Good Policy, Bad Implementation: Evaluating HUD’s Shift Toward Voucherization in Low-Income Housing

This was a paper I wrote in my graduate-level Economics of Regulation class in December 2010. My interest in the subject was sparked by an article on New Orleans’ public housing reforms I had written earlier in the year for The Hullabaloo.

Good Policy, Bad Implementation:
Evaluating HUD’s Shift Toward Voucherization in Low-Income Housing

Abstract

Since the 1970s, the federal government’s approach to low-income housing has been undergoing a substantial change from supply-side to demand-side subsidies. The aim of this paper is to evaluate the cause of the U.S. Department of Housing and Urban Development (HUD)’s shift away from urban public housing high-rises toward current voucherization policies and the effectiveness of these newly adopted programs, regarding the extent to which they accomplish their goals of providing housing to the poor, deconcentrating poverty and allowing for voucher recipients’ choice and mobility. Past literature has exposed geographical unevenness in the results of voucher programs; while some cities have experienced favorable outcomes concerning voucher recipients’ abilities to move into areas of low poverty, other cities showed quite opposite phenomena. This variation among cities suggests that while HUD’s voucherization policies do have the potential to be successful in theory, they are poorly implemented by certain local public housing agencies, reflecting a structural failure within HUD to enact effective centralized oversight.

Introduction

It has been a heated debate for decades: how to best provide public housing assistance for the poor, while fostering growth as a society. American cities—the majority of which are de facto racially segregated—are among some of the most unequal in the world, in terms of adjacent neighborhoods’ varying levels of income and access to educational and economic opportunities. Indeed, much of the rhetoric heard at public housing hearings in New Orleans today involves talk of modern-day slavery and exploitation of the poor black masses for white private gain.
Many proponents of a relatively recent policy shift to voucherization hope that through providing public housing assistance vouchers to the nation’s poorest families, poor children will be able to grow up in areas of low poverty and be able to take advantage of these otherwise unattainable opportunities to ultimately achieve economic success.
This paper will examine the cause of the shift in housing policy from command-and-control to more decentralized voucherization. Subsequently, this paper will attempt to evaluate the policy change’s effectiveness in deconcentrating poverty and promoting voucher recipients’ ability to choose where they want to live, exploring in particular the case of New Orleans as an example of deeply flawed implementation.
Through examining empirical evidence and past research, this paper finds that voucherization has resulted in geographical differentiated outcomes. That the policy succeeded in some cities and failed in others suggests that this policy is not inherently flawed; however, implementation at the local level is key. This incompetence at the local level reflects a structural failure within the U.S. Housing and Urban Development (HUD) agency to accurately oversee its 2,400 local public housing agencies (PHAs).
Background Information
Over the past 40 years, the federal government’s approach to low-income housing has undergone a substantial change from supply-side to demand-side subsidies. While the role of the U.S. Department of Housing and Urban Development (HUD) used to be more of a housing provider (i.e. builder and landlord), HUD is now taking steps to decentralize the housing system by providing vouchers for low-income renters to use in the private market. From the time HUD vouchers were first introduced, the amount of low-income households receiving these subsidies nationwide has grown from 30,000 in 1970 to 1.4 million in 2000 to 3.3 million in 2010. Not only has the amount of vouchers in use increased, but their overall proportion of the total number of HUD-assisted low-income households has increased dramatically as well; while vouchers accounted for only one-third of these households in 2002, they now comprise roughly 66 percent of HUD-assisted households in 2010. Meanwhile, 196,000 public housing units have been demolished across the country since 2006, and 200,000 more are currently planned for future demolition (Cohen 2010).
Table 1. Increasing Section 8 Tenant-Based Assistance Enacted By Congress
Fiscal Years 1975-2000

Many consider this shift toward voucherization to be a critical move in HUD’s effective accomplishment of its central mission of providing decent, safe, and affordable housing. Though HUD was not formally created until 1965 as part of President Lyndon Johnson’s “Great Society,” Congress first established the critical priorities of U.S. housing policy under the Housing Act of 1949 to be “the realization as soon as feasible of the goal of a decent home and a suitable living environment for every American family.” Though HUD has not strayed from the core concepts of its mission, its 2010 stated goals are more specific:

HUD’s mission is to create strong, sustainable, inclusive communities and quality affordable homes for all. HUD is working to meet the need for quality affordable rental homes; utilize housing as a platform for improving quality of life; build inclusive and sustainable communities free from discrimination; and transform the way HUD does business.

