Neighborhood Stabilization Program
HUD and Grantees Are Taking Actions to Ensure Program Compliance but Data on Program Outputs Could be Improved Gao ID: GAO-11-48 December 17, 2010Congress created the Neighborhood Stabilization Program (NSP) to help reduce the number of foreclosed and abandoned properties and restore depressed local housing markets. The Housing and Economic Recovery Act of 2008 (HERA) authorized the program's first round (NSP 1), providing $3.92 billion in grant funds to states and local governments. The Department of Housing and Urban Development (HUD) administers the program. HERA mandated that GAO report on whether grantees were using NSP 1 funds in accordance with the act's criteria. For this mandate, GAO examined (1) grantees' progress and challenges in meeting NSP 1 obligation and income-targeting requirements, (2) HUD's actions to mitigate program risks and ensure grantees' compliance with key NSP 1 requirements, and (3) HUD's efforts to collect program data and assess program performance. To address these objectives, GAO analyzed HUD data and the information system used for NSP 1; interviewed HUD officials and representatives of NSP 1 grantees; analyzed HUD's internal control processes; and conducted limited tests of 8 grantees' compliance with key NSP 1 requirements.
According to HUD data, the vast majority of the 309 NSP 1 grantees obligated their funds within the required 18-month time frame. As a result, over 99 percent of NSP 1 funds were obligated as of early October 2010. Also, consistent with HERA criteria, most grantees obligated at least 25 percent of their funds for housing for low-income households. Some grantees with whom GAO spoke modified their NSP 1 strategies to meet obligation deadlines and overcome other challenges such as competition from private investors in acquiring foreclosed and abandoned homes. For instance, with HUD approval, some grantees expanded the geographic areas they were targeting. Grantees also participated in banks' "first look" programs, which give grantees the chance to bid on bank-owned properties before other potential buyers. HUD provided training, guidance, and technical assistance to grantees to address new requirements and risks posed by NSP 1. Although the grantees GAO spoke with were generally satisfied with HUD's guidance and program support, some said these efforts would have been more useful if provided earlier. HUD officials said that some of the assistance grantees found useful was delivered using funds that HUD received well after the start of NSP 1. HUD also established various internal control processes for NSP 1 and hired additional staff to help oversee the program. HUD field office staff conducted remote monitoring of all grantees and on-site monitoring for 176 grantees that HUD considered to be higher risk. Although HUD is still aggregating the results of its on-site monitoring, available results from the four field offices GAO contacted generally showed compliance with key NSP 1 requirements but also found some financial management deficiencies. HUD is requiring grantees to take corrective actions, where appropriate. GAO's review of records for 32 properties at 8 grantees found no instances of significant noncompliance with key NSP 1 requirements. To collect information on NSP 1, HUD adapted an existing financial and information system--the Disaster Recovery Grant Reporting (DRGR) system--and provided training and guidance on its use. HUD has used the system to monitor NSP 1 grantees' obligations and summarize program outputs for specific types of activities (rehabilitation and construction, demolition, and homeownership assistance). However, variation in the way grantees entered information into DRGR makes it difficult to summarize outputs for each activity (e.g., housing units acquired) without undercounting, and overall outputs (e.g., total benefiting households) without overcounting. HUD has developed a method for addressing the overcounting problem, but insufficient guidance to grantees and HUD field staff may be contributing to variation in data entry that limits the usefulness of DRGR output information. For example, HUD has not provided grantees with specific written guidance on selecting output measures, which can lead to inconsistency among grantees. HUD is planning an assessment of NSP outcomes that will focus primarily on the program's second round (NSP 2) but will also include NSP 1 in geographic areas where the two phases of the program overlap. GAO recommends that HUD provide additional guidance to NSP grantees and HUD field staff to help ensure that information on output measures is collected in HUD's data system in a more consistent manner. HUD agreed with the report's recommendations.
RecommendationsOur recommendations from this work are listed below with a Contact for more information. Status will change from "In process" to "Open," "Closed - implemented," or "Closed - not implemented" based on our follow up work.
Director: Mathew J. Scire Team: Government Accountability Office: Financial Markets and Community Investment Phone: (202) 512-6794GAO-11-48, Neighborhood Stabilization Program: HUD and Grantees Are Taking Actions to Ensure Program Compliance but Data on Program Outputs Could be Improved
This is the accessible text file for GAO report number GAO-11-48
entitled 'Neighborhood Stabilization Program: HUD and Grantees Are
Taking Actions to Ensure Program Compliance but Data on Program
Outputs Could be Improved' which was released on December 17, 2010.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as
part of a longer term project to improve GAO products' accessibility.
Every attempt has been made to maintain the structural and data
integrity of the original printed product. Accessibility features,
such as text descriptions of tables, consecutively numbered footnotes
placed at the end of the file, and the text of agency comment letters,
are provided but may not exactly duplicate the presentation or format
of the printed version. The portable document format (PDF) file is an
exact electronic replica of the printed version. We welcome your
feedback. Please E-mail your comments regarding the contents or
accessibility features of this document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
United States Government Accountability Office:
GAO:
Report to Congressional Committees:
December 2010:
Neighborhood Stabilization Program:
HUD and Grantees Are Taking Actions to Ensure Program Compliance but
Data on Program Outputs Could be Improved:
GAO-11-48:
GAO Highlights:
Highlights of GAO-11-48, a report to congressional committees.
Why GAO Did This Study:
Congress created the Neighborhood Stabilization Program (NSP) to help
reduce the number of foreclosed and abandoned properties and restore
depressed local housing markets. The Housing and Economic Recovery Act
of 2008 (HERA) authorized the program‘s first round (NSP 1), providing
$3.92 billion in grant funds to states and local governments. The
Department of Housing and Urban Development (HUD) administers the
program. HERA mandated that GAO report on whether grantees were using
NSP 1 funds in accordance with the act‘s criteria. For this mandate,
GAO examined (1) grantees‘ progress and challenges in meeting NSP 1
obligation and income-targeting requirements, (2) HUD‘s actions to
mitigate program risks and ensure grantees‘ compliance with key NSP 1
requirements, and (3) HUD‘s efforts to collect program data and assess
program performance. To address these objectives, GAO analyzed HUD
data and the information system used for NSP 1; interviewed HUD
officials and representatives of NSP 1 grantees; analyzed HUD‘s
internal control processes; and conducted limited tests of 8 grantees‘
compliance with key NSP 1 requirements.
What GAO Found:
According to HUD data, the vast majority of the 309 NSP 1 grantees
obligated their funds within the required 18-month time frame. As a
result, over 99 percent of NSP 1 funds were obligated as of early
October 2010. Also, consistent with HERA criteria, most grantees
obligated at least 25 percent of their funds for housing for low-
income households. Some grantees with whom GAO spoke modified their
NSP 1 strategies to meet obligation deadlines and overcome other
challenges such as competition from private investors in acquiring
foreclosed and abandoned homes. For instance, with HUD approval, some
grantees expanded the geographic areas they were targeting. Grantees
also participated in banks‘ ’first look“ programs, which give grantees
the chance to bid on bank-owned properties before other potential
buyers.
HUD provided training, guidance, and technical assistance to grantees
to address new requirements and risks posed by NSP 1. Although the
grantees GAO spoke with were generally satisfied with HUD‘s guidance
and program support, some said these efforts would have been more
useful if provided earlier. HUD officials said that some of the
assistance grantees found useful was delivered using funds that HUD
received well after the start of NSP 1. HUD also established various
internal control processes for NSP 1 and hired additional staff to
help oversee the program. HUD field office staff conducted remote
monitoring of all grantees and on-site monitoring for 176 grantees
that HUD considered to be higher risk. Although HUD is still
aggregating the results of its on-site monitoring, available results
from the four field offices GAO contacted generally showed compliance
with key NSP 1 requirements but also found some financial management
deficiencies. HUD is requiring grantees to take corrective actions,
where appropriate. GAO‘s review of records for 32 properties at 8
grantees found no instances of significant noncompliance with key NSP
1 requirements.
To collect information on NSP 1, HUD adapted an existing financial and
information system”the Disaster Recovery Grant Reporting (DRGR) system”
and provided training and guidance on its use. HUD has used the system
to monitor NSP 1 grantees‘ obligations and summarize program outputs
for specific types of activities (rehabilitation and construction,
demolition, and homeownership assistance). However, variation in the
way grantees entered information into DRGR makes it difficult to
summarize outputs for each activity (e.g., housing units acquired)
without undercounting, and overall outputs (e.g., total benefiting
households) without overcounting. HUD has developed a method for
addressing the overcounting problem, but insufficient guidance to
grantees and HUD field staff may be contributing to variation in data
entry that limits the usefulness of DRGR output information. For
example, HUD has not provided grantees with specific written guidance
on selecting output measures, which can lead to inconsistency among
grantees. HUD is planning an assessment of NSP outcomes that will
focus primarily on the program‘s second round (NSP 2) but will also
include NSP 1 in geographic areas where the two phases of the program
overlap.
What GAO Recommends:
GAO recommends that HUD provide additional guidance to NSP grantees
and HUD field staff to help ensure that information on output measures
is collected in HUD‘s data system in a more consistent manner. HUD
agreed with the report‘s recommendations.
View [hyperlink, http://www.gao.gov/products/GAO-11-48] or key
components. For more information, contact Mathew Scirè at (202) 512-
8678 or sciremj@gao.gov.
[End of section]
Contents:
Letter:
Background:
NSP 1 Grantees Overcame Challenges and Used a Variety of Strategies to
Meet Obligation and Income-Targeting Requirements:
HUD Has Taken Actions to Mitigate NSP 1 Program Risks through
Training, Technical Assistance, and the Establishment of Internal
Controls:
HUD Adapted an Information System for NSP 1, but Data on Program
Outputs Have Limitations:
Conclusions:
Recommendations for Executive Action:
Agency Comments and Our Evaluation:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: Comments from the Department of Housing and Urban
Development:
Appendix III: GAO Contact and Staff Acknowledgments:
Table:
Table 1: Activities Reviewed for 32 NSP 1 Properties:
Figures:
Figure 1: NSP 1 Implementation Time Line:
Figure 2: Total NSP 1 State and Local Government Allocations, by State:
Figure 3: NSP 1 Grantees' Obligations as of June 24, 2010, and as of
October 1, 2010:
Figure 4: Reported NSP 1 Grantee Obligations by Activity, as of
October 1, 2010:
Figure 5: States' Reallocation of NSP 1 Funds to Units of Local
Government That also Received Funds Directly from HUD:
Figure 6: NSP Strategies for Acquiring and/or Rehabilitating Homes:
Figure 7: Selected HUD Internal Controls for NSP 1:
Figure 8: Program Requirements and Controls We Reviewed:
Abbreviations:
AMI: area median income:
ARRA: American Recovery and Reinvestment Act of 2009:
CDBG: Community Development Block Grant:
CPD: Office of Community Planning and Development:
DRGR: Disaster Recovery Grant Reporting:
FERA: front-end risk assessment:
HERA: Housing and Economic Recovery Act of 2008:
HUD: Department of Housing and Urban Development:
NOFA: Notice of Funding Availability:
NCST: National Community Stabilization Trust:
NSP: Neighborhood Stabilization Program:
OIG: Office of the Inspector General:
PD&R: Office of Policy Development and Research:
QPR: Quarterly Performance Report:
[End of section]
United States Government Accountability Office:
Washington, DC 20548:
December 17, 2010:
The Honorable Christopher J. Dodd:
Chairman:
The Honorable Richard C. Shelby:
Ranking Member:
Committee on Banking, Housing, and Urban Affairs:
United States Senate:
The Honorable Barney Frank:
Chairman:
The Honorable Spencer Bachus:
Ranking Member:
Committee on Financial Services:
House of Representatives:
In recent years, declining home prices and weak economic conditions
have contributed to a surge in vacant and foreclosed properties that
has negatively impacted many communities across the United States.
Such properties can destabilize neighborhoods by lowering the value of
surrounding homes and attracting crime. They also can impose costs on
communities, for example by reducing property tax revenue and
straining the resources of local police and code enforcement offices.
To address these issues, Congress created the Neighborhood
Stabilization Program (NSP), which provides grants to states and local
governments to help reduce the number of foreclosed and abandoned
properties and restore depressed local housing markets. The first
phase of this program, NSP 1, was authorized by the Housing and
Economic Recovery Act of 2008 (HERA), which provided $3.92 billion in
grant funds.[Footnote 1] The program is administered by the Department
of Housing and Urban Development (HUD) through its existing Community
Development Block Grant (CDBG) program.[Footnote 2] In October 2008,
HUD allocated NSP 1 funds to 309 CDBG grantees, including all 50
states.[Footnote 3] Under NSP 1, state grantees may use funds directly
or reallocate them to local governments within their states.
