Rooms for Improvement: A Qualitative Metasynthesis of the Housing Choice Voucher Program

Abstract This article synthesizes housing subsidy voucher research to explain why, when in theory vouchers enable users to move out of poor neighborhoods, in practice they often do not. This qualitative meta-analysis presents an examination of the assumptions of the program and their relationship to empirical findings. Two themes emerged from this synthesis: market barriers and product problems. Data from a variety of studies and contexts portray recipients struggling to use vouchers in the private rental market due to market barriers, including lack of public transportation and the presence of discrimination. Product problems constrained freedom of choice about where to move and when to make a housing transition. These constraints manifest as compromised housing quality and low voucher utilization. This synthetic view cannot account for all outcomes or exceptional cases, but results suggest where participant experiences are generalizable and attributable to features of the housing market and structure of the program itself.

The second shortcoming is that only a slim majority (Snell & Duncan, 2006, p. 727) of households approved for vouchers actually take advantage of them to relocate (Finkel & buron, 2001;Jacob, 2003;Sanbonmatsu et al., 2011;Wood, Turnham, & Mills, 2008). although demand for vouchers far exceeds their supply, as evidenced by long wait lists, about one third of recipients return them unused. For example, Finkel and buron (2001), basing their estimates on administrative rather than experimental data, reported that only 69% of the families offered vouchers subsequently moved and made use of their voucher. Researchers using experimental paradigms reported similar success rates of families who were provided vouchers and able to lease a housing unit (Jacob, 2003;Sanbonmatsu et al., 2011;Wood et al., 2008).
although no quantitative meta-analysis of the research exists, the research generally shows a pattern of low mobility and low uptake for qualifying families. yet it remains unclear why this pattern developed. Dozens of studies from a variety of disciplines have delved into housing voucher programs, many generating qualitative data on individual HCVs. This article draws upon the findings from studies that have highlighted how voucher user's location deliberations relate to take-up patterns. These studies have enabled researchers to describe these patterns but have left them less able to explain how or why participants locate where they do. Moreover, most researchers base their studies of take-up patterns on assumptions informed by theory rather than data. Considering extensive empirical findings enables a critical review of the theoretical foundations of how voucher use can contribute to poverty deconcentration.
in an effort to understand the take-up patterns among voucher holders, this article synthesizes the findings from a wide range of qualitative studies on the HCV Program and explores an assumption undergirding much of the research, specifically, that such programs give voucher users the choice to participate in the private markets. it asks the question: How do descriptions of housing voucher applicants' experiences help further explain the take up patterns described in the quantitative literature? To answer this question, this article first provides background on the HCV Program, next describes the conceptual framework and the qualitative method employed in the article, presents the findings in detail, discusses the findings, and concludes with considerations of some policy implications, hypotheses, and directions for future research.

Background
The HCV Program (formerly known as "Section 8") began in 1974 and quickly grew to become one of the dominant programs of subsidized housing in the United States (Galvez, 2010a). Operated by the U.S. Department of Housing and Urban Development (HUD) and administered by over 3,000 local public housing authorities (PHas), the HCV Program is now the federal government's largest housing assistance program, providing rental subsidies to over two million households across the country (Sard & Coven, 2006).
The HCV Program is structured to enable households to access housing in the private market. Participants, usually drawn from a waiting list, identify private market housing that meets the program's quality and affordability standards. The program provides a monthly subsidy to cover the difference between the cost of the housing and what the participant can afford to pay, up to a locally defined Fair Market Rent payment standard. HUD's Fair Market Rent is defined as the dollar amount below which 40% of the standard-quality rental housing units are rented in an area. Participants typically pay approximately 30% of their income for rent and utilities, with the program paying the balance (Green, 2011).
although not the program's original purpose (see apgar, 1990), the HCV Program is often portrayed as a response to the concentrated poverty often associated with bricks and mortar, low-income public housing projects (e.g., Galvez, 2010a). initially, policy makers framed vouchers as a means to enable "very low-income families, the elderly, and the disabled to afford decent, safe, and sanitary housing in the private market" (HUD, 2014). in recent years, an additional goal undergirding the program is to deconcentrate the poor by making it possible for voucher recipients to leave public housing projects and move to lower poverty neighborhoods (McClure, 2008).
according to an influential group of researchers and policy makers, housing and neighborhood quality are central components of social inequality. Poor families frequently live in low-income neighborhoods of concentrated poverty (Massey, 1996). These neighborhoods are believed to negatively affect individual and family well-being (Ludwig et al., 2012) and life outcomes (Sampson, 2012;Wilson, 1996). although the prospect that HCVs might enable recipient households to leave high-poverty neighborhoods has piqued the interest of researchers, housing voucher administrators have exhibited varied levels of commitment to this mission shift (Popkin, 2000;Katz & Turner, 2001).
