Vol.:(0123456789)

Journal of Housing and the Built Environment

https://doi.org/10.1007/s10901-020-09796-8

1 3

ARTICLE

Aordable housing program tenants andtheir access

topublic transportation andemployment

HanJohnPark

· KwangyulChoi

2,3

Received: 23 August 2019 / Accepted: 3 November 2020

Abstract

This research examines the level of access to public transportation and job resources spe-

ciﬁc to tenants residing in rental assistance program units in Harris County, TX, USA. It

evaluates the share of the subsidized housing renters from public transit facilities using pre-

set distances and the availability of jobs from the renters’ homes using the job accessibility

index. Particularly, this research compares the residential outcomes of the two major rental

assistance programs—Housing Choice Voucher (HCV) as a tenant-based program and

Low-Income Housing Tax Credit (LIHTC) as a project-based program—and the degree of

access to public transportation facilities—light rail transit (LRT) stations and transit cent-

ers. Furthermore, it assesses spatial distributions of the assisted housing program units and

jobs that are available tolow-income earners. The research ﬁnds that the HCV program

tenants reside in neighborhoods with greater accessibility to light rail stations and tran-

sit centers based on the network distance using the street routes and connections. It also

ﬁnds that the inner-city HCV residents have a higher level of low-paying job accessibility

than their LIHTC counterparts, whereas the suburban areas oﬀer equal or less favorable job

opportunities to the HCV tenants than the LIHTC counterparts.

Keywords Accessibility· Employment· Housing choice voucher (HCV)· Light rail transit

(LRT)· Low-income housing tax credit (LIHTC)· Transit center

1 Introduction

Aﬀordable housing programs in the United States are intended to provide low-income

households with housing units that meet certain quality standards so that qualiﬁed low-

income households could live in better residential units or settle in more desirable neigh-

borhoods through ﬁnancial support. The US Department of Housing and Urban Develop-

ment (HUD) speciﬁes that the housing choice voucher (HCV) program is to assist very

* Han John Park

[email protected]

Kinder Institute forUrban Research, Rice University, Houston, USA

School ofArchitecture, Planning andLandscape, University ofCalgary, Calgary, Canada

Department ofGeography andEnvironmental Sustainability, University ofOklahoma, Norman,

USA

H.J.Park, K.Choi

1 3

low-income families, the elderly, and the disabled to be able to aﬀord rental units in the

private market and that their goal is to make it possible for the program’s tenants to live

in decent, safe, and sanitary housing (HUD 2020a). The Low-Income Housing Tax Credit

(LIHTC) program issues tax credits to acquire, rehabilitate, or newly construct rental hous-

ing units that can be rented by lower-income households (HUD 2020c).

Good quality housing is an essential element for everyone’s life, regardless of their sta-

tus. Low-income populations need to have a stable income source to meet their daily needs

and enable them to have the opportunity to climb the social ladder. In this regard, it is

important to provide a quality transportation option that helps these populations to be able

to commute or travel without having to worry about owning vehicles for all their family

members. Although many commuters rely on personal vehicles for commuting,

some fam-

ily members in each household still rely on public transportation, and other choice riders

could switch their travel mode to public transportation when transit agencies continue to

enhance public transit frequency and quality. This research aims to assess the degree to

which aﬀordable housing programs are geographically connected to public transit facilities

and conveniently located in the vicinity of low-paying jobs.

In this study, we measure the accessibility to public transportation facilities for the two

major aﬀordable housing programs’ residents. In addition, we assess how much the aﬀord-

able housing programs’ tenants have job opportunities by analyzing how many low-paying

jobs are accessible based on their residences and the distance to such jobs. We examine

whether there is a diﬀerence in the accessibility to public transit and employment between

the two major housing programs: one is a tenant-based housing program (i.e., HCV), and

the other is a project-based housing program (i.e., LIHTC). This study aims to make sev-

eral practical contributions by providing policymakers, housing analysts, transportation

agencies, and urban planners with a better understanding of mobility challenges, particu-

larly for the job accessibility that aﬀordable housing tenants face. Harris County, the third-

most populous county in the UnitedStatesand the home of the city of Houston, is used as a

case study in representing low-density southern US counties. This paper is further divided

into the following sections: literature review, data and methodology, study area, analysis

results of accessibility, and discussion.