These goals translate into three main priorities behind the design of public housing policy: (1) To enable poor households to afford “safe and decent” housing; (2) to allow residents more flexibility and freedom in choosing where they live by improving their access to low-poverty neighborhoods, which entail higher quality job and educational opportunities, and; (3) to deconcentrate poverty through facilitating low-income families’ access to otherwise unattainable areas.
In an effort to fulfill these priorities, HUD has instituted new programs, such as the Housing Choice Voucher Program (HCVP, formerly known as Section 8), the Moving to Opportunity Program (MTO) and the Choice Neighborhood Initiative (CNI). The HCVP and MTO provide vouchers and, in some cases, counseling, to residents earning incomes of 30 percent of the area median income (AMI) or less. These tenants must contribute 30 percent of their income to pay for rent, and then the voucher pays the rest up to a limit set by the local public housing agency. The main difference between public housing units and private market renting with a voucher is that with a voucher, tenants must pay for their own utilities. Meanwhile, the HCI works to deconcentrate poverty by providing grants to cities that can prove they have a plan for economic growth stemming from the demolishing of public housing projects followed by subsidized private redevelopment of the area devoted to mixed-income housing.
All of the various long-term goals of these voucherization policies, however, are contingent on the success of the most fundamental one of allowing voucher recipients to choose where they live—especially to move into areas that would be otherwise inaccessible. The de-concentration of poverty and HUD’s other aims to further the poor population’s access to high quality education and job opportunities are only possible if voucher recipients are truly able to choose where to live. This freedom of choice, however, is not just dependent on the voucher recipients’ financial ability to move into these more expensive areas, as the voucher should theoretically take care of that, but also on other conditions, either within HUD’s domain or externally. These conditions include, but are not limited to: landlords’ willingness to rent to voucher recipients, the local public housing agency’s (PHA) timely management of HUD HCV requirements (such as performing house inspections and paying each month’s rent), affordable housing availability in the market (or lack thereof), and increased costs of living associated with suburbanization (particularly transportation costs).
All these contributing factors coupled with the limited amount of data available (due, in part, to the policies’ recent implementations) make measuring the effectiveness of HUD’s voucherization difficult. In this paper I will attempt, however, to determine the extent to which HUD’s policy goals are fulfilled throughout the country. Through analyzing the results of past research on HUD’s shift to voucherization, I conclude that the results are vastly uneven in different geographical areas with regard to voucher recipients’ ability to choose where they live and whether or not this freedom is reflected in the de-concentration of poverty. While HUD’s steady increase in voucherization has produced positive outcomes in some cities, other cities have experienced quite the opposite. While the policies should be successful in theory, these regional disparities in practice are due to implementation shortcomings by certain local PHAs and, ultimately, structural failure within HUD.

Cause of HUD’s Shift

The advent of HUD’s decentralized voucherization policies is the result of what Stegman (2002) calls a “stunning budget-induced transformation.”
Over the past 30 years or so, there has been a paradoxical change: while demand for public housing has soared throughout the country, political support for HUD has fallen drastically. Congress has substantially cut funding for HUD and housing programs proving the Washington Post’s point that “politically, HUD is about as popular as smallpox” (Gugliotta 1995). Since 1980, subsidized housing’s share of new federal spending commitments has fallen over 80 percent; even within the social spending alone, housing assistance’s share of all budget appropriations targeted to the poor fell from more than 36 percent to less than seven percent (Stegman 2002). Cuts of this magnitude have translated into losses of hundreds of thousands of new households that are allowed to be funded annually in assisted housing. While in 1984, 230,000 new subsidized households were added, that number dwindled to 126,000 in 1995. Currently there are roughly 50,000 new families added to public assisted housing annually.