Researchers have identified negative externalities that appear to be associated with high concentrations of voucher households. Galster, Tatian and Smith (1999), for example, found that the clustering of voucher users in disadvantaged and declining neighborhoods increases the likelihood that property values will be adversely affected. Such concerns added further momentum to the deconcentration goal, even though deconcentration has not been causally linked to individual economic outcomes for participants.
Housing subsidies are not an entitlement, and not all households that qualify receive housing subsidies. Only about one in four eligible households receives federal rental assistance of any type (Leopold, 2012;Sard & averez-Sanchez, 2011). as a consequence, most housing authorities maintain waiting lists for housing subsidies. nationally, applicants wait an average of 2 years to receive notice of an available voucher (Leopold, 2012).
Given the demand signaled by the waitlists, as Pashup, Edin, Duncan and burke (2005) have noted: "it may seem surprising that not all families that can move through these programs choose to do so" (p. 362). Understanding why families approved for the voucher program actually relocate and where they decide to live requires going "beyond housing supply to consider the dynamics of residential mobility and the institutional context of housing policy itself" (DeLuca et al., 2013, p. 271). With a backdrop of long HCV waiting lists presumably reflecting the documented undersupply of affordable housing, policymakers seek to better understand why wait-listed voucher applicants are so reluctant to take advantage of the broader housing options offered by the HCV Program.

Conceptual Framework and Analytical Approach
The mixed and sometimes unexpected outcomes regarding take-up patterns suggest a need to revise the theoretical approaches to understanding how voucher use interacts with the expected outcomes and an opportunity to recalibrate programmatic expectations. as noted, this article synthesizes qualitative data to enhance our understanding of the program.
Researchers generally accept two primary roles for qualitative research: to help explain the results of prior quantitative work, and to develop hypotheses (Small & Feldman, 2012). Qualitative data can help researchers and policy makers understand what was going on inside the "black box" of the voucher program and to generate new hypotheses (Kling, Liebman, & Katz, 2005).
The goal of this metasynthesis is to make some progress toward explaining the take-up patterns observed across HCV studies. To that end, i examine some assumptions undergirding much of the pertinent research and programmatic expectations, specifically those tied to the notion that the HCV Program gives voucher users free and full choice to participate in the private housing market. i then review descriptions of participants' attempts to use the voucher program to access the private housing market and analyze how participant experiences interact with these assumptions.

Qualitative Metasynthesis
The qualitative metasynthesis approach taken here may be particularly relevant in the voucher context because the authors of qualitative studies on vouchers emphasize the unique circumstances of the particular program studied. This necessarily leads to caution in drawing generalizations, developing theory, or devising policy implications. The synthetic approach used here attempts to identify consistent themes across studies in order to make arguments about the validity of a more generalizable hypothesis or theoretical refinement. The method is an attempt, then, through a process of translation and synthesis, to bring individual qualitative studies together with one another at a more abstract level (Jensen & allen, 1996) and analyze the original findings with the goal of generating new interpretations (Thorne & Paterson, 1998).
There are some problems with this approach. The first is methodological: the approach attempts to meaningfully synthesize studies stemming from a variety of methodological approaches to the qualitative data. The second is that a fundamental aspect of the qualitative study is its singularity of context and the consequent exploration of how participants interpret their experiences. Therefore, the concept of bringing a number of qualitative studies to higher levels of abstraction and theory risks violating one of the essential aspects of qualitative inquiry (Jensen & allen, 1996;Sandelowski, Docherty, & Emden, 1997). a metasynthesis attempts to reinterpret other researchers' interpretations. Ultimately, it is the metasynthesist's task to acknowledge these limitations and develop an analysis that is logical and yields compelling and testable theories.