2 Literature review

Housing and transportation are the two most substantial categories of expenditure for

American households, which account for almost half of the average household budget (The

US Department of Labor 2018). The ﬁnancial burden of these two indispensable elements

makes up a signiﬁcant portion of expenditure for low-income families. To lessen the bur-

den of housing expenditure, particularly rental costs, many rental assistance programs have

been created for low-income populations who otherwise would have struggled with high

living costs. In the US, HCV and LIHTC are the two primary aﬀordable housing programs

Every eight-in-ten residents in Harris County drive their own vehicles to workplaces according to the

2018 5-year American Community Survey. Our research area, or Harris County, Texas has less dense devel-

opment patterns in the county, which is unfavorable to serving the county’s residents with good public tran-

sit due to dispersed origins and destinations. Thus, focusing the county’s two major transit characteristics

(light rail stations and transit centers) and work trips could make it straightforward to understand how con-

veniently aﬀordable housing residents commute to work by public transportation.

Aﬀordable housing program tenants andtheir access topublic…

1 3

that subsidize low-income tenants and make rent less burdensome than that of regular mar-

ket units. As a tenant-based subsidy program, HCVs subsidize the diﬀerence between 30%

of recipients’ household income and the contract rent. The maximum available rent for

HCV tenants is determined by an area’s reasonable market-rate residential units and also

varies by the recipient’s household size. HCV holders can choose any type of rental hous-

ing as voucher subsidies are assigned to individual households, not speciﬁc properties. In

theory, voucher recipients could move to less segregated and less poverty-stricken areas.

However, in practice, the HCV program, similarly to other housing mobility programs, has

failed to move voucher tenants into better neighborhoods (Kleit etal. 2016).

Compared to the HCV program, the LIHTC program is a project-based subsidy pro-

gram, which helps private developers provide aﬀordable housing properties to low-income

families. Private LIHTC developers construct multi-family complexes using the allocated

tax credits. The locations of LIHTC multi-family properties are determined by LIHTC

project applicants, considering each state housing ﬁnance agency’s LIHTC qualiﬁed allo-

cation plan (QAP). Each state housing agency approves successful LIHTC proposals that

went through competitive processes and allocates tax credits to each selected project. Many

LIHTC properties have been constructed in areas where developers were able to take extra

tax credits by considering state housing agencies’ rules on additional scores, such as quali-

ﬁed census tracts (QCTs), and therefore many of the existing LIHTC units are located in

poor neighborhoods.

In general, neighborhoods with new LIHTC properties already have

substantial populations who are below the poverty line (Williamson etal. 2009).

The neighborhood quality for aﬀordable housing program tenants has been extensively

studied, and the ﬁndings indicate that it is diﬃcult to conclude that tenants living in HCV

and LIHTC units beneﬁt from the best quality in their neighborhoods. Wang etal. (2017)

have found that voucher holders do not move to high-opportunity neighborhoods. Regard-

ing the poverty level, only approximately 20% of the voucher households live in low-pov-

erty neighborhoods in the 50 largest metropolitan areas (McClure etal. 2015), and it takes

some time for voucher tenants to relocate to lower-poverty neighborhoods after receiving

the subsidy (Eriksen and Ross 2013). Basolo (2013) has shown that voucher movers live

in neighborhoods with lower poverty rates than their previous neighborhoods, although

the improvements were minimal and, in addition, the movers’ employment signiﬁcantly

dropped after moving. In terms of safety in neighborhoods, subsidized housing tenants

tend to live in less safe neighborhoods than non-subsidized populations (Lens 2013; Ellen

et al. 2012). However, voucher holders live in neighborhoods in which crime rates are

considerably lower than traditional public housing and LIHTC tenants (Lens etal. 2011).

Regarding school quality, voucher movers live in neighborhoods with better school quality

than their previous neighborhoods (Basolo 2013), but both HCV and LIHTC tenants have

lower school quality in their residency area than other renters (Deng 2007). More speciﬁc

comparisons among subsidy programs by Horn etal. (2014) show that voucher holders live

in neighborhoods with lower-performing schools than those living in LIHTC developments

and poor households overall but with higher-performing schools than those living in public

housing developments. Lastly, concerning walkable neighborhoods, federally subsidized

housing is located predominantly in lesswalkable neighborhoods with poor access to ser-

vices and amenities (Talen and Koschinsky 2014).