In the same time frame, however, the need for public housing assistance has doubled: the number of American families in critical housing need (i.e. who spend more than half their income on housing or live in seriously substandard housing) increased from 7.2 million to 13.7 million (Stegman 2002). Currently, about 14 percent of all American households have critical housing needs. In addition, 3 million Americans—over half of which are children—are likely to be homeless at least once during the year.
There are many theories as to why this shockingly massive disparity between public demand and political support for low-income housing exists.
Some argue that there has been an ideological shift, manifested in an effort to reform poverty programs in Congress. This is especially evident in the results of former Speaker of the House Newt Gingrich’s 1994 “Contract with America.” In 1995, Congress recaptured $17 billion of funds that had been previously approved but unspent, two-thirds of which came from anti-poverty programs. HUD, as an agency, was clearly a target: out of these cuts, 43 percent, or $7 billion, was cut from HUD (Stegman 2002).
One reason why HUD may have been targeted is because it was seen as incompetent. Structural factors, such as past “inadequate oversight and poor program design” played a strong role in HUD’s losing public confidence. In addition to these structural factors, high violent crime rates, increasing gang membership and racism all contributed to the stigmatization of public housing high-rises. HUD’s incompetence was further highlighted by its long-standing position on the Government Accountability Office (GAO)’s “high-risk” list. Worse still, many HUD officials had been charged with fraud and corruption in the past, most notably the criminal indictment of a sitting HUD secretary during the Reagan administration who “reportedly was so bored that he spent some afternoons watching soap operas” (Stegman 2002).
In addition to an arguably increasing ideological aversion to the concept of public housing for the poor, all these above factors have helped contribute to HUD’s unpopularity to the point where “most people now believe that federal low income housing programs reward a combination of government bureaucrats, politically connected developers, and people who engage in antisocial or self-destructive behavior” (Dreier and Atlas 1996).
These public and political sentiments are symptoms of bureaucratic drift, illustrating HUD’s responses to the multiple problems that can arise from asymmetric information inherent in Congress’ policy delegation to an agency (McCubbins, Noll and Weingast (1987):

One is simple shirking: an agency becomes a Club Med for government officials who undersupply policy decisions. Another is corruption: agency officials allow the bureau to be “captured” by selling out to an external group. Still another is oligarchy: the peculiar political preferences of the agency override democratic preferences. The challenge for political overseers is to prevent these outcomes.

HUD’s budget cuts were also easy for the legislature to pull off because the politically significant middle class does not feel it has a stake in low-income housing or HUD policies. Therefore the target population of HUD’s low-income housing programs is small, fragmented, marginalized, and, most importantly, politically disorganized. Because HUD’s main constituency—the poor—has relatively no political power, there was little opposition in significantly cutting HUD’s budget and scaling back their abilities, and politicians did not have to fear backlash at the polls (Dreier and Atlas 1996).

Another argument as to the cause of HUD’s shift is that there has been a fundamental shift in the nation’s social policy philosophy. HUD’s voucherization “reflects the increasing concern of the U.S. government to refocus attention on individuals rather than on government intervention via housing demolition and construction. This shift reflects a fundamental change in social thinking on how best to combat poverty and help underprivileged populations” (Clark 2008).

Effects of HUD’s Shift to Voucherization

To evaluate the effectiveness of HUD’s voucher policy in accomplishing its goals, this paper will focus on the Housing Choice Voucher Program (HCVP)’s impact on two main preferred policy outcomes: the deconcentration of poverty, and voucher recipients’ abilities to choose where they live.

HCV Impact on Concentration of Poverty
Wang, Varady and Wang (2008)’s findings exemplify the huge regional differences pervasive in HUD’s nationwide shift to voucherization (see Table 1). Through their analysis of public housing “hot spots,” Wang, Varady and Wang (2008) measured patterns of concentration of HCV recipients in eight U.S. cities from 2000 to 2005. First, while HCV recipients showed patterns of moving to the suburbs in Phoenix, Houston, and Chicago, they showed further urbanization in New York, Los Angles, Cincinnati, Miami, and Baltimore. Second, HCV recipients became less concentrated in Chicago and Phoenix, but became increasingly concentrated, meanwhile, in every other city, especially so in New York, Cincinnati and Baltimore. Thirdly, there was no evidence supporting HCVP’s promotion of poverty or racial minority deconcentration in any of the cities studied. These geographically conflicting results suggest HUD policies are not meeting their goals in many areas, possibly due to variation between local PHAs’ implementations of voucherization policies as well as a role for external factors.

