the same time, experience above median neighborhood quality. Unfortunately, households
across both programs live in neighborhoods that tend to have above 20% poverty rates,
with PBV and TBV users actually living in higher poverty neighborhoods in the most urban counties (e.g., Cuyahoga, Hamilton, Summit, and Franklin) when compared to the
HCV population as a whole.
Voucher use in LIHTC units can be perceived as beneficial insofar as the lowest income
and hardest to house population are provided access to quality rental units that would
otherwise not be available. However, it is important for housing administrators and policy
makers to address the low neighborhood quality that prevails for users of both programs.
A further exploration of the county-level variation in neighborhood quality and poverty
illustrated by our analysis may shed some light into how to better coordinate the provision
of rental housing subsidies to improve the living conditions and opportunities of lowincome families.
27
References
Baum-Snow, Nathaniel and Justin Marion, “The Effects of Low Income Housing Tax
Credit Developments on Neighborhoods,” Journal of Public Economics, 2009, 93 (5-6),
654–666.
Buron, Larry, Sandra Nolden, Kathleen Heintzi, and Julie Stewart, “Assessment of
the Economic and Social Characteristics of LIHTC Residents and Neighborhoods,” Abt
Associates, Cambridge, MA: 2000.
CMHA, “PHA 5-Year and Annual Plan,” Columbus, OH: 2013.
Cunningham, Mary K., Susan J. Popkin, and Martha R. Burt, “Public Housing Transformation and the ’Hard to House’,” Metropolitan Housing and Communities Center
Brief, June 2005, (9).
Deng, Lan, “Comparing the Effects of Housing Vouchers and Low-Income Housing Tax
Credits on Neighborhood Integration and School Quality,” Journal of Planning Education and Research, 2007, 27, 20–35.
Eriksen, Michael, “The Market Price of Low-Income Housing Tax Credits,” Journal of
Urban Economics, 2009, 66 (2), 141–149.
Frieden, Bernard J., “Housing allowances: an experiment that worked,” The Public Interest, 1980, 59, 15–35.
Galvez, Martha, “What Do We Know About Housing Choice Voucher Program Locations
Outcomes? A Review of Recent Literature,” Urban Institute. Washington D.C.: 2002.
GAO, “Tax Credits: Opportunities to improve oversight of the low-income housing program,” Report to the Chairman, Committee on Ways and Means; and the Chairman,
Subcommittee on Oversight, Committee on Ways and Means, House of Representatives
1997.
28
, “Federal Housing Assistance: Comparing the Characteristics and Costs of Federal Programs,” GAO Publication 02-76. Washington D.C.: U.S. Government Printing Office.
2002.
Karaca-Mandic, Pinar, Edward C. Norton, and Bryan Dowd, “Interaction Terms in
Nonlinear Models,” Health Services Research, 2012, 47 (1), 255–274.
Lens, Michael C., Ingrid Gould Ellen, and Katherine O’Regan, “Neighborhood Crime
Exposure among Housing Choice Voucher Households,” U.S. Department of Housing
and Urban Development Office of Policy Development and Research. Washington D.C.:
2011.
McClure, Kirk, “Rent Burden in the Housing Choice Voucher Program,” Cityscape, 2005,
8 (2).
NCSHA, “State HFA Factbook: 2012 NCSHA Annual Survey Results,” Washington D.C.:
2014.
Newman, Sandra J. and Ann B. Schnare, “The Failure of Housing Programs to Deliver
on Neighborhood Quality And a Suitable Living Environment,” Housing Policy Debate,
1997, 8 (4), 703–741.
OHFA, “2007 Housing Credit Qualified Allocation Plan,” Columbus, OH: 2006.
Olsen, Edgar, “The Cost-Effectiveness of Alternative Methods of Delivering Housing
Subsidies,” 2009. Presented at the Thirty-First Annual APPAM Research Conference,
Washington D.C.
O’Regan, Katherine M. and Keren M. Horn, “What Can We Learn About the LowIncome Housing Tax Credit Program by Looking at the Tenants?,” Housing Policy Debate, 2013, 23 (3), 597–613.
Richter, Francisca García-Cobián, Mark Snidernam, Matthew Klesta, and Frank
Manzo, “Low-Income-Rental-Housing Programs in the Fourth District,” Working Paper
13-11, Federal Reserve Bank of Cleveland, 2013.
29
Roberts, Buzz, “Beyond Section 42: How the New Housing Bill Affects LIHTC Projects,”
Journal of Tax Credit Housing, 2008, 1 (8).
Williamson, Anne R., “Can They Afford the Rent? Resident Cost Burden in Low Income
Housing Tax Credit Developments,” Urban Affairs Review, 2011, 47 (6), 775–799.
, Marc T. Smith, and Marta Strambi-Kramer, “Housing Choice Vouchers, the LowIncome Housing Tax Credit, and the Federal Poverty Deconcentration Goal,” Urban
Affairs Review, 2009, 45 (1), 119–132.
