The Boston Public Interest Law Journal
Copyright (c) 1996 The Trustees of Boston University
The Boston Public Interest Law Journal
5 B.U. Pub. Int. L.J. 239
LENGTH: 15085 words
ARTICLE: THE HUD REINVENTION: A CRITICAL ANALYSIS
JENIFER J. CURHAN*
* J.D., Harvard Law School, 1995. Candidate for Master in Public
Administration, John F. Kennedy School of Government, Harvard University, 1996.
The author would like to thank Professor Duncan Kennedy of the Harvard Law
School for his advice, encouragement and insights.
... HUD's Reinvention Blueprint is a flawed concept -- the program is not
sufficiently thought-out. ... Through the HCF, HUD would universally replace
federal subsidies to public and assisted housing developments, serving 1.2 and
1.4 million families respectively, with direct tenant assistance in the form of
housing certificates distributed by PHAs. ... Certificate holders would be
entitled to use their certificates, set at fortieth percentile FMRs minus
tenant contribu- tions, to live in former public or assisted housing, or to
rent housing on the private market. ... Second, even if FMRs keep pace with
market rents, a certificate holder's ability to obtain housing depends on
housing vacancy rates. ... Assuming that affordable housing were available,
landlord discrimination might still prevent certificate holders from exercising
free choice in determin- ing where to live. ... The HUD Reinvention fails to
consider the effect that the demand shock that voucherization would have on
existing neighborhood equilibria. ... Moreover, given neighborhood effects and
circular cau- sation, the HUD Reinvention might create housing market
instability, result- ing in the segregation of certificate-holders in new areas
of poverty concentra- tion, and the forced departure of certificate-holders
from gentrifying areas. ...
HUD's Reinvention Blueprint is a flawed concept -- the program is not
sufficiently thought-out. There is no one solution to the problems of pub- lic
housing, and even if the Reinvention Blueprint were successful, we've got to
"What's accomplished by this?" We didn't get into the busi- ness because we wanted to be private market
In response to congressional threats of extinction
n2 and mounting popular criticism of public housing conditions,
n3 the Department of Housing and Urban Development (HUD or the
"Department") has proposed to reinvent itself, thereby transforming federal support for
housing in America. HUD outlines its reorganization plans in the 1994
n4 and the 1995
"HUD Reinvention: From Blueprint to Action."
n5 Both proposals suggest implementing aggressive reforms, including
consolidation of the agency's sixty active programs into three
performance-based funds, and implementing a uni- versal voucher-based system
for subsidized housing programs.
Experts agree that HUD's programs are in need of reform. While dis- missing
the stereotype perpetuated by the popular media that the worst projects typify
all public housing,
n7 industry analysts contend that public and
[*240] assisted housing is in a state of crisis.
n8 The majority of developments provide decent and safe housing for residents;
n9 however, Public Housing Authorities (PHAs)
n10 can be poorly managed,
n11 often deferring maintenance expenditures by failing to collect sufficient
rents to meet ongoing capital needs.
n12 Burdened also by HUD's poor monitoring and contract enforcement record,
project- based developments often fall below the Department's Housing Quality
Stan- dards (HQS).
n13 This has sometimes forced tenants to live in barely habitable conditions.
n14 As a consequence of such deferrals, the total cost of modernizing public
housing currently stands at $ 16.65 to $ 29.2 billion.
n15 In a small but growing percentage of public housing developments, neglect has
led to physical and social distress.
n16 HUD policies have concentrated low-income families in densely populated
high-rise public housing projects. They also have forced families to live in
high poverty inner-city areas where job opportunities are low, two-parent
families are rare, welfare dependency is common, and rates of drug abuse, crime
and violence are high and increasing,
n17 HUD policies are thereby exacerbating already serious problems.
On a theoretical level, voucherization
n18 has been heralded as a cost saving program that provides for individual
n19 Proponents have also observed that voucherization may alleviate the
concentration of poverty by enabling low-income families to move out of poor,
[*241] that members of racial minority groups are more likely than whites to live in
n21 such an exodus theoretically would foster the develop- ment of
racially-integrated neighborhoods. Nevertheless, while HUD's voucherization
proposal is rhetorically compelling, a more careful analysis indicates that, in
practice, the above assumptions would not hold true or be practical on a
universal basis. Not only would housing certificates cost more on a national
level than current HUD subsidies, but also, depending on neighbor- hood
dynamics, certificates might fail to enable recipients to exercise free choice
in deciding where to live, and fail to reduce the concentration of pov- erty.
While voucherization might represent an effective policy for the most troubled
and most unfavorably located PHAs by enabling low-income resi- dents to escape
the worst public housing developments, demand-side programs could create
unfavorable effects in other neighborhoods. HUD should assess more critically
the consequences of universal voucherization before overhaul- ing the
Department and changing the way housing assistance has been admin- istered to
After describing the HUD Reinvention proposal in Part II, Part III of this
article explores the plan's cost, choice, and de-concentration implications. In
Part IV, this article recommends that HUD adopt a more flexible approach to
housing policy, involving both targeted demand- and supply-side programs.
II. THE HUD REINVENTION PROPOSAL
The federal government currently funds four types of housing assistance
programs for low- and middle-income families. Under the United States Hous- ing
Act of 1937,
n22 municipalities establish PHAs that build, own and operate low- and
middle-income housing. PHAs enter into Annual Contributions Con- tracts with
the federal government whereby PHAs fund land purchases and construction costs
through the issuance of long term bonds. In turn, the gov- ernment pays the
debt service on these bonds, including principal and inter- est.
n23 Through its Performance Funding System, HUD provides operating sub- sidies to
PHAs, and PHAs agree to operate in accordance with federal statutes and
n24 HUD also finances low-income housing by providing hous- ing subsidies to
private market landlords. HUD-subsidized housing develop- ments have insured,
assigned, or non-insured mortgages and are subsidized
[*242] under HUD's Section 221(d)(3),
n25 Section 236,
n26 and Section 202
n27 programs. HUD's Rent Supplement,
n29 New Construction,
n30 Substantial Rehabilitation,
n31 and Property Disposition Set-Aside
n32 programs provide rent subsidies.
In addition to public and subsidized housing, HUD operates two demand- side
programs. Under the Section 8 Existing Housing Program ("Rental Cer- tificate"),
n34 HUD signs annual contributions contracts with PHAs or other agencies willing
to administer the program.
n35 Administrating agencies contract directly with the owners of existing units to
subsidize tenant rents, and agree to pay landlords the difference betweenthe
contract rent, which cannot exceed established HUD forty-fifth percentile Fair
Market Rents (FMRs),
n36 and thirty percent of the tenant's adjusted income (the tenant's share of the
rent). The administering agency approves leases between tenants and landlords
and ensures that units meet HQS. The Housing Voucher Program
n37 is similar to the Section 8 Existing Housing Program, except that
administering agencies directly supply tenants with vouchers. Voucher
recipients are responsible for locating rental units that meet HQS, after which
the administering agencies approve leases and enter into contracts with
landlords for assistance payments. Vouchers also differ from certificates in
that voucher recipients are permitted to pay in excess of thirty percent of
their adjusted incomes for rent, and land- lords may charge more than
forty-fifth percentile FMRs. The administering agency continues to limit its
payments to the difference between such FMRs and thirty percent of tenants'
adjusted incomes, allowing tenants to finance any excess rental costs.
Under the proposed HUD Reinvention, HUD would consolidate its sixty active
programs into three performance based funds: the Community Opportu- nity Fund
(COF); the Affordable Housing Fund (AHF); and the Housing Certificates Fund
n39 Through the HCF, HUD would universally replace federal subsidies to public and
assisted housing developments, serving 1.2 and 1.4 million families
respectively, with direct tenant assistance in the form of housing certificates
distributed by PHAs.
n40 The HCF would also pro- vide certificates to the current 1.5 million
n41 Section 8 Existing Housing and Housing Voucher programs' participants.
Certificate holders would be entitled to use their certificates, set at
fortieth percentile FMRs minus tenant contribu- tions, to live in former public
or assisted housing, or to rent housing on the private market.
n42 Certificates would entitle recipients to receive the difference between
fortieth percentile FMRs and thirty percent of their adjusted income.
Moreover, as with the Housing Voucher program, landlords would be able to
charge in excess of FMRs, and certificate recipients would be able to spend
more than thirty percent of their adjusted income on housing.
III. CRITICAL ANALYSIS
A. The Relative Cost of Housing Certificates
If lawmakers genuinely are interested in helping the poor, they would
eliminate the trickle-down programs that throw dollars at the housing industry
and use this money to fund many more vouchers.
HUD's comparison of Section 8 costs with public housing is seriously mis-
leading. Public housing costs less, not more than, Section 8, when true costs
are considered on a long-term basis.
Critics have characterized HUD's current subsidy programs as
n46 Proponents of household-based programs, including voucher and certificate
programs whereby the government gives families the means to purchase housing,
contend that such demand-side assistance is more efficient, and thus less
costly, than project-based, supply-side programs. First, it is argued that more
of each subsidy dollar goes to housing consumption because funds are applied
directly towards family rents.
n47 Funding rents elimi- nates the transaction costs incurred when middlemen --
such as developers, consultants, and landlords -- are involved.