Following Paterson, Thorne, Canam and Jillings (2001), the method of metasynthesis used in this study consisted of four distinct processes. The first step was to define the purpose and research question. The purpose was to synthesize the qualitative studies that help answer the question: "How do descriptions of housing voucher applicants' experiences help further explain the take-up patterns described in the quantitative literature?" For the second step, i conducted a search for relevant literature. i primarily used the search tool Google Scholar to identify articles pertaining to housing vouchers and other related terms (e.g., HCV, Section 8). Relative to other academic search engines, 1 Google scholar allows for a more exhaustive search because it scans the entire text, rather than only abstracts and keywords and includes conference articles, books, theses, and government publications. as others have noted ( Thomas & Harden, 2008;Whittemore & Knafl, 2005) including non peer reviewed literature can be particularly relevant in qualitative meta-syntheses. This project in particular required studies that offered descriptions of participant experiences, whereas peer-reviewed articles tend to be limited to qualitative studies that either build or test theory.
i used search terms to identify articles that reported studies using a qualitative research design. The search yielded 212 items. i then reviewed these items to identify ones to include in the metasynthesis. The inclusion criteria were the following: (a) published articles, reports, dissertations, and book chapters which included descriptive data on participant experiences; (b) investigations of location choice deliberations; and (c) articles about housing vouchers in the United States. i augmented the database search with forward and backward citation tracking and contact with researchers and practitioners in the housing voucher field. after applying these criteria, the data set contained 20 articles as shown in Table 1. i entered the articles into the nVivo qualitative data analysis software for coding and analyzing unstructured data.
in the third step, i grouped the findings by issue. i began by identifying the data analyses method, which was to use the a priori themes of location choice deliberations, such as assessing unit quality and neighborhood quality, to develop sub themes in a grounded fashion. The method was both inductive and deductive, as it articulated existing assumptions and examined them against the body of qualitative evidence, thus allowing for the revision and refinement of existing assumptions and theories. i read the studies three times: first for an initial understanding, second for an initial coding and development of grounded themes, and third for a final grounded coding. The grounded analysis led to 32 distinct sub themes relating to the a priori themes. The third reading of the studies coded for these grounded themes. For metasyntheses with fewer than 50 studies, a single coder may be used (Lipsey & Wilson, 2001), and that was the procedure that i followed for this article. The fourth and final step involved reviewing these themes, their relative frequency in the body of data and their relationship to the theoretical framework. Two themes dominated: market problems and product problems. These two themes served as support for the new theoretical framework and the supporting main concepts. i assessed the remaining themes for their applicability to the dominant concepts and either categorized as sub themes or excluded from final analysis. The analysis below presents these themes and their applicability to explanations regarding location outcomes and program utilization.
The studies highlighted below vary in terms of the circumstances through which the participants obtained their vouchers and the constraints placed contingent upon receiving the vouchers. Some participants came voluntarily, having signed up for a mobility program (i.e., briggs, Popkin, & Goering, 2010;Pashup et al., 2005), and others were subjected to forced relocation (i.e., Popkin, 2000;Varady & Walker, 2000). Still others were randomly assigned to an experiment (Wood et al., 2008). The effort here is to attempt an analysis that identifies themes that apply across circumstances and thus may illuminate programmatic outcomes and policy implications that apply more generally.

Findings
i present the findings below in four sections. The first two focus on how structural problems endemic to the housing market also apply to the housing voucher rental process. The remaining two focus on specific programmatic problems that prevent voucher holders from participating fully in the private rental market. Vouchers are often framed as providing poor households with a subsidy enabling them to pay fair market rents. Many policymakers and researchers assume that the program gives participants options: "a choice on housing location, voucher recipients can decide what best suits the needs of their household, opting to sacrifice space for a better school district, for a house instead of an apartment, or for a shorter commute to their place of employment. There are many factors that go into deciding on a location to live and the voucher approach is seen as providing that freedom of choice for families" (Matthews, 1998, p. 12). Thus, many researchers and policy makers assume that the housing voucher enables low-income households to participate in the mainstream housing market, affording them choices of location similar to those facing middle-income renters.
The qualitative studies provide detailed accounts of participants' attempts to choose where to live, illustrating how problems in the housing market limit choices (DeLuca et al., 2013;Kleit & Manzo, 2006;Ross, Shlay, & Picon, 2012). Data from the cases is organized below to address this free market assumption and details how participants, after obtaining a housing voucher, experience constrained location choice.