Table 2. Summary of Wang, Varady and Wang’s Key Findings (2008)
HCV Recipients’ Suburbanization HCV Recipients’ Clustering HCV Concentration in High Poverty/ High Racial Minority Areas
New York Unfavorable Unfavorable Unfavorable
Cincinnati Unfavorable Unfavorable Unfavorable
Los Angeles Unfavorable Unfavorable Unfavorable
Miami Unfavorable Unfavorable Unfavorable
Baltimore Unfavorable Unfavorable Unfavorable
Chicago Favorable Favorable Unfavorable
Houston Favorable Unfavorable Unfavorable
Phoenix Favorable Favorable Unfavorable
Key:
Bold: favorable outcome (the situation improved between 2000 and 2005)
Italicized: highly unfavorable outcome (the situation worsened substantially between 2000 and 2005)

With regard to using vouchers as a tool for dispersing poverty, William Clark (2008) found mixed results according to geographical area. Clark analyzed tests of differences on data comparing the mobility patterns of both MTO and Section 8 voucher recipients with those of non-recipients in five U.S. cities from 1994 to 2002. He found that the three groups of low-income movers and nonmovers that he studied (MTO voucher recipients, who were required to move to areas of low-poverty; Section 8 voucher recipients, who had total flexibility in where they moved; and the control sample group who did not receive any vouchers at all) all experienced similar gains and distributions throughout the eight years, as exemplified in Figure 2. He says this evidence “suggests viewing the calls for national voucher programs with caution” (Clark 2008), concluding that while “on an individual basis, vouchers can work, and recipients of both MTO and Section 8 vouchers fared better off than those without any assistance in some cases, …this study is not about individual outcomes and localized situations, but about overall policy outcomes of an intervention program” (Clark 2008).

HCV Impact on Recipient Choice
There are many factors that limit HCV recipient choice. Costs are a big constraint—especially transportation costs associated with suburban areas. Landlord discrimination on multiple bases (i.e. racial, social, or perhaps a desire to charge above the Fair Market Rent) is also widely reported. As to drawbacks of the program, HCV recipients have cited low quality of rental housing in the private market, the additional expenses of HCVP (i.e. utilities are not typically included in the voucher, as they are in public housing units), and the lack of quality neighborhoods with available HCV units (Jacob 2004).

New Orleans: Problems with HCV Implementation

HANO

In theory, landlords should be more willing to rent to HCV recipients because their monthly rent payments are government-guaranteed. But in New Orleans, many landlords cite that very reason for choosing not to rent to HCV tenants. Many property owners report that Housing Authority of New Orleans (HANO) is consistently late with both rent payments and mandatory inspections, burdening landlords who choose to rent to HCV tenants both financially and with paperwork.
Tom Hammons, a New Orleans landlord who rents out four houses to voucher recipients, underscored the types of problems that HANO’s mismanagement can cause for HCV tenants:

If I tell my mortgage company that I can’t pay the note for three months because HANO has not paid me, they will take my house. If I have someone working for me and they have delays like that, they will be fired that day. It’s getting very frustrating and I am beginning to have second thoughts about even having HANO homes and possibly giving the tenant a notice to vacate for non-payment. That’s a shame because they did not do anything wrong.

In December 2009, HUD appointed David Gilmore, who had served on multiple cities’ PHAs and a Congressional panel on public housing, to be HANO’s chief and reform the long-dysfunctional agency. In an operational assessment to HUD in February 2010, Gilmore reported the agency’s myriad problems, from poor ethical conduct to being heavily understaffed to a lack of job descriptions. He wrote:

HANO has continued to stand among the few agencies in the country still unable to
manage its own affairs. The cyclical impact is clear both in how HANO is viewed in the
city and how its employees view themselves and their employer; the city’s expectations
of HANO are low so HANO delivers on low expectations. HANO’s employees feel
beleaguered by the negative reputation that plagues them and the agency, so their
performance reflects the resulting low morale.