Second, supporters maintain that funding public housing through portable
certificates would subject landlords to private market discipline, causing them
to act in more profit-maximizing ways. Assuming that private landlords pursue
their self-interest and attempt to maximize the discounted stream of income
from their housing stock, they would reduce waste
n49 and channel assets, including land and housing stock, to their most valued
n50 Buildings would be constructed more efficiently,
n51 and operating costs would be reduced.
n52 Landlords would be more likely to better maintain and rehabilitate their hous-
ing stock to attract tenants. Where economically rational, buildings would be
n53 In short, landlords would be motivated to reduce costs, and would thus better
maintain their properties
n54 because certificate holders would no longer be tied to specific buildings and
could choose where to live.
Supporters of demand-side housing assistance have estimated the cost sav- ings
that would be achieved through voucher-based programs. Based on the 1988 Report
of the President's Commission on Privatization, some contend that project-based
programs cost twice as much as voucherization.
n56 This implies that the government could provide higher quality assistance, or
double the number of families it currently serves, for the same amount of
housing expenditures. Others point to a 1988 Congressional Budget Office study
which estimated that the twenty-year cost of housing an elderly household in
public housing would be thirty percent more costly than housing the same family
through a Section 8 certificate.
n57 Recently, Housing Secretary Henry G. Cis- neros estimated that the HUD
Reinvention would save $ 60 billion over the next five years, including $ 800
million in administrative costs.
In spite of the theoretical merits of demand-side arguments, however,
voucherization would be more expensive than project-based assistance in the
context of the current U.S. housing market. Indeed, demand-side assistance has
been criticized even by its most ardent proponents. A 1992 HUD analysis found
that the average twenty-year cost of maintaining existing public housing
inventory on a per unit basis was less than the cost of converting subsidies to
n59 Moreover, in a 1995 analysis, HUD maintained that the per unit cost of current
supply-side programs and Section 8 certificates were equivalent.
An analysis of the actual costs of voucherization demonstrates that, on a
national basis, the HUD Reinvention plan is more expensive than current HUD
supply-side programming. Based on fiscal year 1995 appropriations for its
Perfomance Funding System (PFS), HUD's Office of Policy Development and
Research calculated the per unit month (PUM) subsidy for an occupied unit of
public housing to be $ 481.
n61 National housing associations have reacted
[*246] with a flurry of cost critiques, universally concluding that voucherization
costs more than current HUD subsidies. On the national level, the National
Associ- ation of Housing and Redevelopment Officials (NAHRO) and the Council of
Large Public Housing Authorities (CLPHA) have estimated that voucheriza- tion
costs exceed current public housing costs by $ 46 and $ 63 PUM, respec- tively.
n62 Other associations have compared subsidy and voucherization costs across a
handful of U.S. states and cities. The Public Housing Authority Directors
Association (PHADA), which compared costs across ten states, con- cluded that,
on average, voucherization costs would exceed subsidy costs by $ 126 PUM.
n63 In a second regional study, the Association of Community Orga- nizations for
Reform Now (ACORN) examined HUD expenditures for eigh- teen PHAs and found the
average cost of voucherization to be $ 75 in excess of current public housing
As calculated by HUD, components of the cost included $ 191 in operating
subsidies, $ 20 for the Public Housing Drug Elimination Program (PH-DEP),
[*247] and $ 270 for modernization expenditures.
n65 HUD compared the $ 481 with its estimated $ 440 PUM of replacing PFS subsidies
with certificates. Voucheriza- tion costs were derived by taking the FMR of $
571 based on the fortieth per- centile of rents in the metro areas for the
bedroom sizes of current public housing tenants, adding a $ 46 administrative
fee calculated at eight percent of such FMRs, and subtracting the $ 177
tenant's share of the rent payment cal- culated at thirty percent of the PFS
HUD's analysis is faulty in several respects. First, the PFS, which allocates
operating subsidies to PHAs, is inadequate. Not only did the program's origi-
nal design underestimate the true costs of operating PHAs, but also PFS sub-
sidies have failed to keep up with inflation.
n67 While the $ 2.618 billion
n68 in HUD's analysis may accurately represent PFS costs, the expenditures should
be adjusted upwards to reflect adequate subsidy payments.
Second, HUD should include the expenditures for its Public Housing Drug
Elimination Program in voucherization costs. Under the HUD Reinvention, such
assistance will continue during the first stage of the transformation through
HUD's Operating Fund.
n69 Moreover, issuing certificates will not guar- antee a reduction in drug use
and crime. Given that police departments com- monly have failed to provide
adequate assistance to housing developments with high concentrations of
n70 and because most tenants will either remain in or move to new
n71 landlords will be forced to raise their rents to pay for the costs of crime.
If such increases resulted in rents at or below fortieth precentile FMRs, HUD
would continue to finance these costs. Even if rents exceeded such FMRs, adding
PH-DEP costs to voucherization expenditures accurately reflects the fact that
the fed- eral government is merely cost-shifting this burden onto private
landlords, rather than eliminating an expense.
Third, HUD's modernization costs, which include expenditures for its Com-
prehensive Improvement Assistance (CIAP),
n72 Comprehensive Grant (CGP),
n73 HOPE VI, and Vacancy Reduction programs, are artificially inflated and mis-
n74 HOPE VI expenses, which would continue under the HUD Reinven-
n75 should be added to voucherization costs to reflect this fact. Further- more,
public and assisted housing requires $ 16.65 to $ 29.2 billion in repairs due
to longtime maintenance neglect.
n76 In recent years, Congress has appro- priated historically high levels of funds
to address such neglect through the newly created CIAP, CGP, and Major
Reconstruction of Obsolete Projects (MROP)
n78 HUD's use of such inflated maintenance figures in its analysis are misleading.
n79 A more accurate assessment of PFS subsidies and certificates would compare the
expected long-term costs of the respective pro- grams. Based on an Abt
Associates/ICF study which demonstrated that, assuming good public housing
stock conditions in 1987, maintenance expendi- tures would be $ 2 billion per
year or $ 125 PUM, CLPHA has calculated PFS subsidized housing maintenance
expenditures to be $ 166 PUM.
n80 Substituting this figure for HUD's $ 270 PUM for modernization expenditures
decreases total PFS-based expenditures to $ 377 PUM, or $ 63 less than HUD's
esti- mated $ 440 PUM voucherization cost.
n81 As an alternative to the above analy- sis, the costs of HUD's CIAP and CGP
programs could be added to certificate costs. These programs would continue
under the HUD Reinvention to enable the public and assisted housing stock to
compete with private market housing.
Even if costs were not adjusted in these ways, HUD's analysis is far from
conclusive. If fiscal year 1995 modernization costs were reduced by $ 562 mil-
lion per year (fifteen percent of current spending), subsidy and predicted
voucherization costs would be equivalent.
n83 In light of probable HUD budget
n84 a reduction in modernization costs of this magnitude is possible. Thus, in
spite of the analysis conducted by HUD's Office of Policy and Development
Research, voucherization costs would exceed current HUD program assistance on a
per unit basis if this reduction obtains.
Assuming supply-side programs are cheaper, the primary argument in sup- port
of voucherization arguably is negated. It should be noted, however, that these
calculations are highly dependent on the data employed and that the comparable
costs of subsidization and voucherization are relatively close. Moreover, the
decision to voucherize is highly political in nature, and Con- gress will
ultimately determine the amount HUD will spend on either type of housing
program. Given these factors, it is questionable as to whether the deci- sion
to voucherize should be based solely on a cost comparison.
The debate over voucherization costs has been centered on the national level.
This focus is largely a response to HUD's universal model analysis.
n85 Nevertheless, critics note that selectively voucherizing some, but not all,
PHAs would provide for a more prudent program.
n86 While voucherization might be an economical way by which to provide housing in
certain circumstances -- for example, in the case of the most troubled and
least favorably located PHAs -- such a strategy could be disastrous in other
contexts. Accordingly, HUD should develop a more flexible approach that would
allow for targeted voucherization among and within public housing developments.
B. Housing Certificates and Tenant Choice
Demand-side approaches directly attack the root of the problem in most markets
-- the inability of some households to afford decent-quality hous- ing at
current rents for such.
Rather than enhancing resident choice, HUD's blueprint would seriously
constrict residents' already-limited options by lowering Fair Market Rents and
by eliminating the only permanent source of affordable housing reserved for low
income families -- an investment of nearly $ 100 billion.
Proponents of demand-side assisted housing have lauded certificates and
vouchers for promoting household choice by cutting tenant ties to particular
buildings and enabling them to live in public or private housing at a variety
of locations. While individuals who are able to escape the inner cities enjoy
n89 studies and anecdotal evidence concerning voucherization indicate that not all
certificate and voucher recipients are able to exercise such free choice.
Although 1994 HUD studies found that eighty percent of certificate and voucher
holders living in large, urban areas found housing,
n90 these success rates were achieved only because of significant assistance from
programs des- ignated for that purpose. In the Gautreaux program,for instance,
a manage- ment agency actively
"finds landlords willing to participate in the program, notifies families as
apartments become available, and counsels them about the advantages and
disadvantages of the move; counselors accompany them to visit the units and
n91 Smaller housing-search assistance programs also contribute to success rates
for finding housing as the movement of fewer assisted families into wealthier
neighborhoods is less apparent than the arrival of large groups.