The LIHTC locations are partly the consequence of state housing agency goals for community revitaliza-

tion through the LIHTC program rather than for LIHTC developments to be located in quality neighbor-

hoods (Deng 2007).

H.J.Park, K.Choi

1 3

Transit accessibility is an essential value that HCV and LIHTC tenants could beneﬁt if

their family members can easily access and use transit in their neighborhoods or the vicin-

ity of their residences, as their car ownership is limited. However, the speciﬁc research

of HCV and LIHTC tenants’ accessibility to individual transit facilities, such as light rail

stations and transit centers, is limited. The tenants of both programs primarily reside in

neighborhoods where rental units are not much more expensive than the median market

rate, where multi-family homes or denser housing units are predominantly available. A cer-

tain level of high-density supports the eﬃciency of public transit, where land use supports

multi-family buildings and attached housing units since dense land uses could improve

people’s mobility while conserving land (Guerra and Cervero 2011). Though the transit

service is not always successful in high-density areas in the American context (Bertaud

and Richardson 2004), a signiﬁcantnumber of low-income families live in high-density

neighborhoods where multi-family units are prevalent, and public transportation often con-

nects residential areas that are at a high density (Glaeser etal. 2008). Unlike those who

can aﬀord to operate private vehicles for every household member to commute, some

low-income families may have to be entirely dependent on public transit.

Blumenberg

and Ong (2001) have illustrated that many welfare participants rely on public transporta-

tion as their only transportation option to commute to work. This group in the population

would have signiﬁcant diﬃculties in life without the accessibility to public transit services,

which could lead to a low employment rate and continuous poverty status (Sanchez 1999;

Sanchez etal. 2004). However, there has been a disparity in the accessibility to public tran-

sit among housing assistance programs. Welch (2013) has explained that HCV households

have more equitable access to Baltimore’s transit stops than their LIHTC counterparts.

Employment is necessary for low-income families to sustain their daily lives. Allard and

Danziger (2002) have shown that the proximity to employment opportunities, especially

low-skill employment opportunities, is associated with welfare and housing assistance

recipients being able to have a higher probability of working. Additionally, Immergluck

(1998) has found that job opportunities near one’s area of residence have a modest eﬀect

on employment rates, although he has pointed out that the required levels of jobs should ﬁt

the skills of the labor force. Additionally, residences in close proximity to both transporta-

tion and job opportunities likely contribute to making it easy for people to ﬁnd and main-

tain jobs, and low-earning workers are more sensitive to transit than high earning workers

(Andersson etal. 2018; Ellen and Weselcouch 2015). Palm etal. (2014)have supported

that households trade-oﬀ housing and commuting costs and that households living in areas

close to jobs and ﬁxed-route transit service pay higher housing costs, which indicates that

neighborhoods oﬀering jobs and transit accessibility are more expensive and that people

are willing to pay more housing costs in such communities.

The locations in which assisted housing tenants have settled are still questionable in

terms of whether they could increase their life quality by gaining access to potential work-

places and main public transportation facilities without diﬃculties. To our knowledge, little

research has been carried out to evaluate how much rental housing assistance program ten-

ants collectively beneﬁt from existing public transportation facilities and jobs, which could

According to the 2018 5-year ACS estimates in Harris County, a majority of families, including low-

income families still rely on automobile. However, we know low-income adults are more likely to use tran-

sit than higher-income adults, and our focus renterhouseholds are primarily extremely low-income or very

low-income families. Therefore, the renterhouseholds in this study are more likely to rely on public trans-

portation than general households and even more so than moderate-income renterhouseholds.

Aﬀordable housing program tenants andtheir access topublic…

1 3

aﬀect their commuting and personal life patterns. Houston etal. (2013) haveassessed the

built environment for the voucher units and LIHTC properties that are located in Orange

County, California, and found that the HCV program is more likely than the LIHTC pro-

gram to be located in high-traﬃc areas which are associated with more transit service. As

most low-income families rely on inadequate income and their mobility is restricted due to

a low rate of private vehicle ownership, well-served public transportation options to their

work are necessary rather than optional. Low-income families subsidized by housing pro-

grams in and near neighborhoods that provide access to public transit facilities and oﬀer a

great number of employment opportunities could decrease transportation expenses. There-

fore, a family’s potential economic conditions could be enhanced (Pendall etal. 2014).