HERA defines a variety of "eligible uses" of NSP 1 funds to allow
grantees to address issues associated with foreclosed and abandoned
properties. For example, grantees may choose to acquire and
rehabilitate properties for rental or resale, or demolish blighted
structures. HERA also imposed several key requirements governing the
use of NSP 1 funds. For example, grantees must use the funds in "areas
of greatest need" within 18 months of receiving them from HUD; for
this requirement, HUD interpreted "use" to mean obligate. In addition,
HERA required that all NSP 1 funds benefit households with incomes at
or below 120 percent of area median income (AMI), and at least 25
percent of the funds benefit households with incomes at or below 50
percent of AMI; for the 25 percent requirement, HUD interpreted "use"
to mean expended.
HERA mandated that GAO conduct periodic audits of whether NSP 1 funds
are being used in a manner consistent with criteria set forth in HERA.
This review examines (1) grantees' progress and challenges in meeting
HERA obligation time frames and income-targeting criteria, (2) HUD's
actions to mitigate program risks and ensure grantees' compliance with
key NSP 1 requirements, and (3) HUD's efforts to collect program data
and assess program performance.
To review grantees' progress and challenges in meeting HERA obligation
time frames and income-targeting criteria, we analyzed HUD data on
program obligations and grantees' uses of funds. We reviewed state
grantees' action plans and reports on NSP 1 activities to determine
the methods states used to distribute NSP 1 funds to subrecipients,
and the factors they considered in making these decisions. We analyzed
Disaster Recovery Grant Reporting (DRGR) data to determine the extent
to which the subrecipients also received NSP 1 funds directly from
HUD. We spoke with selected grantees to discuss their approaches to
using NSP 1 funds and the challenges they faced in implementing the
program.
To determine the actions HUD has taken to mitigate program risks and
ensure grantees' compliance with key NSP 1 requirements, we reviewed
NSP 1 regulations and guidance, examined HUD's efforts to oversee the
use of NSP 1 funds and provide technical assistance to grantees, and
interviewed HUD headquarters and field office officials and
representatives of selected grantees. In addition, we reviewed HUD's
guidance and procedures for monitoring grantee compliance with program
requirements and obtained and summarized on-site monitoring results
(as of September 15, 2010) from four HUD field offices responsible for
overseeing grantees we visited. We also conducted limited tests of
selected grantees' compliance with key NSP 1 requirements and relevant
internal controls through reviews of property-level records of
completed NSP 1 activities (see discussion of site visits below).
Additionally, we spoke with staff from the HUD Office of the Inspector
General (OIG) about their NSP-related audits and reviewed the results
of their completed audits.
To review HUD's efforts to collect program data and to assess the
reliability of the data, we reviewed HUD OIG audits of the information
system HUD adapted for NSP 1 (the DRGR system) and HUD documentation
and guidance for using the system. We also interviewed HUD and grantee
staff responsible for entering and monitoring DRGR data. We also
assessed the reliability of the DRGR information we used by conducting
reasonableness checks to identify any missing or erroneous data and by
interviewing knowledgeable HUD officials to ensure we interpreted the
data correctly. For the purpose of this and the first objective, we
concluded that the data we used were sufficiently reliable for our
purposes. To review HUD's efforts to assess program performance, we
interviewed HUD's Office of Policy Development and Research and Office
of Community Planning and Development (CPD) staff knowledgeable of a
planned NSP assessment.
To address all of our objectives, we interviewed representatives from
18 NSP 1 grantees. We interviewed a group of 11 grantees selected to
represent different housing markets, grant amounts, and grantee types
to discuss their approaches for using NSP 1 funds, progress they had
made, and challenges they faced in implementing the program. We
conducted site visits at another group of 8 grantees selected to
represent a variety of completed NSP 1 activities, types of grantees,
and geographic areas.[Footnote 4] To avoid overburdening grantees, we
focused our site visits on grantees that, on the basis of information
from HUD, were not receiving extensive technical assistance or being
monitored on-site by HUD staff. At these sites, we interviewed grantee
staff to obtain information on their processes for complying with NSP
1 requirements and internal control procedures, and reviewed property-
level records of completed NSP 1 activities. We also interviewed staff
from four HUD field offices: Columbus, Ohio; Miami, Florida;
Philadelphia, Pennsylvania; and San Francisco, California. Appendix I
contains a more detailed description of our scope and methodology and
a list of NSP 1 grantees we contacted.
We conducted our work in two phases. We did our initial audit work
(April through December 2009) in Washington, D.C., and at 15 sites
across the United States, including Florida, Indiana, Maryland,
Nevada, Pennsylvania, and Virginia. This work resulted in briefings to
your staff in December 2009 on the status of HUD's and grantees'
initial implementation of NSP 1. The second phase of our NSP 1 work
was conducted in Washington, D.C., and at 8 sites across the United
States, including locations in Arizona, Florida, Ohio, Pennsylvania,
and Virginia from January through December 2010. This report focuses
on the results of our audit work during the second phase. We conducted
all of our NSP 1 work in accordance with generally accepted government
auditing standards. Those standards require that we plan and perform
the audit to obtain sufficient, appropriate evidence to provide a
reasonable basis for our findings and conclusions based on our audit
objectives. We believe that the evidence obtained provides a
reasonable basis for our findings and conclusions.
Background:
Congress established NSP to try to reduce inventories of foreclosed
homes by providing funding to state and local governments to, among
other things, acquire and rehabilitate or demolish foreclosed and
abandoned homes. HERA allows grantees to use NSP 1 funds to:
* establish financing mechanisms for the purchase and redevelopment of
foreclosed homes and residential properties;
* purchase and rehabilitate homes and residential properties that have
been abandoned or are in foreclosure in order to sell, rent, or
redevelop them;
* establish land banks for foreclosed homes and residential properties;
* demolish blighted structures; and:
* redevelop demolished or vacant properties.[Footnote 5]
Land banks are governmental or nongovernmental nonprofit entities
established, at least in part, to assemble, temporarily manage, and
dispose of vacant land for the purpose of stabilizing neighborhoods
and encouraging reuse or redevelopment of urban property. Land banks
often acquire properties through tax foreclosures, intergovernmental
transfers, nonprofit transfers, and open-market purchases but can also
acquire them through foreclosure.
In addition to these five eligible uses, HERA outlined the following
key requirements for using NSP 1 funds, as follows:
1. All NSP 1 funds must be used within areas of greatest need as
determined by grantees on the basis of the following factors:
* percentage of foreclosures,
* percentage of homes financed by subprime loans, and:
* likelihood of a significant rise in the rate of foreclosures.
2. All NSP 1 funds must be used to benefit households with incomes at
or below 120 percent of AMI.
3. At least 25 percent of NSP 1 funds must benefit households with
incomes at or below 50 percent of AMI.
4. The sale, rental, or redevelopment of abandoned and foreclosed
homes must remain affordable to the above income groups for the
longest feasible term.
5. All NSP 1 funds must be used (obligated) by grantees within 18
months of their receipt.
6. All properties must be acquired at a discount established by HUD.
7. If a property is sold to an individual as a primary residence, the
sale price cannot be more than the total acquisition and redevelopment
costs.
In October 2008, HUD issued guidance on implementing NSP 1 in a Notice
in the Federal Register.[Footnote 6] With respect to the requirement
that NSP 1 funds be used within 18 months, HUD stated, "funds are used
when they are obligated by a state, unit of general local government,
or any subrecipient thereof for a specific NSP activity; for example
the acquisition of a specific property." According to the Notice,
funds are considered to be obligated for an activity when orders are
placed, contracts are awarded, and goods and services are received.
HUD specified that funds are not obligated for an activity when
subawards (for example, grants to subrecipients or units of local
government) are made. The Notice also stated that grantees must expend
their NSP 1 funds within 4 years of receipt. In accordance with HERA,
HUD established a required discount rate that would apply to any
property acquired through NSP 1. Initially, HUD set the discount at 15
percent for grantees' total acquisition portfolios but subsequently
reduced it to 1 percent per property acquired.[Footnote 7] HUD reduced
the required discount to provide grantees with more flexibility in
acquiring foreclosed properties and to avoid the potentially adverse
impact of discounts on neighborhood property values. Among other
provisions, HUD required that each NSP-assisted homebuyer complete at
least 8 hours of homebuyer counseling.
HUD also required that NSP 1 grantees submit "NSP 1 action plans" in
the form of substantial amendments to their existing CPD consolidated
plan.[Footnote 8] HUD required grantees to identify in these plans
areas of greatest need within their states or communities, approaches
for implementing NSP 1 and expected program outputs, and plans for
complying with key requirements. After grantees submitted these plans
and HUD approved them, HUD executed grant agreements for NSP 1.
NSP 1 implementation has a tight time frame (see figure 1). For
instance, HUD had to establish its formula for allocating funds 60
days after the enactment of HERA. As HERA linked NSP 1 to CDBG, HUD
was able to take advantage of the administrative infrastructure of one
of the federal government's largest and most flexible community and
economic development programs. Established in 1974, CDBG provides
states and local entitlement communities--eligible metropolitan cities
and counties--with funding to address locally determined community and
economic development priorities.[Footnote 9] CDBG funds may be used to
rehabilitate housing, improve infrastructure, provide job training,
and fund other community-determined projects.[Footnote 10]
Figure 1: NSP 1 Implementation Time Line:
[Refer to PDF for image: time line]
October 2008:
$3.92 billion allocated to grantees by HUD formula.
December 2008:
Grantees submitted NSP 1 substantial amendments for approval.
March 2009:
Grant agreements signed.
September 2010:
Deadline to obligate NSP 1 funds[A].
March 2013:
Deadline to expend NSP.
Source: GAO analysis of HUD information.
[A] A grantee's specific deadline depended on the date it received NSP
1 funds from HUD, so obligation deadlines varied among grantees.
According to HUD, 300 NSP 1 grantees had obligation deadlines expiring
in September 2010. The obligation deadlines for the remaining grantees
were some time in August, October, or November 2010.
[End of figure]
In accordance with the time frames and criteria established by HERA,
in October 2008 HUD allocated NSP 1 funds using formulas based on the
number and percentage of the following:
* foreclosed homes in each state or locality,
* subprime mortgages in each state or locality, and:
* homes in default or delinquency in each state or locality.[Footnote
11]
As required by HERA, all 50 state governments and Puerto Rico received
an allocation of at least $19.6 million. HUD also allocated NSP 1
funds to local (city and county) governments.[Footnote 12] Figure 2
shows the total state and local government allocations for each state.
Figure 2: Total NSP 1 State and Local Government Allocations, by State:
[Refer to PDF for image: illustrated U.S. map and associated vertical
bar graph]
State: Florida;
Total NSP 1 State and Local Government Allocations: $541.4 million.
State: California;
Total NSP 1 State and Local Government Allocations: $529.6 million.
State: Michigan;
Total NSP 1 State and Local Government Allocations: $263.6 million.
State: Ohio;
Total NSP 1 State and Local Government Allocations: $258.1 million.
State: Texas;
Total NSP 1 State and Local Government Allocations: $178.1 million.
State: Illinois;
Total NSP 1 State and Local Government Allocations: $172.5 million.
State: Georgia;
Total NSP 1 State and Local Government Allocations: $153.0 million.
State: Indiana;
Total NSP 1 State and Local Government Allocations: $151.9 million.
State: Arizona;
Total NSP 1 State and Local Government Allocations: $121.1 million.
State: New York;
Total NSP 1 State and Local Government Allocations: $100.3 million.
State: Pennsylvania;
Total NSP 1 State and Local Government Allocations: $88.1 million.
State: Tennessee;
Total NSP 1 State and Local Government Allocations: $72.5 million.
State: Nevada;
Total NSP 1 State and Local Government Allocations: $71.9 million.
State: Missouri;
Total NSP 1 State and Local Government Allocations: $64.9 million.
State: New Jersey;
Total NSP 1 State and Local Government Allocations: $64.0 million.
State: Minnesota;
Total NSP 1 State and Local Government Allocations: $57.8 million.
State: North Carolina;
Total NSP 1 State and Local Government Allocations: $57.7 million.
State: Massachusetts;
Total NSP 1 State and Local Government Allocations: $54.8 million.
State: Colorado;
Total NSP 1 State and Local Government Allocations: $53.1 million.
State: South Carolina;
Total NSP 1 State and Local Government Allocations: $49.2 million.