Market Barriers: Transportation Constraints
Most housing subsidy recipients lack personal vehicles and must rely on public transit, ride sharing, or travel by foot, and this reliance constrains their housing options. in general, the consequences of uneven availability of public transportation on housing accessibility are well documented (Squires, 2002). a market barrier of this type has not always been integrated into the theoretical framing for voucher programs, despite the qualitative data confirming that voucher users are also subject to similar transportation constraints. One voucher recipient in baltimore explained her logic for choosing a central city neighborhood: "and you move to a location where the nearest grocery store is two miles away. How are you going to go there and get back with a full bag of groceries?" (Varady & Walker, 2000, p. 125). in another study, a voucher recipient from Chicago, illinois, explained her location choice: "i prefer to stay in the city. For one, if my child gets sick and me, i don't have a car. Where i live, i could walk to the county [hospital]" (Popkin & Cunningham, 2008, p. 34).
Some participants view transportation accessibility as a key feature in location choice, as a recipient from buffalo, new york, explained: "it's got to have transportation … the busses don't run there, so it's a bad situation" (Popkin et al., 2003, p. 191). Many voucher holders reported their need for housing with access to public transportation outranked the desire to live in lower poverty communities (Galvez, 2010b;Walker, 2014).
notably, although some researchers anticipated that participants would list access to networks of social support as important considerations when choosing a place to move (Levanthal & brooks-Gunn, 2003), this theme did not appear prominently across studies in the qualitative analysis of location choice. Rather, access to family and friends appeared most prominently as a subtheme within the transportation accessibility theme. Participants did express concern with how changing locations might affect their social relationships, but less in terms of proximity to family and friends than in terms of accessibility to them. Possibly because transportation accessibility was an important feature of location choice, moves did not necessarily negatively affect social relationships. as briggs (2006, p. 37) found in the MTO studies: "it is not yet clear to what degree MTO movers lost social support, or experienced a shift in informal support, based on relocation distance or other access factors. " Thus, in considering a new location, participants seemed interested in assuring accessibility to dense public transportation networks to reach friends and family in addition to work, school, and the institutions required to meet their daily needs.

Market Barriers: Race and Source of Income Discrimination
The data related to voucher recipients' searches for housing show at least two kinds of discrimination: source of income discrimination and racial discrimination. Racial discrimination generally discourages members of low-income minorities from moving to predominantly white or suburban neighborhoods, even if affordable housing is located there (yinger, 1998). Many qualitative accounts of voucher user's location deliberations provided evidence of source of income discrimination, specifically the reliance on a government-subsidized housing voucher. Some hypothesize that source of income discrimination "may artificially limit the housing and neighborhood options available to voucher holders" (Galvez, 2010b, p. 6). Reluctance to accept vouchers appears to be because both negative perceptions of the housing authorities who administer voucher programs and of voucher holders themselves (Greenlee, 2011;Scott, 2013).
Many studies relate participants' accounts of being subject to source of income discrimination. One participant from Chicago recounted an experience with a potential landlord: "She was like, she don't rent to Section 8 [i.e., voucher holders] at all. … and i said, 'Just because a person has a … voucher, that doesn't make the person mean or unorganized … or an unclean person!'" (Pashup et al., 2005, p. 376). a participant in Kansas City, Missouri, said: "The landlords i spoke with were very nice and very positive, but they didn't want Section 8 vouchers" (Varady & Walker, 2000, p. 134). according to another participant: "They [i.e., the owners] had the stigma about everybody that's on Section 8 are nasty, the children tear up the house" (Wood, Turnham & Mills, 2008, p. 392). in a study of a "mid western city, " Teater (2011) reported that one recipient noted: "There is stigma attached that says these are all people that are gonna tear up your unit, that are going to not work, they're gonna have high crime, they're not doing anything but having babies and things like that" (p. 511).
Some participants in the studies described perceived racial discrimination. Popkin and Cunningham (2000) recounted one Chicago resident's experiences with a prospective landlord. at first, the participant explained that the landlord initially encouraged the applicant: "Oh, yes, you come over here, i want to meet you. " However, the applicant explained that: "i got there, i guess she see my black face and changed her mind! She was polite … but i didn't get the apartment … she talked to me on the porch. but i knew what time it was, i'm not stuck on stupid" (Popkin & Cunningham, 2000, p. 44). Studies provide less documentation of direct racial discrimination against voucher holders who are members of minority groups, relative to details about source of income discrimination. Some authors surmise that this is because prospective tenants do not conduct their housing searches in nonminority neighborhoods (Varady & Walker, 2000). Others suggest that discrimination against voucher holders may merely mask racial discrimination (Popkin & Cunningham, 2000). Voucher program participants in many regions are disproportionately nonwhite compared with the overall population. based on a study in Seattle, Washington, Galvez concluded: "landlords in these MSas may view voucher assistance as a proxy for race, and avoid all voucher holders in an effort to avoid nonwhite tenants" (Galvez, 2010b, p. 6).