Hence, this research aims to assess the dynamics among aﬀordable housing tenants,

publictransit, and employment in Harris County, Texas, which includes the city of Hou-

ston, as an example of one of the Sunbelt cities. Houston’s case can apply to many low-

density cities in most southern US areas, where the population is growing fast, and urban

area is expanding to suburbs at a fast pace as well. Most commuters are drivers, and public

transit facilities are not dense in such low-density southern cities. Our Houston case analy-

sis could be an exemplary study in helping to understand the degree to which low-income

subsidized tenants use public transit and access low-paying jobs.

It is critical to compare the diﬀerence between two major assistance housing programs

and show how much the tenants of each housing program have accessibility to public tran-

sit facilities and job opportunities since they each have diﬀerent characteristics. While the

voucher program has oﬀered mobility to individual tenants using the existing private rental

market, the tax credit program has added new aﬀordable housing units by developers who

otherwise might not have voluntarily been involved in buildingaﬀordable housing (Deng

2007). We assume that many aﬀordable housing renters would be able to use transit ser-

vices and access employment more easily if public transportation facilities and employ-

ment opportunities exist in the vicinity of their subsidized homes than those who do not

have such accessibility. Better proximity to jobs connected to convenient public transporta-

tion facilities from home would enable subsidized housing tenants to reduce both monetary

and non-monetary transportation costs. Often, subsidized renters are single parents with

children, and the need for convenient transportation, including public transit, is signiﬁcant

as these families have responsibility for all aspects of household maintenance in addition to

job-related trips.

Therefore, the proximity to transit facilities and convenience in traveling

by transit must be of value to them. Reliable transportation helps to increase the level of

job opportunities, income, and job stability, and for those with limited automobiles, prox-

imity to public transit is essential (Blumenberg and Pierce 2012). In addition, improved

access to transit is positively related to maintaining jobs, and residing in transit-rich areas

is economically beneﬁcial (Blumenberg and Pierce 2014).

In the case of Harris County, 86% of the heads of the HCV families are female, 55% of the HCV fami-

lies are female-headed families with children, and the average household size is 2.59, which is close to the

county’s average household size of renter-occupied units, or 2.66.

H.J.Park, K.Choi

1 3

3 Data andmethodology

The HCV datasets were acquired from the Houston Housing Authority (HHA) in January

2016 and were already aggregated at the census block group level for personal information

conﬁdentiality. The LIHTC dataset was available on the HUD website, and it shows spe-

ciﬁc individual LIHTC project addresses that were in place by 2015. In Harris County, 316

projects existed, of which 300 projects were able to be accurately located using recorded

geographical information. The 16 other recordsdid not include coordinates as these were

not placed in service yet, although the credits of the project were already allocated.

Public transportation datasets were mainly available on the Houston–Galveston Area

Council (HGAC) and Houston METRO websites. We analyzed the street network to derive

the accessibility to public transportation facilities, which was conducted by using an Arc-

GIS analysis tool that assesses real street routes. This analysis was used to measure the

distance from each census block group centroid to the closest metro rail station and transit

center. The street ﬁle was available from 2017 topologically integrated geographic encod-

ing and referencing (TIGER) line shapeﬁles on the Census Bureau website. Although the

network analysis does not directly mean the level of real public transportation use of the

aﬀordable housing program tenants, it measures the true distance to the closest facility,

which could reﬂect the shortest time that the tenants have to spend getting to use public

transportation in reality.

We computed the numbers and shares of HCV and LIHTC tenants within preset dis-

tances to both the closest metro rail station and the closest transit center. The centroids

of the census block groups were used to assess the number of tenants for each program.

Distances were more speciﬁcally categorized within three miles with an accumulated half a

mile (e.g., a half-mile,

one mile, one mile and a half, two miles, two and a half miles, and

three miles), and distance thresholds were more widely but intuitively set outside of the

three-mile range (e.g., ﬁve miles, 10 miles, 20 miles, and 40 miles).

Employment information was available from the US Census longitudinal employer-

household dynamic (LEHD) origin–destination employment statistics (LODES) datasets.