State: Wisconsin;
Total NSP 1 State and Local Government Allocations: $48.0 million.
State: Maryland;
Total NSP 1 State and Local Government Allocations: $46.4 million.
State: Mississippi;
Total NSP 1 State and Local Government Allocations: $46.3 million.
State: Virginia;
Total NSP 1 State and Local Government Allocations: $46.0 million.
State: Kentucky;
Total NSP 1 State and Local Government Allocations: $44.4 million.
State: Alabama;
Total NSP 1 State and Local Government Allocations: $41.8 million.
State: Louisiana;
Total NSP 1 State and Local Government Allocations: $38.8 million.
State: Oklahoma;
Total NSP 1 State and Local Government Allocations: $32.9 million.
State: Washington;
Total NSP 1 State and Local Government Allocations: $28.2 million.
State: Connecticut;
Total NSP 1 State and Local Government Allocations: $25.0 million.
State: Iowa;
Total NSP 1 State and Local Government Allocations: $21.6 million.
State: Kansas;
Total NSP 1 State and Local Government Allocations: $21.0 million.
State: Alaska;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Arkansas;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Delaware;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Hawaii;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Idaho;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Maine;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Montana;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: North Dakota;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Nebraska;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: New Hampshire;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: New Mexico;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Oregon;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Rhode Island;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: South Dakota;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Utah;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Vermont;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: West Virginia;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: Wyoming;
Total NSP 1 State and Local Government Allocations: $19.6 million.
State: District of Columbia;
Total NSP 1 State and Local Government Allocations: $2.8 million.
Source: GAO analysis of HUD data; MapInfo.
[End of figure]
HUD CPD is implementing NSP 1. CPD administers CDBG and other formula
and competitive grant programs from its offices at HUD headquarters
and 43 field offices located throughout the United States. The
headquarters office sets program policy, while field office staff are
responsible for a wide range of grant management activities, including
annual reviews and approval of grantee action plans, execution of
grant agreements, and monitoring of grant recipients. CPD's monitoring
of the CDBG program focuses on high-risk recipients. Each year, CPD
sets a formal monitoring goal. For fiscal year 2010, the overall goal
for CPD and each of its field offices was to monitor a minimum of 870
formula and competitive grant recipients. For each CPD program, field
staff analyze potential risks for grantees in a number of areas,
including: financial (size of grant, financial compliance), physical
(physical control of assets), management (staff capacity),
satisfaction (citizen complaints and grantee responsiveness), and
services (meeting program objectives and progress).
HUD modified the DRGR system for the purposes of collecting and
maintaining information on NSP activities. DRGR is new to NSP 1
grantees and HUD field staff unless they had prior experience with
Disaster Recovery Assistance grants, another program managed by CPD.
[Footnote 13] In its NSP 1 Notice, HUD explained that no other
reporting system was as flexible as DRGR and that statutory time
frames did not allow sufficient time to develop a new system or modify
an existing one to perfectly fit NSP. HUD's key modifications to DRGR
included:
* adding data fields to collect information specific to NSP, such as
land banking;
* removing drop-down menu selections for ineligible uses under NSP;
and:
* adaptations to enable grantees to designate whether properties had
been resold or rented or were single-or multifamily.
DRGR was set up to monitor grantees' progress in implementing NSP 1 by
tracking budgets, obligations, expenditures, and program outputs. DRGR
also serves as a financial management system through which grantees
obligate and draw down funds. At the beginning of NSP 1, grantees
inputted their NSP 1 action plans into DRGR. In their actions plans in
DRGR, grantees set up their budgets and performance measures for
different DRGR activities that were linked to larger projects. The
action plan serves as a template for required progress reports, or
Quarterly Performance Reports (QPR), that grantees must submit to HUD
field staff on a quarterly basis. QPRs show updated information on NSP
1 progress for grantees' projects and activities listed in DRGR.
For each DRGR activity, NSP 1 grantees complete data fields that
include (1) the projected start date, (2) the national objective (the
income group the activity benefits), (3) the responsible organization
(either the grantee or a subrecipient), and (4) the activity type
(e.g., acquisition, rehabilitation, demolition).[Footnote 14] For each
DRGR activity, grantees must also include a narrative and output
measures--for example, the number of housing units or properties
acquired, rehabilitated, land banked, or demolished, or the numbers of
households or persons benefiting from the activity. In addition,
grantees must list the addresses of properties benefiting from the
activity but are not required to include the amount of funds spent on
each property in DRGR. In a June 2009 report, the HUD OIG concluded
that DRGR was sufficient to collect the basic information that HUD
needed to monitor NSP 1.[Footnote 15] In a September 2009 report, the
HUD OIG reviewed selected controls within DRGR in order to examine
whether funds were properly safeguarded. While the OIG did not find
misuse of funds in their review, they did identify several weaknesses,
including access policies and testing of transactions, that CPD staff
have been addressing.[Footnote 16]
NSP 1 Grantees Overcame Challenges and Used a Variety of Strategies to
Meet Obligation and Income-Targeting Requirements:
Most Grantees Met the Deadline for Obligating Funds:
HERA required NSP 1 grantees to obligate all of their funds within 18
months, creating a September 2010 deadline for most grantees.[Footnote
17] In order to obligate their funds, grantees had to satisfy certain
conditions, including providing a specific property address and (where
applicable) a detailed scope of work for rehabilitating, constructing,
or demolishing a property. According to HUD guidance, grantees are
considered to have obligated NSP 1 funds once they have placed orders,
awarded contracts, received goods and services, or executed similar
transactions that will require payment by the grantee. For example,
funds for acquiring a property are considered obligated when the
property seller accepts the grantee's purchase offer.
Data in DRGR indicate that the vast majority of the 309 NSP 1 grantees
met their deadline for obligating funds.[Footnote 18] As a result,
more than 99 percent of NSP 1 funds were obligated as of October 1,
2010. Figure 3 shows grantees' progress toward meeting obligation
deadlines as of June 24, 2010 (about 3 months prior to most grantees'
deadlines), compared with their progress by their deadlines. As of the
earlier date, 27 grantees had obligated an amount equal to 100 percent
of their total NSP 1 grant funds. In contrast, 275 grantees had
obligated 100 percent by October 1, 2010--at which point the vast
majority of grantee's deadlines had passed--and 29 had obligated
between 80 percent and 99.9 percent. These figures are preliminary
because HUD (1) gave grantees 30 days after their deadlines to ensure
that all obligations prior to the deadline were recorded in DRGR and
(2) is in the process of reviewing whether grantees' obligated their
NSP 1 funds properly.[Footnote 19] CPD staff told us that 1 business
day after a grantee's obligation deadline passes, they are blocking
the grantee's access to funds and reviewing samples of obligations to
ensure they are proper. As of the end of September, HUD officials told
us that field office staff had reviewed samples for 20 grantees and
found no improper obligations. They plan to review all grantees for
compliance with obligation guidance. Additionally, the OIG is
reviewing obligations as part of its compliance audits of selected NSP
1 grantees.[Footnote 20] In August 2010, HUD outlined a process for
addressing a grantee's failure to meet its obligation deadline and
anticipated that many such grantees would face a choice of entering
into an agreement with HUD to use the remaining funds or having their
funds recaptured.[Footnote 21]
Figure 3: NSP 1 Grantees' Obligations as of June 24, 2010, and as of
October 1, 2010:
[Refer to PDF for image: combined table and horizontal bar graph]
Percentage of funds obligated: 0% to 19.9%;
Number of grantees, June: 18;
Number of grantees, October: 0.
Percentage of funds obligated: 20.0% to 39.9%;
Number of grantees, June: 28;
Number of grantees, October: 0.
Percentage of funds obligated: 40.0% to 59.9%;
Number of grantees, June: 70;
Number of grantees, October: 1.
Percentage of funds obligated: 60.0% to 79.9%;
Number of grantees, June: 87;
Number of grantees, October: 2.
Percentage of funds obligated: 80.0% to 99.9%;
Number of grantees, June: 77;
Number of grantees, October: 29.
Percentage of funds obligated: 100%;
Number of grantees, June: 27;
Number of grantees, October: 275.
Total:
Number of grantees, June: 307[A];
Number of grantees, October: 307[A].
Source: GAO analysis of HUD DRGR data.
[A] There are 307 entries representing 309 grantees in HUD's DRGR
system because the State of Colorado reports information in DRGR for
both the State and for the City of Colorado Springs, and Clark County
reports information for both the County and the City of North Las
Vegas (Nevada) under their cooperative grant agreements.
[End of figure]
Figure 4 shows the percentage of grantee obligations that fell under
different NSP 1 activities as of October 1, 2010. NSP 1 grantees
obligated most NSP 1 funds for acquisition and rehabilitation, with
smaller amounts obligated for other activities such as construction of
new housing, demolition, and land banking.
Figure 4: Reported NSP 1 Grantee Obligations by Activity, as of
October 1, 2010:
[Refer to PDF for image: pie-chart]
Acquisition and rehabilitation: 66%;
Construction for new housing: 11%;
Administration: 9%;
Homeownership assistance: 6%;
Demolition: 5%;
Other: 4%:
- Public improvement, public facilities, nonresidential structures: 2%;
- Land banking: 2%.
Source: GAO analysis of HUD DRGR data.
Notes: The purpose of this graph is to show the relative magnitude of
NSP 1 obligations by activity type. As discussed later in this report,
some grantees may group multiple activities under a single activity
type (for example, both acquisition and rehabilitation activities
under rehabilitation) when entering obligation data into DRGR. Because
this practice is most likely to occur when grantees enter data on
acquisition and rehabilitation activities, we combined the data for
these two activities. The "disposition" activity type is also likely
to be included under rehabilitation, therefore we included obligations
for disposition within the "acquisition and rehabilitation" grouping.
"Homeownership assistance" includes various types of homeownership
assistance such as down-payment assistance and housing counseling for
those applying for down-payment assistance.
Due to rounding, the percentages in the graphic add up to a total of
101 percent. Funds spent toward public services and relocation
payments and assistance were not included in the graph because they
each rounded to zero percent, as did the total amount of funds not yet
obligated.
[End of figure]
HUD data also show that 298 NSP 1 grantees obligated at least 25
percent of their grant funds for activities benefiting low-income
households. As of October 1, 2010, the data indicate that, in total,
grantees obligated 35.1 percent of funds set aside for activities that
would benefit this group. This pattern is consistent with a HERA
requirement that 25 percent of NSP 1 funds be used to benefit
households at or below 50 percent of AMI. In the NSP 1 Notice, HUD
stated that it will assess grantees' compliance with this provision
prior to and at grant closeout on the basis of grantees' expenditures.
State Grantees Often Used a Combination of Methods to Distribute NSP 1
Funds within Their States:
As discussed earlier, HUD made NSP 1 allocations to 309 units of
government, including each of the 50 states. Collectively, the states
received more than half of NSP 1 funds. States had some flexibility in
how they chose to distribute NSP 1 funds but were required to target
areas of greatest need. States could choose to use funds themselves
through state-administered programs or could reallocate funds to local
entitlement or nonentitlement communities within their areas of
greatest need.[Footnote 22] In their HUD-approved action plans, states
had to identify areas of greatest need using factors that included the
following:
* foreclosure rates,
* percentage of homes financed by subprime loans, and:
* likelihood of a significant rise in foreclosure rates.
Our review of NSP 1 action plans for all states (including the
District of Columbia) found that all of the states included these
factors among their criteria for determining their areas of greatest
need. Some states also used factors not listed in HERA in determining
their areas of greatest need, such as unemployment rates or
percentages of the population at or below 50 percent of AMI. In their
action plans, states had to outline their methods for reallocating NSP
1 funds or for using the funds themselves to make the most impact
within the areas of greatest need within their state. HERA also
allowed states to allocate their funds to local entitlement
communities that may have already received NSP 1 funds directly from
HUD, as well as to nonentitlement communities. According to HUD data
as of June 2010, of the total NSP 1 funds that HUD allocated to the
states, states reallocated approximately 9 percent, or $176 million,
to 57 local governments in 21 different states that had already
received NSP 1 funds directly from HUD (see figure 5).[Footnote 23]
Figure 5: States' Reallocation of NSP 1 Funds to Units of Local
Government That also Received Funds Directly from HUD:
[Refer to PDF for image: illustration]
HUD: Direct funding to:
* 50 states;
* 253 units of local government.
21 of the 50 states gave additional funding to:
* 57 of the 253 units of local government.
Source: GAO analysis of HUD DRGR data; Art Explosion.