Several studies demonstrate that the source of income protections-laws prohibiting housing discrimination based on the source of income being a voucher-do improve outcomes for voucher holders (Finkel & buron, 2001;Freeman, 2012). Galvez found that despite the local source of income protections in the City of Seattle: "Half of the focus group and interview participants experienced or perceived landlord discrimination because of their voucher status" (Galvez, 2010a, p. 2). This suggests that the ordinance alone may not be enough to eliminate discrimination (McClure, 2005).

Product Problems: Partial Subsidy
The granting of a housing voucher does not address all of the conventional ways in which the market evaluates the renters themselves. ideally, "a voucher user can select from a variety of apartments on the private market-offering many more choices within the same budget constraint" (Kling, Liebman, Katz, & Sanbonmatsu, 2004, p. 3). in the private market for rental housing, a prospective tenant inspects a unit, assesses whether the price is competitive and whether the unit fits his or her needs, and, if the unit is deemed suitable, attempts to enter into a contract with the landlord. in the contract, the renter agrees to pay a monthly sum in exchange for the right to use the unit for a fixed period of time, usually a year. This agreement puts the landlord's capital at some risk. The most obvious risk is that the renter will damage the unit or depreciate the landlords' capital beyond what the landlord anticipated when setting the monthly rent. another risk is that the tenant will terminate the lease before the end of the contract, leaving the landlord without the anticipated revenue stream. For these reasons, the landlord often asks to review the tenants' credit history, which is seen as a proxy for timely rental payment, and asks that the tenant share in the risk by putting down a deposit, which the tenant agrees to forgo in the event of serious damage or early departure.
The data demonstrate that possessing a voucher does not allow users to negotiate competitively in that market. To begin with, voucher holders who attempt to use their vouchers in the private market often are subjected to credit inspections. a landlord usually expects a prospective tenant to bring documentation of assets including thier credit score (Marr, 2005). in a study of a voucher program in Chicago, Pashup et al. (2005) commented: "Credit problems were common, and participants did not know how to correct them or present them in a more positive light" (p. 382). because only a portion of the monthly rent is subsidized and the voucher-holding tenant must pay 30% of his or her income to the landlord, landlords may be concerned that the portion of rent to be paid by the renter is still too much risk for them to assume, according to the landlord's standard criteria. (it is impossible to know whether assessment of credit risk masks underlying discrimination.) in addition, the housing units of landlords who lease to voucher holders are subjected to a secondary inspection to determine if the unit meets HCV Program guidelines for unit quality and safety. The landlord bears a cost for this process in terms of lost time on the market and expenditures if improvements are required. The program does not compensate landlords for lost income associated with waiting for the housing unit to be inspected and certified. Many studies documented the additional time spent waiting for inspectors and verification. Varady, Wang, Murphy, and Stahlke (2013) recounted one landlord's perspective: "The fact that their tenants and their program cost us a lot of money-they don't care" (p. 123). according to Pashup et al. (2005), in Chicago: "respondents reported that landlords complained about how extraordinarily long it took for a unit to be inspected and approved" (p. 376).
Other accounts, such as that by Marr (2005), document the frustration of landlords related to the costs associated with delays while waiting for certification.
Moreover, voucher holders usually lack the funds for a deposit (typically 2 months at the market rate rent). Security deposits are explicitly excluded as a matter of policy. The HUD Housing Choice Voucher Guidebook states: "The cost of the security deposit is not covered under the Housing Choice Voucher Program" (HUD, 2014, section 8.2 at p. 8-4). HUD instructs PHas to conduct a "briefing" after the PHa determines that the family is eligible for assistance. The PHa is expected to "explain at the briefing that the family is responsible for any security deposit requested by the owner" (HUD, 2014, section 8.2 at p. 8-4). Galvez (2010b) reported that for participants: "deposit fees, however, were by far the most pressing concern, requiring voucher holders to find landlords willing to accept deposit payments over time, or reduced payments" (p. 11).