Earlier studies of non-search-assisted voucherization programs demonstrate
less favorable results. In a 1986 housing voucher program, forty percent of the
recipients returned their vouchers unused, even with extensions beyond the
sixty-day search period.
n93 Moreover, in a 1990 study, success rates were as low
[*251] as sixty-five percent for rental vouchers, and sixty-one percent for Section 8
n94 These utilization rates are even less impressive since, in most instances,
families do not use certificates and vouchers to move to new geo- graphical
areas. In the 1974 Experimental Housing Allowance Program (EHAP), many voucher
and certificate holders who actually used their assis- tance stayed in their
same neighborhoods, and often in their current apart- ments.
n95 HUD found that while one-third of program participants use their assistance to
stay in place, only fifty-two percent of those who move from their current
apartments find housing.
n96 Voucher recipients searching for housing in rent-controlled areas have an even
lower degree of success.
1. Availability of Affordable Units
The administration . . . wants to cut the dollar value of housing vouch- ers,
in effect pricing poor people out of unshabby areas and keeping them warehoused
within urban slums. The administration hasn't only tossed another unsolicited
bone to Republicans who . . . have never liked the possibility that a few poor
kids might end up in their districts' public schools; it has also forsaken its
professed commitment to policies that help the poor help themselves.
The lack of housing-search assistance is not the only barrier to family
choice. One leading constraint on tenant choice concerns the availability of
affordable, privately-owned apartments.
n99 According to a 1994 Congressional Budget Office study, the supply of
affordable housing is decreasing, particu- larly for the poorest families. In
1975, just enough units rented at thirty per- cent or less of the annual income
of the 5.9 million renters in the lowest quar- tile of the income distribution
to meet housing demand. Subsequently, a gap between the number of renting
families and number of available units has developed and widened, however,
resulting in a shortfall of 3.4 million units for this group as of 1991.
n100 Even though the HUD Reinvention would allow families with housing certificates
to spend more than thirty percent of their annual income on housing, estimates
indicate that even sixty to seventy per-
[*252] cent of their incomes is often insufficient to enable these families to live
where they wish.
n101 Consequently, the privately-owned housing stock may be unable to absorb
significant numbers of the households currently living in public and assisted
housing, especially if tens of thousands of families enter the housing market
at the same time.
A related concern is that voucherization will lead to large reductions in the
number of available units in the public housing stock.
n103 HUD estimates that under the Reinvention, the government will demolish
approximately 80,000 to 100,000 of the worst public housing units, plus
additional units which would have trouble competing on the private market.
n104 PHAs are unlikely to build new public housing, given that the HUD Reinvention
loosens demolition regu- lations
n105 and revokes the current one-for-one replacement rule under which PHAs must
replace demolised units with an equal number of new units.
n106 In addition, experts predict that under the HUD Reinvention fortieth
percentile FMRs will be insufficient to meet maintenance and finance
expenditures. This will force public and assisted housing developments to
increase rents beyond such FMRs, as permitted under the HUD Reinvention,
n107 in order to make necessary repairs.
n108 The fact that public housing is generally of lower quality and often of lesser
n109 than private market units would exacerbate such maintenance expenses. These
dynamics could have the effect of pricing the poorest families out of certain
public and assisted housing markets. Tight public and private housing markets
would lower certificate recipients' chances
[*253] of finding housing even further.
Based on this analysis, in areas where the supply of housing stock shrinks,
landlords will raise their rents in response to the increased demand for hous-
ing. Critics could argue that such dynamics would not affect certificate
holders because FMRs are pegged to the fortieth percentile of local
metropolitan area rents. Nevertheless, this reasoning may not hold true for two
reasons. First, FMRs will not keep up with actual market rents because of the
time lag involved in the annual rate setting process. In highly dynamic,
n111 markets, FMRs might consistently trail market rents, making it even more
difficult for certificate holders to afford housing. Second, even if FMRs keep
pace with market rents, a certificate holder's ability to obtain housing
depends on housing vacancy rates. A NAHRO study found a four percent vacancy
rate and a three percent vacancy rate at the forty-fifth and fortieth
percentiles of local metropolitan area rents, respectively, leading to the
conclusion that such vacancy rates are insufficient to accommodate certificate
n112 A second study, however, determined that units which rent to families with
incomes at or below thirty percent of the average median income show a fifty to
eighty percent vacancy rate, indicating that in these markets housing would be
suffi- ciently available.
n113 Such market specific differentials serve to support the development of a
targeted, as opposed to a universal, voucherization strategy.
HUD's proposal to lower FMRs from the forty-fifth to the fortieth percen- tile
of local metropolitan area rents also would reduce the number and qual- ity
n114 of available affordable units.
n115 While HUD maintains that this five per-
[*254] cent decrease
n116 will not have a great impact on certificate recipients, evidence points to the
n117 Enthusiastic proponents of voucherization admit that, even without such a
decrease in assistance, public housing is more affordable than rents subsidized
by housing certificates.
n118 Rent subsidies have proven insufficient in private market areas with high
n119 and have forced families to forgo moving when confronted with even minimal
n120 Moreover, a reduction in rent subsidies from forty-five to forty percent of
FMRs may force public and assisted housing landlords to raise rents to cover
maintenance and finance costs.
n121 While in theory one might predict that new units would be built to meet
certificate recipient demand, in practice new con- struction has not resulted.
Studies indicate that voucherization neither pre- serves nor adds to the
low-income housing stock.
n122 Private builders are deterred from building low- and mixed-income housing,
given that they are seldom assured of a steady income stream to finance
n123 Traditionally, Congress has had to provide private developers and investors
with tax incen- tives and direct rental project-based subsidies to induce
n124 Because HUD Reinvention certificates are portable, income streams would remain
2. Discrimination Against Voucher Holders and Low-Income and Minor- ity
Assuming that affordable housing were available, landlord discrimination might
still prevent certificate holders from exercising free choice in determin- ing
where to live. Currently, private market landlords discriminate against Section
8 certificate and voucher holders in a variety of ways. A Rutgers Uni- versity
study found that private landlords often refuse to accept tenants by virtue of
their certificate- or voucher-holding status.
n125 Landlords have been deterred from participating in Section 8 programs due to
"cumbersome and unwieldy process imposed by HUD"
n126 and current Section 8 regulations, including the prohibition against tenant
n127 as well as the requirement that units meet HQS and pass annual HUD
n128 Because the HUD Reinvention would not require landlords to accept
n129 once the twenty-year Section 8 certificate and voucher contracts expire as
they will between now and 2002, PHAs and HUD-assisted housing developments that
currently accept such assistance would be free to reject certificate
Private landlords also discriminate against certificate and voucher holders
because they do not want low-income
n130 and minority tenants. Landlords reject poor tenants based on the belief that
such individuals will inject the problems of public housing into their own
n131 Moreover, de facto racial segregation exists in many private housing markets.
n132 As noted in the HUD Reinvention, a recent General Accounting Office study of
the Section 8 certifi- cate program found that
"55 percent of white recipients but only 36 percent of black recipients live in
neighborhoods where fewer than ten percent of the residents are poor."
n133 The EHAP study found that minority participants had lower success rates for
securing housing as compared with white applicants.
[*256] When minorities obtain housing, they often are forced to pay a larger percent-
age of their income for it than whites.
Anecdotal evidence also supports findings of discrimination. One prospective
tenant who attempted to move to a nicer neighborhood explains that she was told
by a landlord
"that I wouldn't be happy because there were no other colored."
n136 When she inspected the unit, she noted that a white neighbor said,
"We can't have colored no more."
n137 The HUD Reinvention, which would deregulate public and assisted housing and
allow landlords to refuse certifi- cates, would give such providers more leeway
to discriminate against certifi- cate holders, as is the practice in private
C. Fighting the Concentration of Poverty
Proponents of demand-side assistance support voucherization as a way to
deconcentrate the number of poor families living in high-poverty neighbor-
hoods by enabling certificate recipients to buy into higher income areas.
n138 PHAs often are located in inner-city neighborhoods where the concentration of
poverty is highest.
n139 Currently, the median income of families living in public housing is less than
$ 6,500, and three-quarters of non-elderly households have incomes below the
poverty level. One-half of these families receive public assistance, and over
ninety percent of the families living in some of the large PHAs receive
n140 The degree of poverty among public housing resi- dents has increased; between
1980 and 1995, tenants' income decreased from thirty-three to seventeen percent
of median income.
n141 The incidence of pov-
[*257] erty is tied closely to race; two-thirds of non-elderly families living in
public housing are black and approximately one-fifth are Hispanic.
n142 Overall, thirty- nine percent of subsidized units are occupied by minorities.
n143 Given the conse- quences of the concentration poverty which, according to
William Julius Wil- son, results in problems of a unique kind and magnitude,
n144 eradication of densely populated, high-poverty areas is a laudable goal.