[End of figure]
Once states determined the areas of greatest need, many of them used
competitions to award NSP funds for specific purposes. For example,
one state used a competition for the acquisition and rehabilitation of
multifamily properties, and another state used a competition
specifically to help it meet NSP 1's 25 percent set-aside requirement
to benefit households with incomes at or below 50 percent of AMI.
Eligible applicants--nonprofits, for-profit developers, and local
governments, among others--often varied by state and by the type of
competition. While some states invited nonprofits and for-profit
developers to compete for funds directly, other states allowed only
local jurisdictions to apply or required nonprofits to get buy-in from
their local governments in order to receive funds.
Our review of NSP 1 action plans found that states distributed funds
using three main methods: (1) reallocation through competitions, (2)
reallocation on the basis of formulas, and (3) through state-
administered programs. Many state grantees used a combination of
methods to distribute funds, while a smaller number relied on a single
method. More specifically, our review found that 14 states distributed
all of their NSP 1 funds by reallocating them on a competitive basis,
7 states used a formula to reallocate all of their funds, and 1 state
used all of the funds itself by administering the program through its
housing finance agency and by contracting with other entities. In
contrast, 26 states used a combination of two or more of the methods.
[Footnote 24] For example, a number of states with some of the largest
NSP 1 grants, including California and Ohio, used a formula to
allocate funds to areas of greatest need within the state and then
held a competition for the 25 percent set-aside to benefit households
with incomes at or below 50 percent of AMI. Six states utilized all
three distribution methods--competition, formula, and use of funds--
through state-administered programs. A number of states decided to
administer some of their NSP 1 funds themselves through state-
administered programs; for example, 1 state created a new statewide
soft second loan financing program (Arizona) and another state
channeled some of its NSP 1 funds into the Low Income Housing Tax
Credit Program and the Permanent Supportive Housing Program (Georgia).
[Footnote 25]
Many states that used competitions to distribute NSP funds used
similar criteria to rank applicants and award funds. For example, on
the basis of our review of their action plans, many states considered
applicants' capacity to carry out projects and project readiness, as
well as whether applicants' proposals focused on projects that would
serve the areas of greatest need within the state. Other frequently
used competitive factors included applicants' experience administering
NSP-related activities and ability to leverage other funding sources.
A few states ranked applicants based on green building activities or
gave priority to applicants that proposed projects that would help
meet the 25 percent set-aside requirement.
Grantees Employed Various Strategies for Using NSP 1 Funds:
NSP 1 grantees with whom we spoke employed various different
strategies for obligating and expending funds for NSP-eligible
activities.[Footnote 26] For example, as shown in figure 6, some
grantees found that homebuyer-driven strategies--that is, strategies
that allowed individual homebuyers to identify NSP-eligible properties
and receive NSP 1 funds from grantees to assist with the acquisition
or rehabilitation--were the most appropriate for the market conditions
in their area. An example of a homebuyer-driven strategy would be one
in which a grantee provided homebuyers with financial assistance--down-
payment assistance, closing costs, rehabilitation costs, or soft
second loans--to enable them to acquire a home within an NSP 1 target
area. Another approach grantees used was a more property-driven
strategy in which grantees or their subrecipients--such as housing
authorities, nonprofits, or for-profit developers--directly acquired,
rehabilitated, and resold or rented NSP properties to individuals or
families meeting NSP-specific income requirements. Finally, in some
cases, grantees' strategies included demolishing blighted structures
and land banking the parcels on which the demolished properties stood
for future use.
Figure 6: NSP Strategies for Acquiring and/or Rehabilitating Homes:
[Refer to PDF for image: illustration]
Homebuyer-driven strategy:
Homebuyer:
Identifies and purchases property.
NSP 1 grantee:
Provides financial assistance (downpayment assistance, closing costs,
etc.) and ensures NSP 1 due diligence has been completed.
Homebuyer:
(meets NSP 1-specific income requirements).
Property-driven strategy:
NSP 1 grantee:
Acquires and/or rehabilitates property within an NSP 1 target area;
Grantee resells or rents property to homebuyer;
Homebuyer (meets NSP 1-specific income requirements).
Source: GAO; Art Explosion.
[End of figure]
NSP 1 grantees we spoke with used different NSP strategies depending
on the particular housing market conditions in their target areas. For
example, one grantee we contacted used a homebuyer-driven strategy in
areas where the generally good condition of the properties could
attract potential purchasers. Under that strategy, qualified
homebuyers looked for properties in NSP 1 target areas and chose NSP-
eligible homes that did not require a lot of rehabilitation. The
grantee then conducted the required due diligence on the home,
including having it appraised and inspected, and completing
environmental reviews. After the due diligence had been completed, the
grantee assisted buyers with the acquisition of a property (at the
required discount) and provided some funding for down-payment
assistance and closing costs. The same grantee implemented a more
property-driven strategy in target areas where homes required a higher
level of rehabilitation--a potential deterrent for some homebuyers.
Under this strategy, the grantee worked with nonprofit developers to
acquire and rehabilitate the homes, then resold the homes to NSP-
eligible buyers. Another grantee with whom we spoke used NSP 1 funds
for demolition projects and the removal of blight to help stabilize
its communities. Some of the land on which the demolished properties
were located will be land banked for future use. This strategy made
the most sense for this particular grantee as it did not have a strong
housing market for sale or rental.
In order to meet the 25 percent set-aside requirement to benefit
households with incomes at or below 50 percent of AMI, most of the
grantees we spoke with employed similar strategies. Several of the
grantees we spoke with planned to acquire, rehabilitate, and rent out
multifamily properties to meet this requirement. A couple of grantees
also had back-up strategies for meeting the 25 percent set-aside as
well, such as acquiring and rehabilitating single-family homes and
renting them out to low-income individuals or families.
Some Grantees Modified Their Strategies to Meet the Obligation
Deadline and Address Implementation Challenges:
Grantees we spoke with faced several challenges implementing NSP 1--
including tight time frames, competition from private investors, and
challenges in acquiring properties--but generally found ways to
address them. While 16 of the 18 NSP 1 grantees that we contacted had
obligated their full grant allocations by October 1, 2010, most said
they had faced some difficulties in trying to meet this requirement.
[Footnote 27] For example, several grantees mentioned they had
encountered competition from private investors in their efforts to
acquire NSP-eligible properties. One grantee said both in-state and
out-of-state investors have seized the opportunity to acquire NSP-
eligible properties at low prices, and another grantee provided an
example where a group of investors acquired about 90 properties at one
time. One grantee official expressed concern that investors looking to
rent out properties may not be responsible landlords, which could
undermine efforts to stabilize neighborhoods. Finally, private
investors sometimes outbid grantees for NSP-eligible properties, and
because investors may pay with cash and do not have to meet the same
due diligence requirements as NSP 1 grantees, they are able to act
more quickly and have more success acquiring properties.
Furthermore, a few of the grantees we spoke with that planned to rent
out multifamily properties said they faced challenges meeting the 25
percent set-aside obligation requirement. Officials from two grantees
told us it was difficult to find foreclosed multifamily rental
projects to purchase in NSP 1 target areas. Additionally, one grantee
told us that they were facing neighborhood opposition to a multifamily
project because neighbors feared that affordable housing units would
depress home values in the area.
A broader study of NSP grantees sponsored by the Board of Governors of
the Federal Reserve System and two Federal Reserve banks cited
challenges similar to those that we identified. For example, the study
highlights competition from private-sector investors, grantees'
difficulties identifying a potential pool of foreclosed properties to
acquire, and the reluctance of some property holders to work with
grantees. The study also noted that local requirements and practices,
such as stringent local standards for publicly financed
rehabilitation, often put NSP grantees at a competitive disadvantage
to investors.[Footnote 28]
In order to overcome implementation challenges such as meeting
deadlines or competition from investors, some grantees we spoke with
modified their implementation strategies by:
* moving away from homebuyer-driven models to directly acquiring
properties,
* expanding target areas,
* working with banks' "first look programs," and:
* working with the National Community Stabilization Trust (NCST).
To overcome implementation challenges, some grantees had to make
changes to their original program strategies. For example, several
grantees we contacted said it would be difficult to meet the
obligation deadline using their homebuyer-driven strategies.
Competition from private investors and tight obligation time frames
required some grantees to move funds from their homebuyer-driven
models towards supporting more property-driven strategies where they
purchased and rehabilitated homes themselves. Using a more direct
strategy allowed some grantees to speed up the pace of obligations. In
addition, one grantee shifted its focus from homebuyers to renters in
order to help it meet the deadline for obligating funds, as potential
homebuyers in its housing market were finding it difficult to obtain
mortgages due to job loss and poor credit.
Some NSP 1 grantees also expanded their target areas to expedite
property acquisition. With HUD approval, some NSP grantees expanded
their target areas to include more ZIP codes or census tracts in order
to expedite acquisitions and circumvent investor competition. Several
grantees also worked with banks' "first look" programs, which gave
them the opportunity to bid on bank-owned foreclosed properties before
other potential buyers. Grantees also worked with NCST, a national
nonprofit organization to acquire foreclosed properties in their
target areas. NCST facilitates communication between NSP grantees and
banks and provides grantees with listings of NSP-eligible properties
from which to choose. Many grantees with whom we spoke found NCST to
be very helpful in acquiring properties, and one grantee we contacted
said working with NCST helped it acquire 80 percent of the homes on
which it had placed bids. In an effort to standardize the acquisition
process for NSP grantees, HUD recently partnered with NCST to create
the "National First Look Program" in September 2010. The program gives
communities participating in NSP a "first look" or right of first
refusal to purchase foreclosed homes before the properties are made
available to private investors.[Footnote 29] According to HUD
officials, the program grew out of an initiative HUD announced in July
2010, which gave NSP-eligible purchasers the opportunity to acquire
foreclosed properties owned by HUD's Federal Housing Administration at
10 percent below the appraised value.[Footnote 30]
HUD Has Taken Actions to Mitigate NSP 1 Program Risks through
Training, Technical Assistance, and the Establishment of Internal
Controls:
As previously noted, HERA established specific program requirements
for NSP 1 beyond those for CDBG, including an accelerated timeline for
obligating funds and the requirement that grantees must use 25 percent
of their NSP 1 funds to benefit households at or below 50 percent AMI.
Additionally, HERA required that NSP 1 grantees ensure, among other
things, that properties are acquired at a discount and properties sold
to individuals as a primary residence cannot be resold for more than
the total cost of acquisition and redevelopment.
At the start of NSP 1, HUD conducted a front-end risk assessment
(FERA) to assess NSP 1 program risks and identify actions required to
reduce control risks to an acceptable level.[Footnote 31] Among other
risks, HUD's review noted that the scale of the effort could overwhelm
some grantees; meeting the requirement to obligate funds within an 18-
month time frame could be challenging for grantees, especially those
with limited real estate experience; and HUD's personnel resources
were not adequate to oversee the program. The FERA concluded that
implementing NSP 1 successfully and within required time frames would
require early and regular deployment of controls, including training,
technical assistance, and monitoring. The FERA also recommended
additional HUD staff to support NSP-related activities, because field
staff were already overextended managing existing programs.
HUD Provided Training and Guidance and Made Program Modifications to
Address Program Challenges, but Some Grantees Said Some of These
Efforts Were Not Timely:
HUD provided training and guidance on NSP 1 using a variety of
methods, including:
* online guidance, webinars, and answers to frequently asked questions;
* e-mail updates;
* teleconferences and increased communication with field staff; and:
* classroom and one-on-one instruction (e.g., on DRGR).
HUD provided eight online video training presentations with associated
printouts ("modules") that outlined NSP 1-specific requirements,
policies, and procedures. For example, the first module provided
overall NSP 1 guidance and rules on eligible uses, activities, and
properties.[Footnote 32] HUD also released webinars, hosted by HUD
staff and technical assistance providers, on a weekly basis that
provided guidance on topics such as obligating funds and meeting the
25 percent set-aside requirement. A majority of the grantees we spoke
with said that, overall, the material covered in the modules and
webinars provided good information and helped them implement their NSP
1 programs.
HUD faced significant challenges in issuing comprehensive guidance and
training to grantees at NSP 1's outset because the program was
relatively complex and had to be implemented quickly. Most of the
grantees that we interviewed said that the combination of HUD's
guidance and training was helpful but some grantees had raised some
concerns. Several grantees that we spoke with about 1 year after the
program began said that the guidance and training they received was
not always timely. For example, one grantee said that the information
it received at an NSP problem-solving clinic in early 2010 would have
been more useful had it been provided 6 months earlier. Additionally,
two grantees said they did not always get timely answers to questions
about program rules when the field office staff they asked had to
obtain clarification from HUD headquarters. HUD officials said that
some of the assistance grantees found useful was delivered using funds
that HUD received well after the start of NSP 1.