Often, the cases demonstrate how the programs must appeal to a landlord's sense of mission or charity, which is not a sound economic basis for a program. Marr (2005) has documented that the program "often involves soliciting landlords' understanding and compassion" (p. 100). Scott (2013) cited the following reason for landlords to participate in the program: "To get rent paid on time and to provide a service for those in need" (p. 72). HUD encourages public housing authorities to identify local community resources "for which families can apply to complement their housing assistance. This might include any services of financial assistance for security deposits" (HUD, 2014, section 8.2 at p. 8-4). These studies documenting that housing counselors stress the charitable aspects of the program suggest that the voucher is not functioning as a market-based product. as a result, voucher holders may rely on a "Section 8 submarket" consisting mainly of landlords who already rent to voucher households or have done so in the past (briggs & Turner, 2006;Kennedy & Finkel, 1994).
Other studies have demonstrated that the guaranteed partial rent provided by vouchers actually motivates certain landlords, especially in weak rental markets. Such landlords may readily be incentivized to make financial concessions in terms of the security deposit or credit requirements. For example, participants in Seattle reported that: "Seattle's relatively weak housing market seemed easier to navigate … and 'move in specials' with fixed application and deposit fees were common" (Galvez, 2010b, p. 12). in neighborhoods where demand for rental housing is strong and vacancy rates are low, however, landlords "may have little incentive to accept Section 8 tenants" (Katz & Turner, 2001, p. 241).
These dynamics affect the quality of housing subsidized that renters are able to secure. Researchers noted that "successful searchers reported that they felt they had settled for a less-than-ideal unit. " in Chicago, researchers reported that participants planned to move as soon as their leases expired because they were living in dangerous neighborhoods. Many cases reported how participants appeared to compromise on preferences because they needed financial concessions from landlords (DeLuca & Rosenblatt, 2010;Pashup et al., 2005;Smith et al., 2002). "Participants frequently discussed the need to find 'good' landlords who would not only accept vouchers but also 'work with you' to negotiate an affordable lease agreement" (Galvez, 2010b, p. 14). in Washington, DC, participants reported reduced choice due to lack of deposit: "not many options because apartments check credit" (Scott, 2013, p. 78). Respondents in a Seattle study concisely stated the locational outcomes associated with poor credit: "The credit checks determine where you live" (Galvez, 2010b). These product problems appear to reduce choices or lead the tenant to accept below market rate quality units at market rate rents.
Other data suggest the product constraints also affect whether applicants use the vouchers at all. Some rental applicants did not use their vouchers because they reported that they were unwilling to live in bad units or neighborhoods (Popkin & Cunningham, 2000). Some users believe that lack of a deposit prevented them from relocating. as one participant in Denver stated: "you got to do your deposits so … it's a lot of responsibility if you're not ready for it, don't do it" (Walker, 2014, p. 104). as HUD acknowledges: "not having budgeted for [the security deposit], the family fails to lease under the Housing Choice Voucher Program. " Thus, a household lacking the money for a security deposit may subsequently be unable to participate in the voucher program.
Even without discrimination from the landlords' side, these product problems are likely to reduce use and drive users to lower than average quality units in lower than average quality locations. The details provided by program participants indicate that landlords' hesitancy to work with the program is not exclusively discrimination in the sense of employing negative stereotypes of all voucher holders. Discrimination may not be at play if landlords are turning away potential tenants because the tenants cannot meet the terms of the contract or the landlord does not like the program design or demands. Rather, the qualitative demonstrate how voucher holders, administrators, and landlords implicitly understand that the HCV is a less competitive alternative to, not a direct substitute for, private market rent.

Product Problems: Constraints on When Participants Can Move
Studies also call into question the freedom of choice assumption as it relates to a voucher user's ability to determine when to move. Private market renters typically have considerable discretion about when they relocate. HCV Program participants, on the other hand, have little choice regarding the time frame of when to use a voucher. as Carlson et al. (2012, p. 102) note: "Obtaining and using a Section 8 voucher is a multistep process that routinely spans multiple years. " These time constraints appear in two forms. The first is the lengthy but unpredictable timing of voucher availability. Voucher applicants most typically place themselves on a waiting list, and the elapsed time from initial application to receipt of a voucher ranges from 2 to 10 years for some housing authorities (Carlson et al., 2012;Finkel & buron, 2001). Various cases confirm this pattern. in Chicago, about 30,000 families were wait-listed for housing assistance, and families waited 7 to 8 years for voucher assistance (Jacob, 2003). in Los angeles, California: "the estimated waiting period for the 150,000 eligible households seeking regular Section 8 vouchers in the city was 8 years" (Marr, 2005, p. 88). Participants expressed frustration with the wait-list process. For example, DeLuca et al. (2013) highlighted one participant: "Keisha, a mother of two, described the long lines and uncertainty of the process: 'Fill out an application, and they'll call you once you reach the top of the list. it may be 1 year, it may be 4 years'" (p. 277). This level of indeterminacy does not allow voucher holders to plan for moves or chose a time frame best suited to their life circumstances.