Nevertheless, the HUD Reinvention ignores the consequences of a massive demand
shock to stable housing markets, as well as the potential significance of
neighborhood dynamics in the context of current patterns of develop- ment.
n146 HUD and other demand-side proponents presume that affordable housing is widely
dispersed across geographic areas, that private landlords will be willing to
participate in the certificate program, that certificate holders will be able
to find housing in a variety of neighborhoods, and that neighborhoods exist in
a state of equilibrium such that they are able to withstand significant shocks
without becoming stabilized. HUD's failure to fully consider the impact the
Reinvention would have on certificate-holding and low-income,
non-certificate-holding households has resulted in an incomplete cost benefit
analysis. Such lack of proper reflection could lead to significant
unanticipated costs which, if considered initially, might alter HUD's
1. Consequences of a Massive Demand Shock
The HUD Reinvention would lead to a massive and sudden demand shock to the
current private sector housing market. This demand shock would result from the
n149 the current public housing stock, and the choice of certificate holders to use
their certificates in what is now the private housing market.
HUD exaggerates the degree of flexibility that certificates would provide to
their holders. Although HUD assumes that certificate holders would be able to
use their certificates to move to a variety of different neighborhoods,
n150 in real- ity they would be forced to concentrate in lower- and
lower-middle-income areas, or might be unable to move at all. First,
certificates would be pegged to fortieth percentile FMRs, and because of their
low-income status, certificate holders would find it difficult to spend much
more than the value of their cer- tificates on housing. Poor tenants would be
priced out of tight markets in which landlords can charge rents well above
n151 According to the ring theory and the filtering model,
n152 this dynamic would occur most often in middle-income neighborhoods where
certificate-holders could be consistently outbid for desirable housing.
n153 Thus, certificate-holders would be limited to areas where housing rents for
certificate rates, neighborhoods which are predominantly lower- and
Second, even if some housing in middle-income neighborhoods rented for
fortieth percentile FMRs, certificate-holders, many of whom are members of
minority groups, would not obtain such housing due to racial discrimination by
the predominantly white population of residents and landlords.
n154 Studies have demonstrated that increases in income, and thus spending power,
among blacks have failed to alter levels of segregation in housing markets.
[*259] residents and landlords who do not discriminate against certificate-holders
because of their race might discriminate against them because of their low-
n156 While some landlords in middle-income areas might rent to a small number of
low-income certificate holders whose entry into the neighbor- hood would be
barely perceptible, other landlords would refuse to rent to cer- tificate
holders fearing negative neighborhood effects.
n157 Fourth, even if certifi- cate holders were not the victims of racial or
socioeconomic discrimination, landlords might discriminate against them on the
basis of their certificate- holding status in order to avoid entanglements with
n158 Such discrimination would occur most often in desirable neighborhoods where
landlords have the highest degree of bargaining power.
The significant influx of certificate-holders into now private sector housing
in lower and lower-middle-income neighborhoods would cause prices in such
neighborhoods to rise.
n159 Current residents in these neighborhoods are low- income families, many of
whom qualify for but do not receive public or assisted housing.
n160 Since housing assistance is not an entitlement program, such families may not
be able to obtain certificates under the HUD Reinven- tion. Although the HUD
Reinvention requires PHAs to issue certificates to current residents of public
and assisted housing and to other low-income households, PHAs may also reserve
half of all incremental and turnover certif- icates for working or work-ready
n161 While higher rents resulting from increased demand would affect both
certificate and non-certificate-holders, certificate-holders would be best able
to bear price increases because they receive rent subsidies.
Some non-certificate-holders would stay in their current neighborhoods and pay
higher rents for their same apartments, representing a net transfer of wealth
from non-certificate-holders to landlords.
n163 Others would leave their current residences due to price increases, and move
to vacated units in the reduced-in-size former public housing market, bidding
competitively for apart- ments with the certificate-holders who remain.
n164 Because such families would lack certificates, they would pay the full cost of
what the market would bear
[*260] for the units, possibly more than thirty percent of their adjusted incomes.
n165 Given their low-income status, non-certificate-holders would be forced to move
into the cheapest and worst projects as certificate-holders move out. Further-
more, because landlords who do not accept certificates would not be required to
comply with HUD's HQS, they could refuse to maintain these apartments properly.
n166 Adverse neighborhood effects would be exacerbated if decrepit buildings were
abandoned. Depending on the availability of units, non-certifi- cate-holders
could be priced out of housing markets altogether,
n167 ultimately becoming homeless.
2. The Impact of Neighborhood Dynamics
The study of neighborhood dynamics is rooted in a discourse that seeks to
explain systemic change.
n168 This discourse begins with the idea that systems consist of a set of factors
which are causally interrelated in complex ways.
n169 Such sets of factors are characterized by a dynamic of
"circular causation" in which change in one factor affects other factors.
n170 Most systems exist in a state of equilibrium, such that disturbances, or
inputs, to the systems do not significantly alter them.
n171 When systems undergo a significant enough shock, however, they can become
destabilized, destroying the former equilibrium.
n172 When a system becomes destabilized, a change in one factor in a certain direc-
tion causes other factors in the system to change in the same direction. This
phenomenon creates a cumulative
"feedback loop" whereby the latter factors that are affected, in turn, impact the initially
affected factors, causing them to change again in the same direction, thereby
"spiraling" effect, mag-
[*261] nifying the initial change.
n173 This cumulative process can continue indefinitely, and can lead to undesirable
Neighborhood dynamics draws on this analysis to explain changes in hous- ing
markets, including deterioration and
gentrification. In particular, neigh- borhoods may be susceptible to changes due to factors
related to stigma or status,
n175 such as the opinions and future expectations of residents, real estate agents
and banks as to the character of a given neighborhood,
n176 as well as media attention.
n177 Moreover, individual neighborhood systems are not iso- lated; the
destabilization of one neighborhood may contribute to the destabilization of
The HUD Reinvention fails to consider the effect that the demand shock that
voucherization would have on existing neighborhood equilibria. While it is
impossible to predict precisely what effect such a demand shock would have,
experience indicates that such a significant market input is likely to cause
widespread destabilization of housing markets. This section explores three pos-
sible scenarios that could result: downward spirals characterized by neighbor-
hood deterioration leading to new areas of poverty concentration and abandon-
ment; upward spirals characterized by
gentrification and displacement of low-
[*262] income tenants; and the interaction between downward and upward spirals.
These possibilities are far from exhaustive. Rather, they are meant to suggest
that by failing to take into account the destabilizing effects of
voucherization on neighborhoods, HUD has neglected to account for a significant
i. Scenario One: Downward Spirals and New Areas of Poverty Concentration
If landlords in
"healthy" middle- and lower-middle-income neighborhoods agreed to rent to certificate
holders, the arrival of a handful of families might be accommodated by the
neighborhoods. However, the entry of a critical num- ber of certificate
recipients could cause one current residential household to move out, based on
its expectations about future neighborhood changes that would be generated by
the arrival of certificate holding tenants. Given circular effects,
n179 the departure of this family would, in turn, influence other original
residents to leave. That is, the concomitant decrease in the number of wealth-
ier residents and influx of low-income and/or minority residents might cause
wealthier residents to fear that neighborhood property values will decrease in
the future, and to cut their losses by relocating.
n180 A cumulative feedback loop would be established with the result that, as more
certificate holders moved in and original neighbors left, more of the original
residents would flee to wealth- ier neighborhoods. In turn, this would ease the
entrance of new certificate holders into the area. Ultimately, these dynamics
would destabilize the area, causing the neighborhood to enter a downward
n181 While units would be required to meet HUD's HQS, in practice landlords might
violate this obliga- tion, forcing certificate recipients, in time, to live in
poor quality apart- ments.
n182 Thus, in the end, predominantly low-income certificate holders would be left
in a newly transformed, poverty-concentrated neighborhood.
ii. Scenario Two: Upward Spirals and
While tenants who live in decent quality public and assisted housing in good
neighborhoods may wish to keep their apartments, this might not be possible if
these neighborhoods become gentrified. Depending on the desirability of the
units and the number of wealthier families who want to move in, landlords might
be able to increase rents in excess of fortieth percentile FMRs.
n183 As a result, certificate holders would be forced to commit greater portions of
their incomes and, therefore, sacrifice other necessities, such as food and
[*263] remain in their apartments.
n184 Given cumulative change and feedback effects, as greater numbers of
higher-income families rent or purchase apartments, the neighborhood would
enter an upwards spiral, and begin the process of gentrifi- cation.
n185 Indeed, the presence of the first higher-income pioneers and early settlers
could cause mainstream households and, ultimately, stragglers,
n186 to move in, permitting landlords to steadily raise their rents, and forcing
more certificate holders out of their current apartments.
n187 Such neighborhoods could eventually reach a state of complete
gentrification, thereby transforming once lower-income areas into middle-income enclaves.
iii. Scenario Three: Interaction Among Downward and Upward Spiral- ing Markets
It is possible that downward and upward spiraling markets would interact
across housing markets, providing for a more complex set of neighborhood
dynamics. For example, certificate holders who are priced out of gentrifying
neighborhoods would be forced to reduce their consumption of housing by
competing for less expensive apartments in more densely populated areas.
Depending on neighborhood dynamics, these households may be required to search
for housing in downward spiraling areas with the other certificate hold- ers
described above. Given their low-income status, they too would ultimately end
up being forced to rent lower quality apartments located in areas of high
In sum, taking into consideration neighborhood stratification by class and
income, and incorporating housing market dynamics, the HUD Reinvention could
fail in its objective of easing the concentration of poverty. While the impact
of voucherization is idiosyncratic to individual housing markets, it is
possible that a universal demand-side program would create new high poverty
neighborhoods and transform lower and lower-middle income areas into mid-
dle-income neighborhoods not affordable by certificate holders. The implica-
tion that voucherization would have different effects on different neighbor-
hoods further supports the adoption of a program of targeted, rather than
[*264] universal, voucherization.