HUD modified NSP policy in two key respects to address challenges
faced by grantees in implementing the program. Grantees we spoke with
said that the first of these modifications helped them overcome
difficulties they were experiencing in acquiring properties. More
specifically, in June 2009, HUD reduced the required discount from the
appraised value at which grantees must purchase NSP-eligible
properties. HUD reduced the discount from a minimum of 5 percent for
individual acquisitions to a minimum of 1 percent, and eliminated the
15 percent discount for aggregate acquisitions.[Footnote 33] This
change was intended to mitigate the potentially adverse impact that
the larger discounts could have on neighborhood property values and
the inability or unwillingness of the holders of foreclosed properties
to sell at prices that reflect the higher discount.
NSP 1 grantees had mixed views about the effect of the other program
modification, implemented in April 2010, which expanded the
definitions of foreclosed and abandoned properties.[Footnote 34] As a
result of the program modification, properties no longer had to have
completed the foreclosure process to be purchased with NSP funds, and
the standards for what constituted an abandoned property eligible for
NSP were loosened--for example, to include tenant-occupied properties.
[Footnote 35] HUD made these changes in response to suggestions from
local communities to increase the reach of NSP by allowing more
properties to qualify. HUD headquarters staff said this change also
enabled some NSP 1 grantees to increase the pace of their acquisitions
and meet their obligation deadlines. They added that, on the basis of
comments from grantees, the expanded definitions had helped grantees
acquire multifamily properties but had not yet been used to acquire
single-family homes. Similarly, some CPD field staff said the expanded
definitions helped grantees that were facing challenges in meeting the
obligation deadline by increasing the number of eligible properties
they could acquire. Specifically, field staff reported that it helped
some of their grantees to fulfill the 25 percent set-aside requirement
by easing restrictions on acquiring multifamily properties, which many
grantees we spoke with intend to rent to low-income households.
Several grantees we spoke with expressed similar views, noting that
the new definitions helped them acquire multifamily properties.
However, others indicated that the broader definitions would not make
a significant impact at this stage in their NSP 1 efforts because they
had made many of their acquisitions prior to the definition changes or
had a large inventory of fully foreclosed homes available. The
expanded definitions may have a greater effect on acquisitions of
properties for the more recent rounds of funding (NSP 2 and NSP 3),
since they are in earlier stages of planning and implementation.
HUD Provided Technical Assistance to Support Grantees in Implementing
NSP 1:
In August 2009, HUD awarded $50 million in ARRA funds through a
competition to provide technical assistance for NSP grantees (under
both NSP 1 and NSP 2). According to HUD, as of October 2010, $28.6
million had been budgeted for NSP technical assistance and $11 million
of that amount had been drawn down. Nine national technical assistance
providers were awarded 89 percent of the funds to operate at the
national level, and three local technical assistance providers
received 11 percent of the funds to operate at the local level.
[Footnote 36] Among other areas, NSP technical assistance providers
were tasked with:
* helping NSP grantees implement sound underwriting, management, and
fiscal controls;
* building the capacity of public-private partnerships;
* developing strategies to serve low-income households;
* providing training on the operation and management of land banks;
and:
* training grantees and their subgrantees on HUD program rules and
financial management requirements.
The providers delivered support to grantees through various methods,
including "direct assistance" (assistance provided in person), "on-
call remote assistance" (assistance provided via phone or e-mail), and
"Web-based" assistance, which provides answers to questions submitted
via HUD's NSP resource exchange Web site.
* Direct. HUD headquarters and field staff identified grantees that
demonstrated a need for support in specific NSP program areas or
exhibited capacity concerns in meeting the obligation deadline.
Technical assistance providers conducted needs assessments for direct
assistance for 98 grantees during late 2009 and early 2010. As of
August 2010, 59 grantees had requested direct assistance, 58 of which
received it.
* On-call remote. As of August 2010, technical assistance providers
had received 93 requests for on-call remote technical assistance and
responded to 84 of those requests (with 27 completed and 57 in
progress). HUD officials said they received approximately 3 to 4 on-
call remote technical assistance requests weekly. Technical assistance
providers also facilitated NSP problem-solving clinics that were open
to all grantees and field staff.
* Web-based. HUD officials said that technical assistance providers
had established the NSP resource exchange Web site as a place for
grantees to submit technical assistance requests and share knowledge.
The Web site also has a "frequently asked questions" link that is
regularly updated with answers to questions from grantees that can be
searched by topic. According to HUD, technical assistance providers
answered 2,389 out of 2,544 NSP questions submitted between April and
August 2010.
HUD officials said that NSP 1 grantees mostly requested technical
assistance in two major areas: (1) program design and (2) financial
underwriting skills. In terms of program design, HUD staff said that
some grantees had selected unsuitable NSP 1 approaches for their local
housing markets. For example, some grantees had planned to implement
strategies for selling homes or increasing home ownership in what were
primarily rental markets. In another example, one grantee was
initially not planning to use NSP 1 funds for demolition, even though,
according to HUD staff, demolition was an appropriate use of NSP 1
funds given the grantee's local housing market conditions. Grantees
also requested technical assistance to enhance their knowledge of
property financing.
HUD set up accounts in DRGR for each technical assistance provider to
track the use of technical assistance funds and the performance of
grantees receiving the assistance. HUD staff told us they plan to use
this information to assess whether the technical assistance improved
grantees' performance. HUD staff also said they will conduct post-
technical assistance monitoring to assess grantees' compliance with
NSP 1 requirements and determine if additional assistance is
necessary. Additionally, HUD officials said that the agency is
collecting feedback on the quality of technical assistance at training
events and through the NSP resource exchange Web site, among other
means. HUD officials said they would use this feedback to determine
potential needs for NSP 2 and 3, and to help determine whether current
technical assistance providers should continue to participate in HUD
programs. Given the critical role that HUD assigned to technical
assistance for mitigating program risks, it will be important for HUD
to follow through with these planned assessments to ensure that the
assistance had the intended effect.
HUD Hired Staff and Implemented Internal Control Processes for NSP 1:
The FERA conducted for NSP 1 cited the lack of HUD staff capacity to
absorb the additional workload and recommended the addition of 20 to
25 staff to support NSP 1. With funding from the fiscal year 2009
continuing resolution, as of November 2010, HUD had hired 10 term
staff as NSP specialists--whose duties include monitoring and
oversight of grantees--placing 7 in field offices that generally had
more than 20 NSP 1 grantees and 3 in headquarters.[Footnote 37] Also
as of November 2010, the agency had hired an additional 32 field staff
with funding provided by the American Recovery and Reinvestment Act of
2009 (ARRA) to work on implementing NSP 2 but also to contribute to
NSP 1 efforts, such as monitoring grantee compliance with program
requirements.
The FERA also noted the importance of implementing controls to
mitigate program risks. HUD adapted a range of internal control
processes already in place for the CDBG program to ensure compliance
with program requirements and mitigate the risks posed by NSP 1 (see
figure 7). Before releasing grantees' NSP 1 funds, HUD required all
grantees to submit a substantial amendment (or "NSP 1 action plan") to
their CPD consolidated and annual action plans by December 2008 to
describe how they would use NSP 1 funds. As noted earlier, NSP 1
action plans identify areas of greatest need, establish expected
program outputs, set forth plans for complying with key requirements,
and define relevant NSP 1 terms (such as blighted structures, housing
rehabilitation standards, and affordable rents). HUD reviewed NSP 1
action plans to determine if grantees' planned uses of funds were in
accordance with key HERA requirements. After approving the action
plans, HUD signed the grant agreements (most were signed in March
2009, according to HUD officials, resulting in 18-month obligation
periods ending in September 2010).
Figure 7: Selected HUD Internal Controls for NSP 1:
[Refer to PDF for image: illustration]
I. Prior to access to NSP 1 funds:
HUD reviewed NSP 1 Action Plan (substantial amendments to consolidated
CPD plan) then signed grant agreement; Grant Agreement Approved.
Grantees input NSP 1 action plan into DRGR; High-risk grantees must
also submit NSP 1 Management Plan.
HUD reviewed using checklist: NSP 1 Action Plan DRGR Checklist.
Funds released.
II. NSP 1 funds monitoring:
HUD reviewed and approved QPRs submitted by all grantees, using Guide
for QPR Checklist.
HUD identified high-risk grantees in CPD analysis.
HUD conducted additional on-site monitoring of high-risk grantees
using NSP 1 Program Exhibit.
Funds can be frozen in cases of noncompliance.
Source: GAO analysis of HUD information.
[End of figure]
Also prior to releasing NSP 1 funds, HUD implemented additional
controls for grantees at higher risk of noncompliance with program
requirements. For example, in the initial stage of the program, HUD
designated 104 grantees as high risk because their NSP 1 grants were
at least three times greater than their annual CDBG allocations or
they had audit findings or other performance problems in similar HUD
grant programs. HUD added a "special condition" to the NSP 1
agreements of these high-risk grantees that required them to submit
management plans along with their NSP 1 action plans. The management
plans detail the number, types, and responsibilities of staff
positions supporting the grantee's NSP 1 program.[Footnote 38]
Furthermore, after grantees inputted their action plans into DRGR, CPD
field office staff used a checklist to confirm that the action plans
were entered in a manner that would allow HUD to track NSP 1 funds and
monitor whether some key reporting requirements were met (see figure
7). Field staff used an action plan checklist to verify that grantees
had identified eligible NSP 1 uses, responsible organizations, and
income-based national objectives associated with activities in DRGR.
[Footnote 39] Once field office staff reviewed and approved a
grantee's action plan in DRGR, HUD headquarters released the grantee's
NSP 1 funds.
After releasing grantees' funds, HUD has used remote and on-site
monitoring of grantees' ongoing performance to assess compliance with
program requirements. HUD field staff used a QPR checklist to review
areas covered by the action plan checklist and to monitor grantees'
progress in meeting their obligation deadlines.
Additionally, CPD field office staff conducted on-site monitoring of
176 grantees (57 percent of NSP 1 grantees). The field offices
selected these grantees on the basis of risk scores calculated using
five risk factors: (1) financial soundness, (2) overall management
capacity, (3) client satisfaction with services, (4) extent to which
services benefit targeted areas and clientèle, and (5) NSP 1 program
progress.[Footnote 40] The factor for management, which constituted
the largest part of the score, considered grantees' capacity to
implement NSP 1 based on HUD monitoring information and grantees' use
of subrecipients. The factor for NSP 1 program progress--the second-
largest scoring component--considered other aspects of grantee
capacity, including how much NSP 1 funding the grantee received
relative to its regular CDBG allotment, the type of activities
undertaken, and the grantee's capacity to manage its NSP 1 program.
Although CPD field offices primarily used the risk analysis scores to
select grantees for monitoring and identifying specific compliance
areas to review, they also considered HUD OIG audit findings, where
applicable.[Footnote 41]
To prepare for on-site monitoring, CPD issued a revision to HUD's
Community Planning and Development Monitoring Handbook in April 2010
that provided HUD staff with guidance for monitoring grantees'
compliance with NSP 1 requirements. The guidance includes "exhibits,"
or checklists that cover key NSP 1 requirements, including the income
qualifications of program beneficiaries, rules on continued
affordability, obligation deadlines, and the 25 percent set-aside for
low-income families.
HUD field offices conducted the majority of on-site monitoring from
April through September 2010. As of September 2010, HUD officials told
us they were still in the process of aggregating on-site monitoring
results. As a result, program-wide data on the extent to which
grantees are complying with NSP 1 requirements are not readily
available. In the absence of centralized data, we reviewed the
monitoring results for 40 NSP 1 grantees overseen by the four CPD
field offices that we contacted during the course of our work (the
field offices are in Philadelphia, Pennsylvania; Columbus, Ohio;
Miami, Florida; and San Francisco, California).[Footnote 42] The on-
site monitoring found that 32 of the 40 grantees were in full
compliance with the requirements reviewed. However, CPD field staff
identified a total of 13 findings (potential deficiencies requiring
corrective action) for the other 8 grantees.[Footnote 43] Most of the
findings were related to financial management and accounting issues--
for example, deficiencies in accounting systems for personnel and
other administrative costs--and others related to oversight of
subrecipients. Additionally, 1 grantee did not clearly document that
two homebuyers received homebuyer counseling prior to obtaining their
mortgages loans. The same grantee did not take adequate steps to
ensure that a multifamily rental property would remain affordable to
income-eligible households. HUD required the 8 grantees to take
corrective actions on the findings and has an established process for
following up on the status of corrective actions. Because many
grantees are still completing their NSP 1 activities, following
through on this process will be key to ensuring grantees' compliance
with program requirements.