We used the 2015 employment datasets, including origins of employers, or residential char-

acteristics, and destinations of employers, or work characteristics, with three types of pre-

categorized income levels: jobs with earnings of $1250 per month or less; $1251 to $3333

per month; and $3333 per month or greater. We chose the lowest level of earnings ($1250

per month or less) for both origins and destinations of employers. For job accessibility,

network distances between a pair of centroids for all census block groups in Harris County

were calculated, based on the street connection, in a similar way to the public transporta-

tion accessibility analysis. To assess the accessibility for both renter groups’ residences,

we used HUD’s formula, which computes the accessibility that is assigned to each census

block group (HUD 2020b). More speciﬁcally, HUD’s original job proximity index uses a

gravity model, where the job accessibility of each residential block group is a collective

We used the centroids of census block groups as locations of the residences of subsidized housing tenants

instead of individual addresses, mainly because the information of the HCV program locations was aggre-

gated at the census block group level for the conﬁdentiality of individual voucher families.

Although thresholds that are smaller than a half-mile are empirically important, the thresholds smaller

than a half-mile did not capture a reasonable number of residents from the closest transit facilities because

the average distance from each census block group’s centroid to its boundary was, in general, larger than a

half mile except for a few census block groups in or near Downtown Houston.

Aﬀordable housing program tenants andtheir access topublic…

1 3

job index considering the distance to all job locations with a positive weight of the size of

employment, or job opportunities, and all other residential locations with a negative weight

of the size of labor supply, or competition. A shorter distance to a job block group impacts

a higher job opportunity, which positively aﬀects the index, whereas a shorter distance to

a residential block group impacts a higher labor competition, which inversely aﬀects the

index. In our study, having known the limitation of HUD’s distance that is measured with

the air distance, we computed the street-based distance instead, which general people actu-

ally experience in real life. Additionally, we further developed HUD’s concept to ﬁt with

accessibility for low-income aﬀordable housing tenants by accounting for employees in the

lowest income category (origins) and jobs in the lowest wage category (destinations).

We followed the speciﬁcations of the proximity indexes of HUD (2020b) but used the

network distances instead of air distances to assess meaningful accessibility than simple

proximity. We also utilized the low-paying jobs and low-income residents in order to have

a closer understanding of the subsidized renters’ access to employment opportunities.

The index i stands for a residential census block group, and j indicates all n census

block groups within Harris County. Distance, D, is measured based on the network dis-

tance between census block groups i and j, with distances less than1 mile substituted for 1.

Employment, E, represents the number of jobs paying less than $1250 per month in census

block group j. Residents, R, is the number of low-income workers making less than $1250

per month in census block group j.

4 Housing assistance programs, public transportation, andjobs

intheresearch area, Harris County, Texas

4.1 Aordable housing programs inHarris County

More than half of the census block groups (1103 out of the county’s total 2144 block

groups) have accommodated either HCV or LIHTC tenants or both. A total of 16,246

households with the HCV program are identiﬁed in 1072 census block groups, which is

50% of the census block groups in Harris County, with an average of 15.2 households per

census block group.

A total of 37,171 households supported by the LIHTC program are

found in 176 census block groups, with an average of 211.2 households per census block

Job Accessibility

∑

j=1

(

∕D

i,j

)

∑

j=1

(

∕D

i,j

)

We used census block groups as the unit of analysis of our study, as census block groups are the most

granular level of geography to show our analysis at the neighborhood level given that we acquired the HCV

residents at the census block group level. Census block groups also represent neighborhoods at a ﬁne level

to compute distances between the aﬀordable housing units and the job locations that were available origi-

nally at the census block level.

In addition to the job accessibility index with the raw value unit, we also used the percentile unit that

ranges from 0 to 100, meaning that the higher the job accessibility index, the better the access to low-

paying job opportunities for low-income residents in a neighborhood (HUD 2020b). We included the results

from the percentile-based job accessibility index in footnote 17 but did not speciﬁcally explain the results in

the main text in detail, for brevity.

The voucher data was acquired from the Houston Housing Authority in January 2016.

H.J.Park, K.Choi

1 3

group.