While HUD and other demand-side enthusiasts claim that converting cur- rent
federal housing subsidies to certificates would result in a less costly pro-
gram that would allow for tenant choice and provide for the de-concentration of
poverty, in practice these objectives would not be achieved. Although certifi-
cate proponents aspire to commendable goals, as the above analysis demon-
strates, voucherization is not the correct method for achieving them. Indeed,
contrary to HUD's Office of Policy Development and Research's cost study and
the theoretical support for demand-side assistance, certificates, as pro- vided
for in the HUD Reinvention, would be more expensive than current HUD subsidies.
Certificate holders would often be prevented from exercising free choice in
where to live due to constraints on the availability of affordable apartments
and landlord discrimination. In addition, certificate holders would be forced
to use their certificates in lower- and lower-middle-income neighbor- hoods,
raising housing prices in such areas and adversely affecting low-income,
non-certificate-holders. Moreover, given neighborhood effects and circular cau-
sation, the HUD Reinvention might create housing market instability, result-
ing in the segregation of certificate-holders in new areas of poverty
concentra- tion, and the forced departure of certificate-holders from
Although universal voucherization is not the correct solution, it is important
to recognize that HUD programs are in need of reform. Especially in light of
impending budget cuts,
n188 it is critical that HUD increase its efficiency in pro- viding housing
assistance to low-income families. HUD should embark on a more careful
reassessment of its current programs, and develop a tightly-rea- soned program
that will enable the Department to effectively meet its praise- worthy goals.
As this article demonstrates, a flexible policy, incorporating sup- ply- and
demand-side programs both within and among public and assisted housing
developments, would be most effective in addressing differing neigh- borhood
conditions and dynamics. Indeed, rather than subjecting all federally assisted
housing to the rigor of the private market system, HUD's investment in public
and assisted housing should be strategically sustained.
n1 Telephone Interview with Dan Wuenschel, Executive Director, Cambridge Hous-
ing Authority (Apr. 27, 1995) [hereinafter Wuenschel Interview].
n2 See Catherine S. Manegold, Secretary Proposes Reshaping HUD to Save It, N.Y.
TIMES, Mar. 20, 1995, at 17; PUBLIC HOUS. AUTHS. DIRECTORS ASS'N, REINVENTION:
THE PHADA PLAN AND THE HUD MYTH 17 (1995) [hereinafter PHADA PLAN].
n3 See, e.g., HUD Dud, THE NEW REPUBLIC, Jan. 2, 1995, at 8.
n4 U.S. DEP'T OF HOUS.
& URBAN DEV., REINVENTION BLUEPRINT 3 (Dec. 1994).
n5 U.S. DEP'T OF HOUS.
& URBAN DEV., HUD REINVENTION: FROM BLUEPRINT TO ACTION 11 (Mar. 1995)
[hereinafter HUD REINVENTION].
n6 See id.; REINVENTION BLUEPRINT, supra note 4.
n7 See Stephen B. Kinnaird, Abandon HOPE, But Not Privatization,
103 YALE L.J. 961, 966 (1994) ("Public housing is far more diverse than the crime-ridden, deteriorat- ing,
urban high-rise projects that haunt the popular imagination."); PHADA PLAN, supra note 2, at 9 (noting that the worst problems are found in
less than three percent of public housing authorities).
n8 See Kinnaird, supra note 7, at 966.
n9 See Michael H. Schill, Distressed Public Housing: Where Do We Go from Here?,
60 U. CHI. L. REV. 497, 501-02 (1993); Fred Fuchs, Introduction to HUD Conven- tional Public Housing, Section 8
Existing Housing, Voucher, and Subsidized Housing Programs, Part I:
Conventional Public Housing,
25 CLEARINGHOUSE REV. 782 (1991).
n10 PHAs contract with HUD to provide low- and moderate-income public housing.
See infra notes 23-24 and accompaying text.
n11 Interview with William C. Apgar, Jr., Executive Director, Joint Center for
Hous- ing Studies, John F. Kennedy School of Government, Harvard University, in
Cam- bridge, Mass. (Apr. 18, 1995).
n12 See Kinnaird, supra note 7, at 969.
24 C.F.R. section 882.109 (1995).
n14 See Rochelle L. Stanfield, Slum Fund, NAT'L J., Sept. 3, 1994 at 2017, 2017
(describing HUD's Section 8 project-based assisted Edgewood Terrace I in
Bethesda, Maryland as follows:
"Half the windows are boarded. Weeds choke the grounds. Graf- fiti and trash are
everywhere. A foul odor is discernible.").
n15 See Kinnaird, supra note 7, at 968. Note that this figure only represents the
cost of rehabilitating PHAs, and not HUD-assisted buildings.
n16 See Schill, supra note 9, at 501.
n17 See HUD REINVENTION, supra note 5, at 1, 6.
n18 In this article
"voucherization" refers to assisting families with either housing cer- tificates or vouchers.
n19 See Ronald D. Utt, Improving America's Housing by Shutting down HUD, Heri-
tage Found. Rep., Mar. 16, 1995, available in LEXIS, Nexis Library, NWLTRS file
n20 See id. at *9.
n21 See WILLIAM JULIUS WILSON, THE TRULY DISADVANTAGED: THE INNER CITY, THE
UNDERCLASS, AND PUBLIC POLICY 58 (1987) (noting that in 1980, 68% ofpoor
whites, 15% of poor blacks and 20% of poor Hispanics lived in non-poverty areas
of the five largest U.S. cities, and that only 7% of poor whites lived in
extreme poverty areas, as compared with 39% of poor blacks, and 32% of poor
42 U.S.C. sections 1437-1437aaa (Supp. V 1993).
42 U.S.C. section 1437b(a).
n24 See 24 C.F.R. section 990 (1995) (establishing standards and policies for the
distribu- tion of operating subsidies). For a more detailed analysis of the
public housing pro- gram, see Schill, supra note 9, at 501-02; Fuchs, supra
note 9, at 782.
12 U.S.C. section 17151(d)(3) (1994).
n26 Id. section 1715z-1.
12 U.S.C. section 1701q (1994).
n28 Id. section 1701s.
24 C.F.R. section 215.10 (1995).
n30 24 C.F.R. section 880 (1995).
n31 24 C.F.R. section 881 (1995).
24 C.F.R. sections 886.301-.338 (1995).
n33 See Fred Fuchs, Introduction to HUD Conventional Public Housing, Section 8
Existing Housing, Voucher, and Subsidized Housing Programs, Part II: Section 8
Existing Housing, Voucher, and Subsidized Housing,
25 CLEARINGHOUSE REV. 990, 996-97 (1991).
42 U.S.C. section 1437f (1988
& Supp. V 1993).
n35 Section 8 certificates are portable within the state. See Fuchs, supra note
33, at 990.
n36 FMRs are set on a county basis in areas other than New England, and on a
comparable basis within New England. FMRs under the Section 8 program are cur-
rently set at the 45th percentile of newly rented properties in the area.
Telephone Inter- view with Gerald Benoit, Director of Rental Operations,
Section 8 Housing, U.S. Department of Urban Development (Apr. 24, 1995)
[hereinafter Benoit Interview].
42 U.S.C. section 1437f(o).
n38 See Fuchs, supra note 33, at 995.
n39 The purpose of the COF would be to stimulate revitalization. The fund would
consolidate all current HUD community and economic development programs,
includ- ing Community Development Block Grants and Economic Development
Initiative Grants. The COF would also include a new job creation performance
program. See HUD REINVENTION, supra note 5, at 11, 17. The AHF would support
the production and rehabilitation of affordable housing, and would include all
current HUD programs for the development of low- and moderate-income housing.
See id. at 11.
n40 See id. at 11, 28, 33; Stanfield, supra note 14, at 2018.
n41 Benoit Interview, supra note 36.
n42 See HUD REINVENTION, supra note 5, at 28.
n43 Benoit Interview, supra note 36.
n44 Stuart M. Butler, Only Vouchers Can Stop
"HUDscams', L.A. DAILY J., Aug. 15, 1989, at 6.
n45 See Council of Large Public Hous. Auths., Every Time You Turn Around, Some-
one from HUD Insists That Public Housing Costs More Than Section 8. Is This
Statement True?, 15 CLPHA NEWS 5 (Apr. 13, 1995) [hereinafter CLPHA NEWS].
n46 See, e.g., Utt, supra note 19, at *1.
n47 See Kinnaird, supra note 7, at 965.
n48 See id. at 985 (noting that in HUD's Housing Supply Experiment undertaken in
the 1970s, 85% of program dollars went to serve low-income voucher recipients,
as compared with 34% of public housing dollars that went to tenants under
supply-side assistance); Butler, supra note 44, at 6.
n49 See Butler, supra note 44, at 6. Butler notes that under the current system of
"developers and owners reap enormous profits yet face virtually no financial
risk, while loan and rent guarantees encourage inflated housing costs. With a
captive market of tenants, the landlord has little incentive to keep his
property well maintained and attractive." Id. See also Schill, supra note 9, at 535 (noting that
"the government is frequently considered to be an inefficient provider of goods
and services, for only private entrepreneurs will be driven by the profit
motive to minimize costs and maximize productivity").