In cases where HUD identifies major compliance problems, it may freeze
a grantee's access to NSP 1 funds. HUD officials said they had not
needed to take this action as of September 2010.
Selected Grantees Complied with Key NSP 1 Requirements for Properties
We Reviewed:
We reviewed selected grantee's compliance with key NSP 1 requirements
for a nonstatistical sample of 32 NSP 1 properties. We selected eight
grantees in five states covering different geographic regions and
housing market conditions and reviewed records for four properties at
each grantee.[Footnote 44] Our review focused on requirements relevant
to activities (e.g., acquisition, rehabilitation, etc.) that had been
completed when we conducted our selection process.[Footnote 45] We did
not conduct a financial audit of the grantee or the funds expended on
the properties.
As shown in figure 8, we reviewed grantee compliance with key NSP 1
requirements and relevant internal controls concerning the use of
grant funds. Some of the requirements and controls were relevant to
all types of NSP 1 activities. Other requirements and controls applied
only to specific activity types.
Figure 8: Program Requirements and Controls We Reviewed:
[Refer to PDF for image: illustrated table]
All activities:
Property was located within a defined area of greatest need;
NSP 1 statutory or regulatory requirement.
Property was eligible for all relevant NSP 1 activities;
NSP 1 statutory or regulatory requirement.
Property address was included in grantee‘s Quarterly Performance
Report in DRGR, where applicable;
Other requirement or control[A].
Grantee had relevant procedures and checklists;
Other requirement or control[A].
Acquisition:
Acquisition price was at least 1 percent below the appraised value;
NSP 1 statutory or regulatory requirement.
Appraisal was conducted no more than 60 days prior to purchase offer;
NSP 1 statutory or regulatory requirement.
Proof of purchase and title transfer;
Other requirement or control[A].
Documentation of title insurance;
Other requirement or control[A].
Documentation that the required environmental review process was
completed prior to acquisition and had been certified by the
responsible official;
Other requirement or control[A].
Rehabilitation:
Documentation of work write-ups and appropriate approvals;
Other requirement or control[A].
Documentation of estimated and final rehabilitation costs;
Other requirement or control[A].
Final inspection reports documented that all rehabilitation work was
completed in a satisfactory manner;
Other requirement or control[A].
Demolition:
Documentation indicating the structure met the grantee‘s definition of
blighted;
NSP 1 statutory or regulatory requirement.
Evidence of the structure prior to demolition;
Other requirement or control[A].
Documentation of estimated and final demolition costs;
Other requirement or control[A].
Resale:
Provisions for primary residence and continued affordability were in
place[B];
NSP 1 statutory or regulatory requirement.
Property was resold at a price lower than the sum of the acquisition
and rehabilitation costs;
NSP 1 statutory or regulatory requirement.
Proof of purchase and title transfer;
Other requirement or control[A].
Income eligibility of homebuyers was determined and documented;
NSP 1 statutory or regulatory requirement.
Evidence that homebuyers attended at least 8 hours of homebuyer
counseling;
NSP 1 statutory or regulatory requirement.
Homebuyers obtained a traditional mortgage product;
NSP 1 statutory or regulatory requirement[C].
Financing (assistance provided to homebuyers for purchase or
rehabilitation):
Provisions for primary residence and continued affordability were in
place;
NSP 1 statutory or regulatory requirement.
In cases where down payment assistance was provided, the assistance
was less than 50 percent of the home‘s sale price;
NSP 1 statutory or regulatory requirement.
Income eligibility of homebuyers was determined and documented;
NSP 1 statutory or regulatory requirement.
Evidence that the homebuyers attended at least 8 hours of homebuyer
counseling;
NSP 1 statutory or regulatory requirement.
Homebuyers obtained a traditional mortgage product;
NSP 1 statutory or regulatory requirement[C].
Source: GAO.
[A] Other requirements and controls are based on those from other
relevant HUD programs (e.g., the CDBG and HOME Investment Partnerships
programs) or are practices consistent with federal internal control
standards.
[B] Some grantees addressed the primary residence requirement and
continued affordability restrictions by inserting clauses in purchase
and loan contracts. Under the continued affordability requirement for
NSP 1, grantees must ensure, to the maximum extent practicable and for
the longest feasible term, that the sale, rental, or redevelopment of
abandoned and foreclosed-upon homes and residential properties remain
affordable to individuals or families whose incomes do not exceed 120
percent (or 50 percent if property qualifies as low income) of area
median income.
[C] NSP 1 grantees are required to ensure that homebuyers obtain
mortgage loans from lenders that agree to comply with federal banking
regulator guidance for nontraditional mortgages. (Nontraditional
mortgages include loan products that allow borrowers to defer payment
of principal and, in some cases, interest. These features create the
potential for "payment shock" when the monthly payments adjust to a
fully amortized amount.) We did not review compliance with this
requirement directly because of potential variation in what would
constitute such assurance. However, we did determine whether
homebuyers obtained traditional mortgage products (e.g., 30-year fixed-
rate loans). For the 10 properties in our sample that had been
purchased by a homebuyer (either directly or from the grantee), the
homebuyer obtained a 30-year fixed-rate mortgage.
[End of figure]
We found no instances of significant noncompliance with the key NSP 1
requirements we reviewed for the 32 properties in our nonstatistical
sample.[Footnote 46] The results of our review cannot be generalized
to the total population of NSP 1 activities or grantees.
Table 1 shows the types of activities that had been completed at the
properties we reviewed. Seventeen of the 32 properties we reviewed had
been acquired and undergone rehabilitation. Six of those 17 properties
were resold by the grantees, and a seventh property was directly
acquired and rehabilitated by the homebuyer with NSP 1 assistance.
Table 1: Activities Reviewed for 32 NSP 1 Properties:
Activities reviewed: Acquisition by grantee;
Number of properties: 7;
Number of properties with homebuyer assistance: n/a[A].
Activities reviewed: Acquisition and rehabilitation by grantee;
Number of properties: 10;
Number of properties with homebuyer assistance: n/a[A].
Activities reviewed: Acquisition, rehabilitation, and resale by
grantee;
Number of properties: 6;
Number of properties with homebuyer assistance: 6.
Activities reviewed: Demolition by grantee[B];
Number of properties: 5;
Number of properties with homebuyer assistance: n/a[A].
Activities reviewed: Acquisition by homebuyer;
Number of properties: 3;
Number of properties with homebuyer assistance: 3.
Activities reviewed: Acquisition and rehabilitation by homebuyer;
Number of properties: 1;
Number of properties with homebuyer assistance: 1.
Activities reviewed: Total;
Number of properties: 32;
Number of properties with homebuyer assistance: 10.
Source: GAO.
[A] The properties in these categories were not yet at the resale
stage (or in the case of demolitions not intended for resale) when
homebuyer assistance is provided.
[B] Of the five demolished properties, only one was acquired using NSP
1 funds. The others were donated or blighted structures.
[End of table]
HUD Adapted an Information System for NSP 1, but Data on Program
Outputs Have Limitations:
HUD Modified DRGR and Provided Training and Guidance:
HUD made several modifications to DRGR--a system that was designed to
assist in managing Disaster Recovery grants--in order to collect
information for NSP 1 and subsequent rounds of the program. For
example, HUD modified system menus to include items unique to NSP,
such as land banks; removed inapplicable items; and enabled grantees
to designate whether properties had been sold or rented or were single-
or multifamily. HUD also provided training and technical assistance to
HUD field staff and NSP grantees that were unfamiliar with DRGR. These
efforts included one-on-one troubleshooting workshops, online
information, and a help desk. In addition, HUD technical assistance
providers held 10 joint sessions for HUD field staff and grantees on
DRGR.
As we have seen, grantees established action plans in DRGR, creating
projects and linking activities to them. For activities, HUD required
detailed information, including:
* a national objective (e.g., targeting low-, moderate-, and middle-
income persons),[Footnote 47]
* an activity type (e.g., acquisition, rehabilitation, demolition,
homeownership assistance, land banking, administration),[Footnote 48]
* a responsible organization (grantee or subrecipient carrying out the
activity), and:
* performance measures (e.g., output measures such as the number of
properties, housing units, or households or persons benefiting).
[Footnote 49]
In addition, HUD instructed grantees to enter projected and actual
output measures for each activity in DRGR and provided a predetermined
list of outputs for each activity type. HUD officials said they
considered the number of households benefiting to be the primary
output measure for NSP 1. Under each activity, HUD also required
grantees to enter a property address at the point they obligated funds
for a property.
HUD headquarters developed training and checklists for CPD field
office staff to use in reviewing the information grantees entered in
DRGR, including action plans and QPRs.[Footnote 50] The action plan
checklist required field offices to ensure that (1) grantees
established an activity for each responsible organization, (2)
activities met income-based national objectives, (3) every activity
type had a corresponding NSP eligible use, and (4) each multifamily
project had its own activity. The QPR checklist covers these areas and
requires field staff to monitor grantees' progress in meeting their
obligation deadlines and projected outputs.
HUD uses DRGR for a number of monitoring and reporting purposes. As
discussed earlier, HUD uses information grantees enter into DRGR to
monitor grantees' progress in meeting obligation deadlines and set-
aside requirements for low-income households. HUD also has put program-
wide and grantee-specific "snapshot" reports on its NSP Resource
Exchange Web site to provide financial and other information about NSP
1 to the public. In addition, the Administration has included DRGR-
generated information on NSP 1 in its monthly housing
scorecard.[Footnote 51] Thus far, the scorecard has reported three
output measures: (1) number of housing units constructed or
rehabilitated, (2) number of housing units demolished or cleared, and
(3) number of housing units for which direct homeownership assistance
was provided. In the October housing scorecard, HUD reported that by
March 2013, about 36,000 units would be rehabilitated or constructed;
8,000 units would be demolished; and direct homeownership assistance
would be provided for about 18,000 units.[Footnote 52] For reasons
described in the next section, it is difficult to know the number of
unique housing units or benefiting households these data represent,
and the data may tend to understate activity-specific outputs.
Reporting Flexibilities and Shortcomings in HUD Guidance Have Limited
the Usefulness of Output Data in DRGR:
Although DRGR has been a useful tool for monitoring grantees'
obligations, variation in the way grantees were allowed to classify
certain activities and select output measures in DRGR complicates the
analysis of program outputs. As under the CDBG program---which was
designed to give grantees substantial flexibility---HUD officials said
they permit NSP 1 grantees to group activities in different ways. For
example, HUD officials told us they instructed grantees to enter
information for property acquisitions under a rehabilitation activity
if the acquisition and rehabilitation are to be carried out by the
same responsible organization and are to be rehabilitated within a
short time frame. Accordingly, our review of DRGR information for 18
grantees identified a number of variations, including: (1) one
activity (classified as "rehabilitation") covering both the
acquisition and rehabilitation of housing units; (2) two activities
(one classified as "acquisition" and the other as "rehabilitation")
for acquiring and rehabilitating units; and (3) one activity
(classified as "acquisition for land banking") for acquiring,
demolishing, and land-banking housing units.[Footnote 53] Due to these
variations, totaling outputs for individual activity types can result
in undercounting program outputs (because grantees that combine two or
more activities report the associated outputs under a single activity
type) and totaling outputs across multiple activity types can result
in double counting program outputs (because grantees that do not
combine activities may report the same outputs under multiple activity
types).[Footnote 54]
HUD officials said that the variation did not affect their ability to
monitor grantees' compliance with program requirements but
acknowledged that it complicated analysis of program outputs. HUD
officials said they were in the process of identifying duplicate
property addresses in DRGR, which will allow them to report total
numbers of NSP 1 housing units and benefiting households without
double counting these outputs. Undercounting is likely primarily an
issue with the acquisition activity type, because, as previously
noted, HUD allows grantees to combine acquisition and rehabilitation
under a rehabilitation activity in some circumstances. Additionally,
HUD officials said they were more focused on reporting outputs for
"end uses" of NSP 1 funds, such as the number of housing units
rehabilitated and demolished and the number of benefiting households.
However, we found that grantees did not always group acquisition and
rehabilitation under a rehabilitation activity. For instance, contrary
to HUD's instructions, we identified several instances in which
grantees combined acquisition and rehabilitation under an acquisition
activity. HUD officials indicated they had found similar cases in
their reviews of DRGR information. As a result, totaling output data
for rehabilitation activities may understate actual outputs. HUD
training and its QPR review checklist developed for CPD field staff
responsible for reviewing activity data in DRGR do not adequately
address this issue. In particular, the checklist that field staff use
to review grantees' QPRs does not require field staff to determine
whether grantees grouped activities in accordance with HUD guidance.