As shown in Figs.1 and 2, demonstrating the distribution of the tenants with the

two programs, the HCV tenants are more dispersed than the LIHTC residents. Tenants

with a voucher subsidy can choose a private rental unit without any geographical contain-

ment as long as the selected rent charges are below the Fair Market Rent (FMR). On the

other hand, LIHTC tenants have to reside in units in multi-family complexes where private

developers have already designated and completed construction. Therefore, LIHTC resi-

dents are found in fewer census block groups with a larger average number of units in each

hosting census block group than the HCV case, as the HCV tenants are identiﬁed in many

census block groups with a fewer average number of voucher units in each accommodating

census block group.

Figure1 demonstrates the HCV distribution by census block group. Many HCV house-

holds are clustered in neighborhoods of the county’s southern area and near the George

Bush Intercontinental Airport. The maximum number of HCV holders in a census block

group is 298, and half of the census block groups did not host any single HCV family.

Fig. 1 HCV households in Harris County’s census block groups

The Low-Income Housing Tax Credit unit information was acquired from HUD National LIHTC Data-

base, and we used the LIHTC units that were placed by 2015.

Voucher renters can reside in LIHTC units, and the statistics of such voucher families are not known

because of the unavailability of the detailed addresses of the HCV voucher families.

HUD’s picture of subsidized households shows that, inHarris County as of 2015, 95% of the county’s

HCV holders are very low-income families and that household income per year is $13,607. About 34% of

HCV households reported wages are the major source of income. Approximately 18% of either a household

head or their spouse is 62years or older. Half of the HCV households are single-parent families without a

spouse but with one or more children under 18years, and 23% of the populations in the HCV units have a

Aﬀordable housing program tenants andtheir access topublic…

1 3

Locational diﬀerences in distribution between the two programs are signiﬁcant. Accord-

ing to Fig. 2, the distribution of the LIHTC renters is more clustered in certain census

block groups. The maximum number of LIHTC tenants in a census block group is 940,

which is almost three times larger than that of the HCV tenants. The diﬀerence between

the two programs is primarily caused by the highest number of LIHTC renters residing in

a few census block groups where LIHTC apartment complexes have been constructed. The

diﬀerence in the two housing programs reveals that LIHTC residents have less locational

ﬂexibility in their residence than HCV renters. Far more than half of the county’s census

block groups do not have any LIHTC residents.

Furthermore, we divided the county into three sub-areas by using major concentric

highways that separate the county into three segments. The ﬁrst sub-area is an urban area

within Interstate 610 (I-610) with limited land which is more expensive and supports more

employment opportunities than the other two sub-areas. The second sub-area is a donut-

shaped area between I-610 and Beltway 8, where urban and suburban areas coexist, as

some census block groups are within the city of Houston and others are not. The third sub-

area is beyond Beltway 8, where most census block groups are outside the city of Houston

and located in the suburbs.

Fig. 2 LIHTC households in Harris County’s census block groups

disability. On average, the current voucher families had waited for 26months on the waiting list and have

been on the active subsidy recipient list for 92months.

Footnote 12 (continued)

H.J.Park, K.Choi

1 3

The HCV tenants are relatively evenly distributed as voucher holders ﬁnd locations

where rental units are available and wherevoucher subsidies are accepted. Despite some

clustering census block groups, the maximum number of HCV renters does not exceed

300 units in any census block group. HCV renters choose subsidized units without a strong

geographic limitation, though there is the rent cap, and HCV recipients should ﬁnd own-

ers that accept HCV subsidies as part of renters’ income source. On the other hand, the

LIHTC units are spotted in fewer census block groups throughout the county than the case

of the HCV units. The largest populous LIHTC census block group surpassed 900 units,

and LIHTC units are observed more frequently in the second and third sub-areas than the

ﬁrst sub-area. In the inner I-610 sub-area, it is challenging for LIHTC private developers to

construct multi-family buildings where the land in the urban core is pricier than the other

two outer sub-areas. LIHTC developers, instead, looked out the program’s development

sites where competition for multi-family development is less intense and the land is less

expensive, which are mostly located in the second and third sub-areas, in addition to tax

credit incentive areas taken into consideration (e.g., QCTs and diﬃcult development areas,

or DDAs, albeit less frequently).

4.2 Public transportation facilities inHarris County

As illustrated in Fig.3, the Houston area has three metro rail lines: the Red, Green, and

Purple Lines. The Red Line links Downtown Houston, Houston’s largest job center, and

the Texas Medical Center, the second-largest job center. The Green Line links the east

Fig. 3 Light rail stations in Harris County