n50 See Kinnaird, supra note 7, at 982.
n51 See id. at 972 (noting that
""a 1982 study done for HUD found that it cost 40% more per-unit to build
conventional public housing than unsubsidized FHA housing"). But see id. at 984 n.164 (describing the Supply Side Experiment, and noting
"predictably, vouchers did not stimulate new housing production").
n52 See id. at 972 (describing a second 1982 HUD study that found PHA per-unit
operating costs were 61% higher than private sector operating costs); Schill,
supra note 9, at 536-37 (noting that statistical studies have shown that the
operating costs of public housing are approximately 30% higher than the
operating costs of similar pri- vate sector housing).
n53 See Kinnaird, supra note 7, at 982.
n54 See Schill, supra note 9, at 535.
n55 See Kinnaird, supra note 7, at 971 (arguing that
"the public housing official does not have to be directly responsive to tenant
needs because (a) the public housing tenant cannot transfer his subsidy to
another supplier and (b) PHA revenues depend largely on budgetary allocations,
and only partly on tenant rents").
n56 See REPORT OF THE PRESIDENT'S COMM'N ON PRIVATIZATION, PRIVATIZATION: TOWARD A
MORE EFFECTIVE GOVERNMENT 11 (1988) (comparing the 20-year cost of
per-household household-based assistance programs -- $ 27,892 for the Housing
Voucher Program and $ 27,955 for the Section 8 Existing Housing Program -- with
similar costs for project-based assistance programs -- $ 35,210 for the
Farmer's Home Section 515 Program, $ 53,500 for the Rental Housing Development
Grant Program, $ 53,575 for the Section 8/292 Housing Program, and $ 69,863 for
public housing); Butler, supra note 44, at 6; Utt, supra note 19, at *7 n.5.
n57 See, e.g., Schill, supra note 9, at 538.
n58 See Manegold, supra note 2, at 17.
n59 See Department of Veterans Affairs and Hous. and Urban Dev. and Indep. Agen-
cies Appropriations, Fiscal Year 1993: Hearings Before Comm. on Appropriations,
102d Cong., 2d Sess. 310 (1992); Kinnaird, supra note 7, at 988.
n60 See CLPHA NEWS, supra note 45, at 5.
n61 See OFFICE OF POLICY DEV.
& RESEARCH, U.S. DEP'T OF HOUS.
& URBAN DEV., ISSUE BRIEF, HUD REINVENTION: FROM BLUEPRINT TO ACTION, WILL IT
COST MORE TO REPLACE PUBLIC HOUSING WITH CERTIFICATES? 1 (1995) [hereinafter
REPLACE- MENT COSTS]. Michael Stegman, the Assistant Secretary of the Office of
Policy Devel- opment and Research, has apparently retreated from his former
endorsement of expanding the public housing production program. See Michael A.
Stegman, The Role of Public Housing in a Revitalized National Housing Policy,
in BUILDING FOUNDA- TIONS: HOUSING AND FEDERAL POLICY 333, 334 (Denise
& Langley C.Keyes eds., 1990).
n62 NAHRO estimates the cost of current subsidies to be $ 425 and the cost of
voucherization to be $ 471 PUM. See NATIONAL ASS'N OF HOUS.
& REDEV. OFFICIALS, COMPARING PUBLIC HOUSING AND SECTION 8 COSTS 1 (1995). See
also Letter from Richard Y. Nelson, Executive Director, National Association of
Housing and Redevel- opment Officials, to Michael Stegman, Assistant Secretary
for Policy Development and Research, U.S. Department of Housing and Urban
Development 2, 6 (Apr. 7, 1995) (on file with author) [hereinafter Nelson
Letter]. CLPHA assesses the current subsi- dized cost to be $ 377 PUM, and
compares this expense with HUD's estimated $ 440 PUM under voucherization. See
CLPHA NEWS, supra note 45, at 5.
n63 See PHADA PLAN, supra note 2, at 25-26. The states PHADA examined were
Alabama, California, Florida, Maryland, Massachusetts, Michigan, Missouri, New
York, Ohio, and Tennessee. While cost differentials varied from 104% in
Missouri to virtual equality in Ohio, average subsidy costs were $ 343 and
average voucherization costs were $ 469 PUM.
n64 See Memorandum from Jerry Jones, Executive Director, Association for Commu-
nity Organizations for Reform Now, to Michael Stegman, Assistant Secretary for
Pol- icy Development and Research, U.S. Department of Housing and Urban
Development 2 (Feb. 12, 1995) (on file with author). PHADA conducted an
examination of the following PHAs: Little Rock, Ark.; Oakland, Cal.;
Washington, D.C.; Miami/Dade, Fla.; Atlanta, Ga.; New Orleans, La.; Chicago,
Ill.; Baton Rouge, La.; Baltimore, Md.; Boston, Mass.; Minneapolis, Minn.; St.
Paul, Minn.; St. Louis, Mo.; Kansas City, Kan.; New York, N.Y.; Dallas, Tex.;
Houston, Tex.; and Milwaukee, Wis. Average current public housing expenditures
were found to be $ 491 and average voucherization costs to be $ 566 PUM.
n65 See REPLACEMENT COSTS, supra note 61, at 4.
n66 See id.
n67 See Kinnaird, supra note 7, at 989.
n68 See REPLACEMENT COSTS, supra note 61, at 4.
n69 See HUD REINVENTION, supra note 5, at 39. The HUD Reinvention proposal would
be implemented according to a three-stage process, and all subsidized housing
would be converted to voucherization by the year 2002. See id. at 43.
n70 Telephone Interview with Carolyn Samuels, Publisher, CLPHA NEWS (Apr. 26,
1995) [hereinafter Samuels Interview].
n71 See infra notes 151-59 and 180-83 and accompanying text.
42 U.S.C. section 1437l(b) (1988).
42 U.S.C. section 905.102 (1992).
n74 Telephone Interview with Peter Henderson, Housing and Community Develop- ment
Officer, National Association of Housing and Redevelopment Officials (Apr. 19,
1995) [hereinafter Henderson Interview]; Telephone Interview with Wayne
Sherwood, Council of Large Public Housing Officials (Apr. 20, 1995); Samuels
Interview, supra note 70.
n75 Nelson Letter, supra note 62, at 2.
n76 See supra note 15 and accompanying text. The $ 16.65 billion would cover only
mandatory backlog modernization needs. See Department of Veterans Affairs and
Hous. and Urban Dev. and Indep. Agencies of the House Comm. on Appropriations,
103d Cong., 1st Sess. 85 (1993) (statement of Michael B. Janis, General Deputy
Assis- tant Secretary, Office of Public and Indian Housing); Kinnaird, supra
note 7, at 968.
57 Fed. Reg. 27330 (1992).
n78 CIAP, CGP, and MROP were initiated in 1988, 1991, and 1992, respectively. See
Schill, supra note 9, at 522.
n79 See Nelson Letter, supra note 62, at 2 ("A primary reason for the backlog in modernization today is that federal funding
prior to the Comprehensive Grant Program . . . had been inadequate for meeting
accrual needs. Thus the cost of public housing was kept low in the past and now
appears to be growing because of this change.").
n80 CLPHA increased the Abt Associates/ICF costs by one-third to account for
infla- tion and the continued aging of public housing, obtaining an annual
maintenance expenditure of $ 2.8 billion, or $ 166 PUM. See CLPHA NEWS, supra
note 45, at 5.
n81 See id.
n82 See Nelson Letter, supra note 62, at 2.
n83 With a subsidy cost of $ 440 and a certificate cost of $ 481 PUM, certificate
costs would have to decrease by $ 41 PUM, or $ 492 per year, to equal subsidy
costs. Multi- plying $ 492 per year by the 1.142 million units of occupied PFS
housing, moderniza- tion expenses would need to decrease by $ 562 million, a
15% reduction from current modernization costs. Adding $ 20 PUM to certificate
costs for the PH-DEP program would require only a $ 288 million dollar, or
7.8%, reduction in modernization expendi- tures for a certificate program to
equal PFS subsidy costs.
n84 See NATIONAL HOUS. LAW PROJECT, SEVERE RESCISSIONS OF HUD FUNDS 1 (Feb. 1995)
(reporting that the House Appropriations Subcommittee covering HUD approved a
rescission bill that would reduce HUD's current budget of $ 26 billion by $ 7
billion, or by 27%, and noting that
"even if the Senate or the President were to reject [the bill], the handwriting
for the [fiscal year] 1996 budget deliberations is on the wall").
n85 Telephone Interview with Peter Henderson, Housing and Community Develop- ment
Officer, National Association of Housing and Redevelopment Officials (May 16,
1995) [hereinafter Henderson Interview II].
n86 Id. Henderson notes that many are opposed to the one-size-fits-all programs
cur- rently administered by HUD, and that the HUD Reinvention merely calls for
the replacement of universal HUD programs with a new one-size-fits-all model.