As a result, HUD staff may not be detecting errors in how grantees are
classifying activities, which can negatively affect the accuracy and
reliability of NSP 1 output data and therefore the usefulness of these
data for monitoring program progress and results.
We also found variation in how grantees selected output measures for
different activity types. For example, one grantee selected "number of
properties" as its sole output measure for the majority of its
activities (including acquisition and rehabilitation) and did not
select numbers of households benefiting, which HUD told us it
considers the program's primary output measure. For the same
activities, another grantee selected multiple output measures--number
of properties, housing units, households benefiting, and persons
benefiting. HUD officials said they explained the output measures
grantees should input into DRGR as part of DRGR training and strongly
encouraged grantees to select all applicable output measures. However,
the documentation HUD staff cited as guidance for grantees, including
the NSP 1 Notice, did not provide specific instructions on which
output measures should be linked to each activity type. Additionally,
while HUD's QPR checklist requires field staff to determine whether
grantees used the "right" output measures for each activity, it does
not specify which output measures should be entered for each activity
type. As a result, HUD may not be collecting consistent output data
for each activity, which impairs the agency's ability to summarize and
report on program outputs in an accurate and consistent manner.
GAO's Standards for Internal Control emphasizes the need for federal
agencies to collect reliable information with which to manage their
programs and to review the integrity of performance measures.[Footnote
55] Due to limitations in HUD guidance to grantees and field office
staff, HUD lacks assurance that these standards are being met.
A Planned Assessment of NSP Outcomes Will Focus on NSP 2 but Will Also
Include NSP 1-assisted Areas:
HUD is in the process of contracting for an assessment of NSP that
will evaluate the impact of the program by tracking outcomes in the
neighborhoods where NSP-assisted activities took place.[Footnote 56]
HUD's Office of Policy Development and Research (PD&R) is overseeing
the assessment in collaboration with CPD. The assessment will focus
primarily on the impacts of NSP 2--which funds the same types of
activities as NSP 1--but will also incorporate the results of NSP 1
where the two rounds of the program overlap. According to PD&R staff,
many neighborhoods receiving NSP 2 funds also received NSP 1 funds.
HUD does not plan to conduct a separate assessment of NSP 1 outcomes,
in part because HERA did not provide funding for an evaluation.
However, ARRA, which authorized NSP 2, did provide funds for program
evaluation.
HUD has established outcome measures for NSP. In its May 2009 Notice
of Funding Availability (NOFA) for NSP 2, HUD established short- and
long-term outcome measures as guidance for grantees in their
application process. CPD staff said they would also apply these
measures to NSP 1. The outcome measures are:
* short-term: (1) arrest decline in home values and (2) reduce or
eliminate vacant and abandoned properties; and:
* long-term: (1) increase sales in target areas and (2) increase
median property values.
PD&R staff said they have been working closely with CPD to develop the
objectives and scope of the assessment. They said the assessment will
track conditions and trends in NSP-assisted neighborhoods in relation
to the short-and long-term outcome measures. They also will attempt to
identify suitable comparison areas (e.g., similar areas that did not
receive assistance) to demonstrate the impact of NSP, but expect this
to be more challenging. Additionally, PD&R staff noted that DRGR was
structured to track spending by DRGR activity rather than by property,
making it difficult to determine the amount of NSP 1 funds that were
spent in a particular geographic area. PD&R staff plan to obtain and
review additional records from NSP 2 grantees to determine the amount
of funds spent on NSP 2 properties and may collect similar information
for NSP 1 properties in areas served by both rounds of the program.
HUD is anticipating that the assessment will be completed no earlier
than January 2014. Given that NSP 1 and NSP 2 share a common set of
eligible activities, the results of the assessment should be useful
for understanding the impact of both rounds of the program on assisted
neighborhoods.
Conclusions:
NSP 1 provided a mechanism for state and local governments to mitigate
the destabilizing effects of mortgage foreclosures, but HUD and
grantees faced a number of implementation challenges, including the
program's tight time frames and the limited capacity of some grantees
to undertake real estate activities. HUD took actions to help grantees
meet these challenges though guidance, training, and technical
assistance. Additionally, HUD established internal control procedures
to mitigate risks and promote compliance with program requirements.
Our work suggests that these efforts helped grantees obligate funds in
a timely manner, adopt strategies appropriate to their communities,
and follow program rules. Nevertheless, because many NSP 1 activities
have not been completed, continued HUD oversight will be required to
ensure that any implementation and compliance problems are identified
and addressed in an effective manner. In particular, HUD will need to
follow through on its efforts to ensure that grantees are taking
corrective actions on findings from on-site monitoring visits.
As NSP 1 and other rounds of NSP progress, assessing program outputs
and outcomes will become increasingly important. HUD took a number of
important steps to collect key program data in DRGR, including
information on grantees' activities and performance measures. However,
variation in how grantees entered this information make it difficult
to accurately summarize program outputs without undercounting (in the
case of activity-specific outputs) or overcounting (in the case of
program-wide outputs). While HUD has developed a method to resolve the
overcounting issue, limitations in HUD's written guidance to grantees
and field staff may be contributing to variation in data entry and
impairing HUD's ability to accurately summarize program outputs.
Existing guidance does not require field staff to review whether
grantees properly grouped activities and does not specify which output
measures grantees should select for each type of activity. Similarly,
HUD has not provided grantees with written guidance specifying the
output measures they should select for different activity types. In
the absence of such guidance, HUD lacks assurance that it is
collecting consistent information from NSP 1 grantees and that it is
summarizing and reporting program outputs in an accurate manner.
Because grantees involved in NSP 2 and 3 also are using DRGR,
addressing limitations in written guidance would benefit HUD's
analysis of output data for all rounds of NSP.
Recommendations for Executive Action:
To ensure the consistency of data collection in DRGR and enhance the
reporting of program outputs for all rounds of NSP, we recommend that
the Secretary of HUD take the following two actions:
* Update the QPR review checklist to include reviews of whether
grantees are (1) grouping activities in accordance with HUD
instructions and (2) selecting the appropriate output measures for
different activities.
* Issue written guidance to NSP grantees on the output measures they
should select for different activities.
Agency Comments and Our Evaluation:
We provided a draft of this report to HUD for its review and comment.
We received written comments from the Deputy Assistant Secretary for
Grant Programs, CPD, that are reprinted in appendix II. We also
received technical comments from HUD that we incorporated into the
final report as appropriate. In its written comments, HUD stated that,
in general, the draft report accurately represented its efforts to
implement NSP 1. HUD also agreed to implement both of the report's
recommendations in the first half of 2011.
We are sending copies of this report to the appropriate congressional
committees, the Secretary of Housing and Urban Development, and other
interested parties. In addition, the report will be available at no
charge on GAO's Web site at [hyperlink, http://www.gao.gov].
If you or your staffs have questions about this report, please contact
me at (202) 512-8678 or sciremj@gao.gov. Contact points for our
Offices of Congressional Relations and Public Affairs may be found on
the last page of this report. GAO staff who made key contributions to
this report are listed in appendix III.
Signed by:
Mathew J. Scirè:
Director, Financial Markets and Community Investment:
[End of section]
Appendix I: Objectives, Scope, and Methodology:
Congress created the Neighborhood Stabilization Program (NSP), which
is administered by the Department of Housing and Urban Development
(HUD), to help reduce the number of foreclosed and abandoned
properties and restore depressed local housing markets. The Housing
and Economic Recovery Act of 2008 (HERA) authorized the first phase of
this program (NSP 1), providing $3.92 billion in grant funds to states
and local governments.[Footnote 57] HERA mandated that GAO report on
whether NSP 1 funds were being used in a manner consistent with
criteria set forth in the act. To respond to this mandate we examined
(1) grantees' progress and challenges in meeting NSP 1 obligation time
frames and income-targeting criteria, (2) HUD's actions to mitigate
program risks and ensure grantees' compliance with key NSP 1
requirements, and (3) HUD's efforts to collect program data and assess
program performance.
Grantee Progress and Challenges:
To examine NSP 1 grantees' progress and challenges in meeting HERA
obligation time frames and income-targeting criteria, we reviewed data
from HUD's Disaster Recovery Grant Reporting (DRGR) system on grantee
obligations, analyzed states' methods for distributing funds to other
entities for obligation, and interviewed representatives from selected
NSP 1 grantees (see selection criteria below). We also reviewed
relevant HERA provisions and program guidance from the Department of
Housing and Urban Development (HUD), including the NSP 1 Notice.
We analyzed DRGR data as of June 24, 2010, and October 1, 2010, to
calculate the percentage of their grant allocation each grantee had
obligated at both points in time. Using the October data, we also
calculated the overall percentage of NSP 1 funds obligated by grantees
and the proportion of funds obligated for different activities. To do
these calculations, we used the approach set forth in the NSP 1
Notice, which does not differentiate between obligation of grant funds
and program income in assessing progress toward obligation deadlines.
[Footnote 58] For the analysis of obligations by activity type, we
combined the data for acquisition, rehabilitation, and disposition
because--as described in the body of this report--grantees sometimes
group two or more of these activities together under a single activity
in DRGR.
Because state grantees could reallocate their funds to other entities
(e.g., local governments) for obligation as long as the funds targeted
areas of greatest need, we examined the methods states used to
distribute their NSP 1 funds. We did this by analyzing information in
DRGR on states' HUD-approved NSP 1 action plans to determine whether
the states used required criteria for determining areas of greatest
need and the types of distribution methods they used, including
reallocation by competition, reallocation by formula, and through
state-administered programs. Additionally, by analyzing DRGR data as
of June 24, 2010, we also determined the number of states that
reallocated funds to local governments that had also received NSP 1
funds directly from HUD. We also determined the number of local
governments that received NSP 1 funds both directly from HUD and from
state reallocations, as well as the amount and percentage of program
funds this represented. The funding amounts and percentages are
approximate because in two states it was unclear from HUD data how
reallocations were divided between two recipients, only one of which
was a direct grantee. We included the entire reallocations for these
two cases ($3.6 million in total) in our nationwide total of $175.8
million as it seemed likely that the funds were used by the direct
grantee.
We also assessed the reliability of the DRGR information (discussed
later in this section). We concluded that the data we used were
sufficiently reliable for our purposes.
We interviewed 11 grantees to discuss their approaches to implementing
NSP 1, the progress they had made, and challenges they faced in
meeting obligation time frames and income-targeting criteria. We
selected these grantees to cover areas with substantial foreclosure
problems and provide some variation in geographic location, housing
market conditions, and grantee types. We also focused on grantees that
had made some progress in implementing their NSP 1 programs. The
grantees were: State of Indiana, City of Fort Wayne, and City of
Indianapolis (Indiana); Lee County, Orange County, and City of Tampa
(Florida); and State of Nevada, Clark County, City of Henderson, City
of Las Vegas, and City of North Las Vegas (Nevada). We had previously
spoken with these grantees during the first phase of our work, which
we conducted in April through December 2009.
We also interviewed officials from four HUD field offices about
grantees' progress and challenges in implementing NSP 1. We selected
these field offices to provide some geographic variation and because
they were responsible for overseeing some of the grantees that we
visited on-site. The field offices were located in Columbus, Ohio;
Miami, Florida; Philadelphia, Pennsylvania; and San Francisco,
California.
We also spoke with staff from NeighborWorks® America and the National
Community Stabilization Trust to obtain their perspectives on
grantees' progress and challenges.
HUD's Actions to Mitigate Risks and Ensure Compliance:
To examine the steps HUD has taken to ensure grantees' compliance with
key NSP 1 requirements and mitigate program risks, we reviewed NSP 1
statutes and regulations and HUD's front-end risk assessment (FERA)
for the program. We reviewed relevant documentation and interviewed
HUD officials about the agency's efforts to hire additional staff to
address gaps in capacity to oversee NSP 1 grantees (a program risk
identified in the FERA) and to provide training and technical
assistance to grantees. Additionally, we interviewed staff from the 11
grantees described above about HUD's training, technical assistance,
and oversight processes for NSP 1 to obtain their perspectives on
these efforts.
We reviewed HUD's internal controls for NSP 1, including HUD's
guidance and procedures for monitoring grantee compliance with key
program requirements. In addition, we reviewed results of on-site
monitoring of NSP 1 grantees conducted by the four field offices we
contacted. More specifically, we reviewed the results for the 40
grantees for which the field offices had completed monitoring letters
as of September 15, 2010.[Footnote 59] In addition, we interviewed HUD
headquarters and field office officials on the status and results of
their on-site monitoring.