Various accounts discussed some consequences of the lag times between filling out an application and issuance of a voucher. an obvious one is attrition: a voucher applicant ready to relocate at the time of application may experience changes in life circumstances during the intervening period. DeLuca et al. (2013) have detailed how participants in baltimore, Maryland, discussed having spent years "on the wait list, which made for a seemingly random initiation of the housing search for families" (p. 288). in Washington, DC, Scott (2013) found that there are a "significant amount of people who have moved from the area or whose situation has improved by the time their name gets to the top of the list" (p. 77). Certain positive changes in life circumstances may have meant they were no longer motivated to use a voucher or had less time to execute a search. it has been difficult to assess the percentage of vouchers issued to applicants that remain unused due to changes in circumstances because many applicants do not reply to their notification of availability. To date, most studies have produced limited information about voucher holders who did not make use of a voucher to relocate.
Moreover, voucher applicants face another timing obstacle in addition to indeterminate wait-listing. Once notified that a voucher is being issued, a voucher recipient has limited time to locate housing. They usually must move in 60 to 120 days. if the applicant does not locate housing within that window of opportunity, the authority rescinds the voucher. Studies provide extensive details about the "use it or lose it pressure" (briggs & Turner, 2006, p. 40) and its consequences. The research commonly reported that voucher recipients felt pressed for time and that "your time is just a clicking" (Smith et al., 2002, p. 23) and often chose lower quality housing as a result (Wood, Turnham & Mills, 2008). Popkin and Cunningham (2000) reported that a voucher recipient in Chicago chose an apartment but "it was in the middle of winter. and like i said, i ended up just having to take something that i really didn't want" (p. 50). according to DeLuca et al. (2013), one resident of baltimore explained her compromise because of both time and deposit constraints, which led her to accept a unit, despite a number of "red flags, " as she put it. but she "took him up on it because i knew i had to get somewhere, if i didn't i was going to lose out on my voucher all together" (p. 288).
Time constraints affected the quality of the neighborhoods where participants move. Kansas City housing participants in Varady and Walker's study (2000) "also stressed the time constraint placed on residents there, which meant limited assistance to move into less known neighborhoods located farther away" (p. 137). in illinois, Greenlee (2011) noted that the demand for recipients to quickly locate a unit meant that households were not "able to take full advantage of making more informed housing searches due to time limitations" (p. 400). Wood et al. (2008) similarly concluded: "as a result, many made expedient housing and neighborhood decisions that later proved unsatisfactory" (p. 392). Voucher holders find themselves competing in an open rental market although hampered by time restrictions, and this limits where they can live (Guhathakurta & Mushkatel, 2000).
in fact, program time constraints contribute to the low utilization rate of vouchers. as Galvez explained in her analysis of the Seattle Housing authority voucher program: "the housing success rate hovers around 60%, meaning that approximately four of 10 voucher holders fail to find housing within 120 days and lose their vouchers as a result" (Galvez, 2010b, p. 17). DeLuca and Rosenblatt (2010) too posited that some residents do not use the voucher because of the long wait.

Discussion
in the HCV Program-as is often the case in policy research-some of the measured outcomes of didn't meet the anticipated ones. although certainly the implementation of programs in real-world situations introduces an element of randomness, sometimes unexpected outcomes are due less to the absence of systematic mechanisms than to the operation of alternative ones. This article synthesized the qualitative data on housing choice deliberations of voucher holders to determine if alternative processes led to some of the observed outcomes. as such, it identified key structural and programmatic assumptions underlying the research framing the HCV Program and examined how qualitative studies of the program generally challenged these assumptions. The first assumption is that by subsidizing the cost of rental housing, the HCV facilitates access to the private rental market and allows freedom of choice by participants about where to move. The second assumption is that the HCV Program functions to make the voucher holder equivalent to a renter who can pay fair market rent.
To examine how these assumptions align with empirical data, i have analyzed voucher program participant experiences across a variety of regions, market conditions, and programmatic configurations. Data across study contexts served to challenge assumptions about how program participation intersects with the structure of the housing market. My analysis underscores that two forms of market barriers constrain participants' mobility across a variety of housing markets: (a) the inaccessibility of public transportation in many neighborhoods and (b) implicit forms of discrimination-either racial or source of income discrimination-that exert a profound, yet often unacknowledged, effect on program outcomes.