Id.; see also Wuenschel Interview, supra note 1.
n87 JEROME ROTHENBERG, ET AL., THE MAZE OF URBAN HOUSING MARKETS: THE- ORY,
EVIDENCE, AND POLICY 355 (1991).
n88 PHADA PLAN, supra note 2, at 9.
n89 See HUD Dud, supra note 3, at 8 ("People who escape the inner city stay in school and keep jobs at higher rates
than people who remain in poor neighborhoods.").
n90 See HUD REINVENTION, supra note 5, at 36.
n91 James E. Rosenbaum et al., Can the Kerner Commission's Housing Strategy
Improve Employment, Education, and Social Integration for Low-Income Blacks?,
71 N.C. L. REV. 1519, 1522 (1993). The Gautreaux Assisted Housing Program in Chicago resulted from a 1976 United
States Supreme Court decision that found that the Chi- cago Housing Authority
(CHA) had engaged in racially discriminatory practices. See
Hills v. Gautreaux, 425 U.S. 284 (1976). In response, the CHA developed a program whereby it assists current public
housing residents, as well as those on waiting lists, in finding housing in
low-poverty areas partially funded by Section 8 subsidies. See Kin- naird,
supra note 7, at 986. In spite of this search assistance, in the case of the
Gau- treaux program, only half of participating families were able to find
housing in one of the six outlying counties of Chicago. See PHADA PLAN, supra
note 2, at 29. As described in section B.2, infra, discrimination may have
played a role.
n92 See Laurie Abraham, Foes Kill Housing Plan Funds: Low-Income Renters Call Move
Racist, CHI. TRIB., Dec. 15, 1994, at 33.
n93 See Joann S. Lublin, Uncertain Solution: Vouchers for Housing Help Some of the
Poor, Fail to Benefit Others, WALL ST. J., Nov. 19, 1986, at 23.
n94 See Department of Veterans Affairs and Hous. and Urban Dev. and Indep. Agen-
cies Appropriations, Fiscal Year 1992: Hearings Before the Senate Comm. on
Appro- priations, 102d Cong., 1st Sess. 523 (1991).
n95 See Schill, supra note 9, at 532.
n96 See HUD REINVENTION, supra note 5, at 36.
n97 See ROTHENBERG ET AL., supra note 87, at 354; Wuenschel Interview, supra note
1. Landlords of rent-controlled buildings cannot raise their rents; thus the
market for rent-controlled units is usually tight. Section 8 certificates,
which exempt landlords of rent-controlled buildings from adhering to
rent-control regimes, enable assisted tenants to find housing at higher rates.
n98 HUD Dud, supra note 3, at 8-9.
n99 This section does not address the problems of discrimination discussed in
Section B.2, infra.
n100 See PHADA PLAN, supra note 2, at 17, 28.
n101 See National Hous. Law Project, A Critique of HUD's Reinvention Blueprint, 25
HOUS. LAW BULLETIN 29, 33 (1995) [hereinafter NHLP Critique].
n102 See id. at 29.
n103 This concern is of particular significance, given that many families will stay
in their current apartments or move to alternative public housing under the HUD
Reinvention. See HUD REINVENTION, supra note 5, at 35.
n104 See id. at 37. The National Housing Law Project estimates that 150,000 units
would be demolished. See NATIONAL HOUS. LAW PROJECT, THE CHANGES FOR PUBLIC
HOUSING RECOMMENDED IN HUD'S REINVENTION: FROM BLUEPRINT TO ACTION 8 (Apr. 7,
n105 Currently, projects may not be demolished unless they are
"unusable for housing purposes, and no reasonable program of modifications is
feasible to return the project or portion of the project to useful life."
42 U.S.C.A. section 1437p(a)(1) (West Supp. 1993).
n106 While some support the one-for-one replacement rule since it preserves the
num- ber of public housing units in the housing stock, others note that the
rule has prevented
"crime-ridden, Stalinesque projects" from being torn down. See, e.g., HUD Dud, supra note 3, at 8. The one-for-one
replacement requirement will likely be replaced with a relocation requirement
under which PHAs would be required to assist families in find- ing alternative
housing. See CLPHA NEWS, supra note 45, at 4.
n107 Benoit Interview, supra note 36.
n108 See Wuenschel Interview, supra note 1; NHLP Critique, supra note 101, at 33.
n109 See Schill, supra note 9, at 502-03 (noting that
"the 1937 Act placed relatively low ceilings on construction expenditures to
ensure that the units would not be desira- ble enough to compete with
n110 See Lublin, supra note 93, at 1 (noting that while in San Antonio, Texas, the
20% vacancy rate forced landlords to reduce rents, enabling most voucher
holders to obtain housing within 30 days, many tenants in areas with tighter
housing markets are forced to return their vouchers unused).
n111 See infra note 185 and accompanying text.
n112 Henderson Interview II, supra note 85.
n113 Id. Henderson notes that even if housing were available, legislators still
need to consider whether landlords will be willing to participate in the
program, and where landlords who do accept certificates are located. Often,
affordable low-income housing is not dispersed widely. Id.
n114 See WALLACE F. SMITH, FILTERING AND NEIGHBORHOODS CHANGE 17-33 (1964).
According to the filtering model, wealthier households will outbid poorer
households for the same apartments. Id.
n115 Some fear that certificates will ultimately end up being lower than the 40th
per- centile. See, e.g., Wuenschel Interview, supra note 1. Peter Henderson
explains that while private landlords would receive the full value of the
voucher under the HUD Reinvention, HUD has stated that PHAs would keep less. He
further points out that PHAs will still be responsible for modernization costs,
and notes that it is unlikely that PHAs will receive enough money during the
transition period to make all needed repairs. The fact that PHAs will receive
less that the full value of the certificates will only serve to exacerbate
their current modernization problems. Henderson Interview II, supra note 85.
n116 See NATIONAL HOUS. LAW PROJECT, THE TIME OF RECKONING FOR NATIONAL HOUSING
POLICY: TO BE OR NOT TO BE 8 (Jan. 1995) (estimating that this figure averages
$ 15 per family per month).
n117 Benoit Interview, supra note 36; HUD REINVENTION, supra note 5, at 35 (stat-
ing that based on an analysis of census data for 13 metro areas, at least 40%
of all two-bedroom units rent at or below FMR). Others note that HUD carefully
selected these metro areas. See, e.g., Henderson Interview, supra note 74.
n118 Schill, supra note 9, at 527.
n119 Estella Sanchez, a warehouse worker earning $ 5.25 per hour, became the
nation's first housing voucher recipient. Ironically, she was unable to find
decent affordable housing despite high vacancy rates, and returned her voucher.
See Lublin, supra note 93, at 23.
n120 See id. (noting that one family was unable to use its voucher because
alternative housing cost an additional $ 41 per month).
n121 Lowering FMRs might also frustrate the objective of decreasing the concentra-
tion of poverty, as discussed in Section B.2. In an earlier response to the
Office of Management and the Budget's budget passback, HUD noted the following:
A reduction in FMRs to the 40th percentile will have adverse program impact.
We are still receiving complaints from many PHAs that families continue to
encounter problems in finding units below . . . FMRs at the 45th percentile.
The reduction in FMRs would greatly impede the Department's efforts to
facilitate families' moves to areas with low concentration of poverty because
these areas tend to have higher rents. Since units at the 40th percentile rent
are usually con- centrated in the lower income neighborhoods, a reduction of
the FMR would have the effect of decreasing family choices for housing outside
areas of low income concentrations and will further concentrate families in
very low income neighborhoods.
PHADA PLAN, supra note 2, at 30.
n122 See id.; Stanfield, supra note 14, at 2018; Lublin, supra note 93, at 1.
n123 See NHLP Critique, supra note 101, at 10.
n124 See Stanfield, supra note 14, at 2018 (noting that
"in 1974, Congress added Section 8 to the 1937 U.S. Housing Act to use generous
tax incentives along with direct rental subsidies to get private investors to
do what neither the housing industry nor the federal government wanted to do").
n125 See PHADA PLAN, supra note 2, at 27.
n126 See id. at 28.
n127 See Fuchs, supra note 33, at 993 (noting an exception for serious or repeated
lease violations, violations of federal, state, or local law, or other good
n128 See Lublin, supra note 93, at 23.
n129 See Stanfield, supra note 14, at 2018.
n130 See Lublin, supra note 93, at 23.
n131 See Utt, supra note 19, at *10.
n132 See Kinnaird, supra note 7, at 985.
n133 HUD REINVENTION, supra note 5, at 28.
n134 See Schill, supra note 9, at 531. An analysis of one 1990 New York City pro-
gram demonstrated that while 64% of white applicants found housing, only 23% of
blacks and 30% of Hispanics found apartments. See id. at 532. But see Meryl
& Stephen D. Kennedy, Racial/Ethnic Differences in Utilization of Section 8
Existing Rental Vouchers and Certificates, 3 HOUS. POL'Y DEBATE 463, 480 (1992)
(noting that racial differences
"essentially disappear" outside of New York City). One Pittsburgh experiment noted that 21% of black
families encountered discrimination. See Kin- naird, supra note 7, at 985-86
n135 See John C. Boger, Race and the American City: The Kerner Commission in
Retrospect -- An Introduction,
71 N.C. L. REV. 1289, 1335-36 (1993).
n136 Lublin, supra note 93, at 23.
n138 See, e.g., Utt, supra note 19, at *9; Schill, supra note 9, at 499. HUD argues
that its proposal will lead to the deconcentration of poverty. See HUD
REINVENTION, supra note 5, at 33.