30
Low-income renters
paying >30% of income
in rent (%)
Lucas
> 44 to 58
Cuyahoga
> 58 to 62
> 62 to 66
Summit
> 66 to 70
Mahoning
> 70 to 78
LIHTC to poor population ( % )
<5
> 5 to 10
> 10 to 15
Franklin
> 15
Montgomery
HCV to poor population ( % )
<5
> 5 to 10
Hamilt on
> 10 to 15
> 15
Figure 2: Distribution of LIHTC and HCV tenants in Ohio as a percentage of the county
population in poverty. 2011 LIHTC data is from OHFA, HCV data is from A Picture of
Subsidized Housing, income and rent data from 2009-2013 American Community Survey.
31
Income Distribution of Rent−subsidized Tenants
0
.2
.4
.6
.8
1
Tenant level data for Ohio, 2011
0
10,000
20,000
30,000
annual income in dollars
PBV
TBV
LIHTC−only
HCV
40,000
50,000
Figure 3: Income Distribution of Rent-subsidized Tenants. 2011 LIHTC data is from the
Ohio Housing Finance Agency. 2012 interval income data of HCV users is from the Public
Use Microdata Sample.
32
Share of Elderly−headed Households in LIHTC, 2011
0
.2
.4
.6
by year in which the building was placed in service
1995
2000
2005
2010
date
PBV
TBV
LIHTC−only
Figure 4: Share of Elderly-headed Households in LIHTC by year placed in service. 2011
LIHTC data is from the Ohio Housing Finance Agency.
33
Share of Persons−with−disabilities−occupied Units in LIHTC, 2011
0
.1
.2
.3
by year in which the building was placed in service
1995
2000
2005
2010
date
PBV
TBV
LIHTC−only
Figure 5: Share of Persons-with-disabilities-occupied Units in LIHTC by year placed in
service. 2011 LIHTC data is from the Ohio Housing Finance Agency.
34
Average Neighborhood Quality for HCV and HCV−LIHTC Users
80
100
County averages − Ohio, 2011
60
Clermont
Fairfield
40
Miami
Hancock
Lorain
Greene
Athens
Portage
Butler
Jefferson Warren
Franklin
Clark
Montgomery
Trumbull
Hamilton
Allen
Cuyahoga
Lucas
Summit
Mahoning Stark
0
20
TBV+PBV users
Delaware
0
20
40
60
80
100
All HCV holders in county
Figure 6: Average Neighborhodd Quality for HCV (x-axis) and HCV-LIHTC Users (yaxis). Neighborhood quality measured as quantiles of first principal component of census
tract level variables from Census 2010. We include %poor, %employed, %in labor force,
%high school, %bachelors. Bubble size represents relative share of HCV use in LIHTC
units across counties. 2011 LIHTC data is from Ohio Housing Finance Agency. 2011
HCV data is from A Picture of Subsidized Housing.
35
Average Neighborhood Quality for TBV and PBV Users
80
100
County averages − Ohio, 2011
60
Delaware
Hancock
40
Portage
Athens
Fairfield
Miami
Greene
Jefferson
Clark
Trumbull
Lorain
Franklin
Warren
Allen Hamilton
Mahoning Cuyahoga
Montgomery
Lucas
Butler
Summit
Stark
0
20
PBV holders
Clermont
0
20
40
60
80
100
TBV holders
Figure 7: Average Neighborhodd Quality for TBV and PBV Users. Neighborhood quality measured as quantiles of first principal component of census tract level variables from
Census 2010. We include %poor, %employed, %in labor force, %high school, %bachelors. Bubble size represents relative share of HCV use in LIHTC units across counties.
2011 LIHTC data is from Ohio Housing Finance Agency.
36
Average Neighborhood Poverty Rate for HCV and HCV−LIHTC Users
.5
County averages − Ohio, 2011
Mahoning
.2
.3
Summit
Franklin Athens
Clark
Montgomery
Trumbull
Greene Jefferson
Lorain
Butler
Allen
Portage
Warren
Hancock
Fairfield
Miami
Clermont
Delaware
0
.1
TBV+PBV users
.4
Stark Lucas
Hamilton
Cuyahoga
0
.1
.2
.3
.4
.5
All HCV holders in county
Figure 8: Average Neighborhood Poverty Rate for HCV and HCV-LIHTC Users. Census
tract neighborhood poverty rates are from the Census 2010. Bubble size represents relative
share of HCV use in LIHTC units across counties. 2011 LIHTC data is from Ohio Housing
Finance Agency. 2011 HCV data is from A Picture of Subsidized Housing.
37
Average Neighborhood Poverty Rate for TBV and PBV Users
.5
County averages − Ohio, 2011
Lucas
.4
Hamilton
Cuyahoga
Summit
.3
Montgomery
Athens
Clark
Mahoning
Lorain
Jefferson
Trumbull
.2
Warren
Allen
Delaware
FairfieldHancock
Miami
Clermont
Portage
0
.1
PBV holders
Franklin
Butler
Greene
Stark
0
.1
.2
.3
.4
.5
TBV holders
Figure 9: Average Neighborhood Poverty Rate for TBV and PBV Users. Census tract
neighborhood poverty rates are from the Census 2010. Bubble size represents relative
share of HCV use in LIHTC units across counties. 2011 LIHTC data is from Ohio Housing
Finance Agency
38