The data also challenge the programmatic assumption that possessing a voucher makes a prospective renter equivalent to someone who can pay fair market rent without a subsidy. Voucher holders often lack a security deposit and good credit standing. in addition, voucher holders are under pressure to find suitable housing within 60 to 120 days and thus lack control over timing the relocation relative to their familial obligations and work schedules. Freedom of choice regarding where to move and when to make the transition appear to lower quality of housing users can secure, both in terms of the housing unit itself and its location. This analysis also suggests that such constraints likely contribute to lower voucher use. Taken together, these data do not support the fundamental assumption that the program allows recipients to "exercise free and full location choices" (Katz & Turner, 2001).

Policy Implications and Future Research
Many factors constrain where low-income households are able to live-regardless of whether their housing is subsidized. However, the role of housing policy in residential decision-making is highly relevant to housing location outcomes, yet poorly understood. This analysis suggests several areas for policy and research. additional meta-analyses and syntheses of the program are needed. For example, a quantitative meta-analysis of the experimental and quasi-experimental approaches might lead to new insights or promote consensus regarding programmatic outcomes. additionally, researchers might undertake a qualitative metasynthesis focused on other aspects of the program, such as participant experiences subsequent to relocation.
additionally, policy makers and researchers might enact a number of responses to the structural and programmatic constraints affecting program participants. On the one hand, structural housing market constraints reflect systemic problems that may be beyond the scope of program-level adjustments. as such, participants in the program will likely benefit from policies that succeed in reducing identity discrimination and improving access to destinations that people need and want: school, work, shopping, friends, family, medical, and recreation facilities and the like, whether that happens through improved public transportation, taking advantage of technological change that enables car sharing, or some other approach.
To the degree that discrimination reflects product problems rather than prejudices, problems could be addressed in a variety of ways. although applicants for housing subsidies are wait-listed for many months, housing agency administrators could provide credit counseling to voucher applicants to improve their apparent risk profiles for prospective landlords. To make the voucher product a more attractive option to landlords, repair assistance programs and other forms of financial assistance are ideas worth exploring.
To make the program more appealing to low-income households, the HCV Program could allow voucher holders to have greater control over when they move. because moves during the summer months are optimal times to change schools (Rumberger, 2003), moves during this time may have superior outcomes in terms of both take-up rates and location outcomes. Consequently, this analysis suggests a new hypothesis: The time of year in which the voucher is distributed may affect the take-up rate and the quality of the move. This hypothesis can be tested through the analysis of historical administrative data or through the implementation of a randomized control pilot program.
although this qualitative metasynthesis focused on the examination of certain theoretical assumptions and the relevant empirical realities, other factors are clearly at play in regard to the location outcomes of voucher recipients. These factors present many areas for future policy and research. Some have proposed the idea of instituting small-area fair market rents (FMRs) intended to address inequities arising from variations in rent differences in various markets. Small-area FMRs would enable vouchers to be used more readily in higher rent neighborhoods, while simultaneously ensuring that landlords of poorer quality housing do not receive excess rent (Olsen, 2014).
Researchers also assume that characteristics of the individuals who are selected into a housing voucher program and the housing counseling that they receive are also factors in where housing voucher holders relocate (Jacob, 2003;Walker, 2014). Many assumptions about participant selection bias suggest that participant motivation is also a factor in location choice, as well as a factor in whether someone who receives a housing voucher actually makes use of it. To date, understanding of participant motivation is limited, particularly how a participant's life circumstances and subsequent motivations differ at the time of applying for housing assistance and months-or years-later when a voucher is actually issued. How participant motivation drives participation or changes over time could be other avenues for future qualitative research.
Distinctions in how poor and non poor people are treated and how they behave in the housing market are unlikely to be eliminated by any subsidy or assistance program. nevertheless, the research described here demonstrates that the expectations placed upon the voucher program do not always align with structural realities or programmatic constraints. Researchers and policymakers could adjust both the theoretical framing and programmatic structure of HCV programs in ways that reflect the reality of program participants' experiences and housing market dynamics. a HCV program can and should help to deconcentrate poverty and improve the life prospects for lower income households. Progress toward these goals is best served by gaining a more accurate understanding of program dynamics.

Notes on Contributor
Erin Graves is a policy analyst at the Federal Reserve bank of boston.