n139 The number of people with incomes below the poverty level living in
inner-cities increased 29.5% -- from 1.9 to 2.4 million individuals -- between
1970 and 1980. See Paul A. Jargowsky
& Mary Jo Bane, Ghetto Poverty in the United States 1970-1980, in THE URBAN
UNDERCLASS 235, 252 (Christopher Jencks
& Paul E. Peterson, eds., 1991). The concentration of poverty has continued to
grow in urban areas since 1980. See John D. Kasarda, Inner-City Poverty and
Economic Access, in REDISCOVERING URBAN AMERICA: PERSPECTIVES ON THE 1980S 4-1,
4-39 to 4-44 (Jack Sommer
& Donald A. Hicks, eds., 1993).
n140 See Schill, supra note 9, at 517-18.
n141 See NHLP Critique, supra note 101, at 2.
n142 See Schill, supra note 9, at 518.
n143 See NATIONAL HOUS. LAW PROJECT, WHAT'S IN THE FUTURE FOR FAIR HOUS- ING? 2
(Apr. 7, 1995).
n144 See WILSON, supra note 21, at 58. According to Wilson,
"the communities of the underclass are plagued by massive joblessness, flagrant
and open lawlessness, and low-achieving schools, and therefore tend to be
avoided by outsiders. . . . The resi- dents of these areas . . . have become
increasingly socially isolated from mainstream patterns of behavior." Id.
n145 Families who were able to move to Chicago suburbs under the Gautreaux pro-
gram were positively impacted.
"Not only has the program given families the chance to upgrade their housing,
but it has had significant positive second-order effects on the educational
performance and employment status of children whose families moved to less
segregated suburbs." Kinnaird, supra note 7, at 986.
n146 Current patterns of development are explained by the ring theory and filtering
model. According to the ring theory of development, given differences in income
level and the historical development of public transportation, households are
segregated by class in concentric rings around city centers, with the poorest
living in inner city areas and the wealthiest living in the most distant
suburbs. See SAM BASS WARNER, JR., STREETCAR SUBURBS: THE PROCESS OF GROWTH IN
BOSTON, 1870-1900 at 46-64 (2nd ed. 1978). For an explanation of the filtering
model, see SMITH, supra note 114, at 17- 33.
n147 See William C. Apgar, Jr., Which Housing Policy Is Best? HOUS. POL'Y DEBATE 1,
27 ("Externality effects, be they pecuniary or real, must be accounted for in any
complete assessment of alternative subsidy approaches.").
n148 See supra Section B.1.
n149 See infra Section C.2.ii.
n150 HUD maintains that
"in most housing markets there is an adequate supply of housing available for
certificate recipients . . . units renting for less than the FMR are quite
widely dispersed geographically, accounting for at least 30 percent of the
rental stock in over two-thirds of all residential Census tracts." HUD REINVENTION, supra note 5, at 35. As discussed earlier, however, this
these data may be highly selective.
n151 See supra notes 109-110.
n152 See supra note 146. Thus, certificate holders would be outbid in neighborhoods
where residents could spend larger sums of money on housing than certificate
holders, who would be able to spend the value of their certificates plus any
additional portion of their income above the thirty percent they could afford.
n153 See Duncan Kennedy, Neither the Market Nor the State: Housing Privatization
Issues, in A FOURTH WAY? PRIVATIZATION, PROPERTY AND THE EMERGENCE OF NEW
MARKET ECONOMICS 253, 257 (Gregory S. Alexander
& Grazyna Skanska eds., 1994).
n154 See supra Section B.2.
n155 DOUGLAS S. MASSEY
& NANCY A. DENTON, AMERICAN APARTHEID: SEGREGA- TION AND THE MAKING OF THE
UNDERCLASS 85 (1993) ("No matter how much blacks earned they remained spatially separated from whites.").
n156 See supra Section B.2.
n157 See Utt, supra note 19, at *10 ("Rent subsidies . . . aim at dispersing poor fami- lies rather than
concentrating them. In practice, this has often meant injecting the problems of
public housing into previously stable neighborhoods.").
n158 See id.
n159 See Apgar, supra note 147, at 9.
n160 The number of needy families has been estimated at five million. See NATIONAL
HOUS. LAW PROJECT, THE BOTTOM LINE FOR HOUSING REFORM 1 (Feb. 1995).
n161 See HUD REINVENTION, supra note 5, at 28.
n162 See Apgar, supra note 147, at 13.
n163 Much of the rent increases would result from rent inflation as opposed to
improvements in apartments. See id.
n164 See id.
n165 See id.
n166 See Kennedy, supra note 153, at 259 (noting that
"low-income apartments will deteriorate and have a very low
"floor,' meaning that conditions in the worst-maintained units will be likely to
violate cultural norms about squalid living conditions"). See also Duncan Kennedy, The Effect of the Warranty of Habitability on Low
Income Hous- ing:
"Milking" and Class Violence, FLA. ST. U. L. REV. 485, 489 (1987) (explaining that in
deteriorating neighborhoods, landlords may find it economically rational to
"milk" their buildings by charging tenants what the market will bear and refraining
from providing building maintenance. This practice results in building
deterioration and early retirement from the housing stock.). Although most
jurisdictions formally impose a warranty of habitability, in practice such
warranties are selectively enforced.
n167 See NHLP Critique, supra note 101, at 13-14.
n168 See MICHAEL A. GOLDBERG, ON SYSTEMIC BALANCE: FLEXIBILITY AND STABIL- ITY IN
SOCIAL, ECONOMIC, AND ENVIRONMENTAL SYSTEMS 12, 14, 15, 18 (1989) (applying
this discourse to grasslands, pesticide spraying, urban transportation and rent
control); GUNNAR MYRDAL, AN APPROACH TO THE ASIAN DRAMA: METHODOLOGICAL AND
THEORETICAL 1859 (1970) (explaining dynamics of development in South Asian
n169 See MYRDAL, supra note 168, at 1846.
n170 See id.
n171 See GOLDBERG, supra note 168, at 22.
n172 See id.
n173 See MYRDAL, supra note 168, at 1847.
n174 See GOLDBERG, supra note 168, at 19. For this reason, policy analysts often
tend to disfavor top-down, wide scale initiatives, in favor of more localized
and gradual approaches to change. Goldberg states:
Future decisions should be made on a smaller scale and implemented more
slowly. Smaller and slower interventions permit the systems, upon which these
distur- bances have been imposed, to evolve a systemic response which honors
the com- plexity and stability of the system itself.
They are top-down system-wide responses, often aimed at surprisingly specific
and localized problems. Better targeted and more localized responses would
appear to be in order. Such responses are less likely to disturb the workings
of entire systems and can avoid the simplified and ultimately counterproductive
See id. at 19, 24.
n175 See ROLF GOETZE, UNDERSTANDING NEIGHBORHOOD CHANGE 31 (1979) ("The stigma or status conferred on living in a particular neighborhood distorts
how resources are allocated. People are now willing to pay much more than
average to live in what is considered the right neighborhoods.").
n176 See id. at 11-12 (describing how a young couple decides against buying a house
in a neighborhood when a current resident, real estate agent and bank have low
expec- tations about the neighborhood's future, and how the same couple wishes
to buy in the same neighborhood eight years later when public opinion is highly
n177 See id. at 60-61 (explaining how the deteriorating neighborhood of Jamaica
Plain in Boston, Massachusetts, was transformed by a highly positive television
special and advertising by the Boston Redevelopment Authority. This media
attention resulted in reinvestment in the neighborhood and upgrading of the
n178 See Peter Marcuse,
Gentrification, Abandonment, and Displacement: Connec- tions, Causes, and Policy Responses in
New York City, 28 J. URB.
& CONTEMP. LAW 195 (1985) (discussing the relationship of
gentrification and abandonment in New York City).
n179 See MYRDAL, supra note 168, at 1847.
n180 See M. Edel, Filtering in a Private Housing Market, in READINGS IN URBAN
ECONOMICS 204-15 (M. Edel
& J. Rothenberg eds., 1972).
n181 See GOETZE, supra note 175, at 95-100.
n182 See NHLP Critique, supra note 101, at 3 (noting that such activities occur in
n183 See Lawrence K. Kolodney, Eviction Free Zones: The Economics of Legal Brico-
lage in the Fight Against Displacement, FORDHAM URB. L.J. 507, 512 (1991).
n184 See Kennedy, supra note 153, at 258.
Gentrification involves the process by which low-income residents are bought or forced out by
upper-income families who are able to outbid them. See id. at 260; GOETZE,
supra note 175, at 100-03. Peter Marcuse maintains that:
Gentrification occurs when new residents -- who disproportionately are young, white,
professional, technical, and managerial workers with higher education and
income levels -- replace older residents -- who disproportionately are
low-income, working-class and poor, minority and ethnic group members, and
elderly -- from older and previously deteriorated inner-city housing in a
spatially concentrated manner, that is, to a degree differing substantially
from the general level of change in the community or region as a whole.
Marcuse, supra note 178, at 198-99.
n186 See GOETZE, supra note 175, at 101-02.
n187 See Kolodney, supra note 183, at 511.
n188 See NATIONAL HOUS. LAW PROJECT, supra note 84.
Prepared: January 24, 2003 - 5:02:29 PM
Edited and Updated, January 25, 2003
Kristen A. Stelljes