The operational assessment also reported long, costly delays in HANO inspections and disorganized paperwork so that when inspections are performed, there are rarely records of them. HANO’s website did not provide up-to-date information on HCV units that were available; one study found that nine out of ten listings were inaccurate, regarding either phone numbers or unit availability. A HUD audit claimed that HANO fails to hold landlords accountable for federally mandated quality standards. In 2008 and 2009, HUD found that eight out of 10 voucher households lived in units that did not meet program standards (Cohen 2010). HANO’s problems with implementation prevent the Housing Choice Voucher Program from working successfully in New Orleans.

Concentration of New Orleans Voucher Recipients
In New Orleans, HCV recipients have shown the tendency to be clustered, regarding proximity to both other HCV recipients and their former public housing projects. In a city where all the public housing units have been demolished except the Iberville Development (which is slated to be demolished in the future as well), there are supposedly more HCV units than the number of public housing units that have existed in the past. Though a total of 20,841 subsidized units in the city suggests some degree of neighborhood variety, the 14,000 HCV recipients in New Orleans are currently offered a disproportionate number of available listings in neighborhoods that are among the city’s poorest and least racially diverse. This helps explain why so far, HCV recipients have shown a tendency to be most clustered in three of the poorest and racially blackest ZIP codes in particular: 70117 (9th Ward), 70119 (7th Ward, Esplanade, and parts of Mid-City), and 70122 (Gentilly).
In fact, more than 60 percent of available HCV apartments were located in areas that the Greater New Orleans Fair Housing Action Center (GNOFHAC) classified as “low opportunity,” taking into account indicators for education, jobs, housing, health and the environment (see Figure 6). Additionally, the sites of former public housing projects are also the areas with the largest concentration of HCV units as well as the best access to public transportation and grocery stores, suggesting that HCV recipients have little incentive to stray from these areas.
Figure 6. High Concentrations of HCV (Section 8) Housing in Areas of Low Opportunity

According to a study by GNOFAHC, 82 percent of landlords would not accept vouchers. James Perry, GNOFAHC executive director, has said that without structural reforms within HANO, the voucher system will not work: “HANO has performed so poorly that in some respects it’s valid for a landlord to not want to accept vouchers. If you don’t have a functioning housing authority, you can’t have a voucher program that desegregates poverty.” Despite legitimizing landlords’ reasons for rejecting voucher-holders as tenants, Perry has also called for lawmakers to enact legislation barring landlords from discriminating based on income or source of income.
Another key finding by GNOFAHC’s study was a 57.5 percent rate of racial discrimination: six out of 10 Black testers encountered less favorable treatment than the white tester, all else equal.

Conclusion
The effects of HUD’s change in its approach to low-income housing from being command-and-control to decentralized have triggered notably diverse results depending on the area of the country. While HUD’s steady increase in voucherization has produced positive outcomes in some cities, other cities have experienced quite negative results. This vast disparity between geographical areas suggests that the most pervasive of the nation’s low-income housing problems—concentration of poverty, voucher recipient choice, affordable housing availability, landlord willingness to accept HCVs and access to socio-economic opportunity—do not stem from HUD’s blanket policies, but rather from these policies’ locally differentiated implementations. As there are 2,400 local public housing agencies (PHAs), HUD is unable to effectively oversee each PHA’s functioning. As in the case of New Orleans, even a new and experienced HUD-appointee can falter in implementing a sound policy when faced with an understaffed, disorganized agency that is historically notorious for being corrupt and ineffective.
In any case, HUD’s shift to voucherization is still in its fledgling phase and thus it is too soon to determine completely the effects of this switch in policy. Measuring the effects of housing policy is difficult, and the robustness of the studies used in this paper can be questionable because of a variety of factors: the sizes of studies, assumptions that neighborhoods are the causal agent of certain issues, inadequate controls for family background and characteristics, factors that contribute to neighborhood choice. In addition, as all the studies mention, more empirical evidence is needed to fully ascertain stated conclusions.
HUD’s decentralized structure and top-down implementation is clearly ineffective in ensuring certain PHAs function. More federal oversight and structural reforms are needed for voucher policies to be in any way successful across the country—but success is a possibility.