Additionally, we visited 8 grantees to conduct limited tests of
compliance with key program requirements. We purposefully selected the
grantees to cover different eligible NSP 1 activities, types of
grantees, and geographic areas where characteristics of housing
markets may vary. The grantees we selected were: Collier County and
Lee County (Florida); City of Columbus and City of Dayton (Ohio);
Maricopa County and City of Phoenix (Arizona); Philadelphia,
Pennsylvania; and Prince William County, Virginia. We primarily
focused on grantees with completed activities, which intentionally
overrepresented grantees that had made the most progress in
implementing NSP 1 at the time of our review and ensured we would
review a variety of activities for compliance with NSP 1 key
requirements. Further, to avoid overburdening grantees, we focused our
site visits on grantees that, on the basis of information from HUD,
were not receiving extensive technical assistance or being monitoring
on-site by HUD staff.
At each grantee, we reviewed records for four properties, for a total
of 32 properties in our sample. To select the four properties to be
reviewed at each grantee, we obtained a list of NSP 1 properties with
completed activities and applied the following steps: (1) We
identified the group of properties with the largest number of
completed activities (e.g., acquisition, rehabilitation, resale,
etc.). We grouped the remaining properties by type of activity. (2) We
allocated the number of properties to be selected from the different
groupings to intentionally capture properties with the most completed
activities, while also covering the grantee's full range of
activities. (3) We randomly selected properties from each group up to
the predetermined allocation.
We developed a checklist to review compliance with program
requirements and internal controls for the following types of
activities for each property: acquisition, rehabilitation, demolition,
financing, and resale. We selected the checklist components by
reviewing federal internal control standards and analyzing NSP 1
requirements in statutes, regulations, and relevant HUD program
guidance. Figure 8 in the body of this report includes the
requirements and controls we reviewed.
We are not able to generalize the results of our compliance testing to
all NSP 1 grantees or activities. We did not conduct financial audits
of the 8 grantees or the 32 properties.
HUD Data Collection and Performance Assessment:
To examine HUD's efforts to collect program data and assess the
reliability of NSP data, we reviewed HUD Office of Inspector General
(OIG) audits of the DRGR system. We interviewed OIG staff about these
audits and reviewed documentation on HUD's actions to address the
OIG's recommendations. We also reviewed documentation on HUD's
modifications to DRGR and efforts to train system users. This
information included HUD training materials and guidance to grantees
and field staff on entering and reviewing data on program activities
and output measures. We also interviewed HUD headquarters, HUD field
office, and grantee staff with responsibilities for inputting or
monitoring DRGR data. We also conducted reasonableness checks of DRGR
data to identify any missing or erroneous data and by interviewing
knowledgeable HUD officials to ensure we interpreted the data
correctly. For the purpose of this and the first objective, we
concluded that the data we used were sufficiently reliable for our
purposes.
Additionally, we reviewed the NSP output data that HUD posted on its
NSP Web site and included in the Administration's monthly housing
scorecard. Furthermore, we examined DRGR action plans and Quarterly
Performance Reports (for the second quarter of calendar year 2010) for
18 grantees--the 11 cited previously plus others we visited on-site
(described in the previous section of this appendix). Our examination
focused on the consistency with which grantees set up activities and
selected program output measures in DRGR when entering their action
plans and performance reports, and the extent to which they followed
applicable HUD guidance. Finally, we reviewed federal internal control
standards relevant to data quality and controls.
To obtain information on HUD's plans to assess NSP 1, we reviewed the
scope of work for a planned HUD study of NSP program outcomes. In
addition, we interviewed HUD staff knowledgeable of the study,
including officials from HUD's Office of Policy Development and
Research and Office of Community Planning and Development.
We conducted this performance audit from January to December 2010, in
accordance with generally accepted government auditing standards.
Those standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe
that the evidence obtained provides a reasonable basis for our
findings and conclusions based on our audit objectives.
[End of section]
Appendix II: Comments from the Department of Housing and Urban
Development:
U.S. Department Of Housing And Urban Development:
Office Of Community Planning And Development:
Washington, DC 20410-7000:
December 6, 2010:
Mr. Matthew J. Scirè:
Director, Financial Markets and Community Investment:
US Government Accountability Office:
441 G St., NW:
Washington, DC 20548:
Dear Mr. Scirè:
Thank you for the opportunity to comment on the draft report entitled
"Neighborhood Stabilization Program ” HUD and Grantees Are Taking
Actions to Ensure Program Compliance but Data on Program Outputs Could
be Improved."
In general, the Department finds that the draft report accurately
represents the Department's efforts to implement the Neighborhood
Stabilization Program (NSPI) since its enactment as part of the
Housing and Economic Recovery Act of 2008 (HERA). Editorial comments
on the draft report have been informally conveyed to your staff via a
conference call on December 2, 2010. This letter provides a response
to the Recommendations for Executive Action identified in the draft
report.
The first recommendation calls for HUD to "Update the QPR review
checklist to include reviews of whether grantees are (1) grouping
activities in accordance with HUD instructions and (2) selecting the
appropriate output measures for different activities." The Department
will revise the QPR checklist in draft format and issue it to
Community Planning and Development (CPD) field staff for testing by
March 31, 2011. Based on comments received from the field testing
effort, HUD intends to issue a final revised QPR checklist not later
than June 30, 2011.
The second recommendation call for HUD to "issue written guidance to
NSP grantees on the output measures they should select for different
activities." The Department concurs with the recommendation and will
issue guidance on this point not later than March 31, 2011.
The Department appreciates the willingness of your staff to
accommodate CPD staff schedules and workload considerations over the
duration of this engagement as we have worked to implement not only
NSP I but also NSP2 and NSP3. Please contact me at (202) 708-2111 if
you have any questions regarding this matter.
Sincerely,
Signed by:
Yolanda Chavez:
Deputy Assistant Secretary for Grant Programs:
[End of section]
Appendix III: GAO Contact and Staff Acknowledgments:
GAO Contact:
Mathew J. Scirè, (202) 512-8678 or sciremj@gao.gov:
Staff Acknowledgments:
In addition to the individual named above, Steven Westley, Assistant
Director; Allison Abrams; Emily Chalmers; Mya Dinh; Ying Long; John
McGrail; Meredith Moore; Jasminee Persaud; and Carl Ramirez made key
contributions to this report.
[End of section]
Footnotes:
[1] P.L. 110-289. The American Recovery and Reinvestment Act of 2009
(P.L. 111-5) provided an additional $2 billion in NSP funds (referred
to as NSP 2) and changed several aspects of the program. Later, the
Dodd-Frank Wall Street Reform and Consumer Protection Act (P.L. 111-
203) enacted in July 2010 provided an additional $1 billion in funding
for the program (referred to as NSP 3). This report focuses on NSP 1.
[2] The CDBG program was first established in 1974 and is one of the
federal government's largest and most flexible community and economic
development programs. CDBG funds may be used to rehabilitate housing,
improve infrastructure, provided job training, and fund other
community-determined projects.
[3] Funds went to 50 states, 4 insular areas, Puerto Rico, the
District of Columbia, and 253 county and city governments. The four
insular areas are American Samoa, Guam, Northern Marianas, and the
Virgin Islands.
[4] One grantee was included in both of these groups.
[5] NSP 1 grantees may redevelop demolished or vacant properties for
residential and nonresidential uses, commercial use, or mixed
residential and commercial use. The American Recovery and Reinvestment
Act limits redevelopment of such properties under NSP 2 and NSP 3 to
residential purposes.
[6] 73 Fed. Reg. 58330 (Oct. 6, 2008).
[7] 74 Fed. Reg. 29225 (June 19, 2009).
[8] CPD requires grantees to submit consolidated plans covering up to
four of its major grant programs: CDBG, HOME Investment Partnerships,
Emergency Shelter Grants, and Housing Opportunities for Persons with
AIDS.
[9] Generally, entitlement communities are (1) principal cities of
metropolitan areas, (2) other metropolitan cities with populations of
at least 50,000, and (3) qualified urban counties with populations of
at least 200,000 (excluding the population of entitled cities).
[10] Some examples include the building of community centers or
firehouses, or repairing sewage treatment plants.
[11] HERA required HUD to establish a formula for allocating NSP 1
funds to states and eligible local governments no later than 60 days
after the law's enactment and to distribute these funds no later than
30 days after establishing the formula.
[12] For details on HUD's methodology for allocating funds to state
and local governments, see 73 Fed. Reg. 58343-9 (Oct. 6, 2008).
[13] In response to disasters, Congress may appropriate additional
funding for CDBG to provide disaster recovery grants to rebuild
affected areas and provide seed money to start the recovery process.
[14] Under the CDBG program, grantees' activities must meet one of
three national objectives. They must either (1) benefit low-and
moderate-income persons, (2) eliminate slums or blight, or (3) meet
urgent needs. Under NSP, there is one national objective grantees'
activities must meet--they must benefit low-, moderate-, and middle-
income persons. Low-income is defined as 50 percent of AMI or lower,
moderate-income as 51 percent to 80 percent of AMI, and middle-income
as 81 percent to 120 percent of AMI. The middle-income category
applies to NSP but not CDBG, which defines low-and moderate-income
persons as those being at or below 80 percent of AMI.
[15] HUD Office of Inspector General (OIG), HUD's Disaster Recovery
Grant Reporting System Can Collect the Basic Information Needed to
Monitor the Neighborhood Stabilization Program, 2009-FW-0001
(Amended), June 25, 2009.
[16] HUD OIG, Review of Selected Controls within the Disaster Recovery
Grant Reporting System, 2009-DP-0007, September 30, 2009.
[17] As discussed previously, a grantee's deadline depended on the
date it received NSP 1 grant funds from HUD, so obligation deadlines
varied among grantees. HUD officials confirmed that the majority of
obligation deadlines expired in September 2010.
[18] HUD made NSP 1 grant allocations to 309 units of state and local
government. However, there are only 307 grant agreements because the
State of Colorado and City of Colorado Springs (Colorado) entered into
a cooperative grant agreement, as did Clark County and the City of Las
Vegas (Nevada).
[19] As indicated in the NSP 1 Notice, HUD did not distinguish between
obligation of grant funds and program income (e.g., income a grantee
received from reselling a property) in assessing compliance with the
18-month requirement. Therefore, grantees met the requirement as long
as they obligated grant funds and program income in an aggregate
amount equal to their NSP 1 allocation. HUD officials said they took
this approach so as not to penalize grantees that progressed faster in
implementing NSP 1, potentially resulting in larger amounts of program
income.
[20] In its audits of the State of Kansas and the City and County of
Denver, the OIG concluded that these grantees had improperly obligated
NSP 1 funds by reporting its funds as obligated without linking these
funds to a specific address or household, as required. See HUD OIG,
The State of Kansas Did Not Properly Obligate Its Neighborhood
Stabilization Program Funds, 2010-KC-1006, August 20, 2010; and HUD
OIG, The City and County of Denver Did Not Properly Obligate and
Report NSP 1 Funding, 2010-DE-1006, September 17, 2010. However,
several other OIG compliance audits found that the grantees generally
complied with NSP 1 requirements.
[21] Depending on the grantee's NSP 1 performance and the amount of
unobligated funds, HUD may enter into a memorandum of agreement with
the grantee designed to enable use of the funds for the purposes
intended in the NSP 1 Notice. See 75 Fed. Reg. 52772 (Aug. 27, 2010).
[22] Under CDBG, HUD provides funds to metropolitan cities and urban
counties, known as entitlement communities, and provides funds to
states for distribution to nonentitlement communities. States may not
use CDBG funds directly. However, the NSP 1 Notice spells out the
various ways in which states may use NSP 1 funds directly for
activities, including using their own employees, procuring
contractors, or providing grants through nonprofit subrecipients,
among other direct uses. In prior work, we examined how the states
distribute CDBG funds. See GAO, Community Development Block Grants:
Entitlement Communities' and States' Methods of Distributing Funds
Reflect Program Flexibility, [hyperlink,
http://www.gao.gov/products/GAO-10-1011] (Washington, D.C.: Sept. 15,
2010).
[23] The percentage (9) and dollar amount ($176 million) are estimates
because in two cases it was unclear exactly how much NSP 1 funds were
reallocated to direct local government NSP 1 grantees. However, we
included the amount in the estimates--totaling $3.6 million--as it
seemed likely that the funds were used by the direct grantee.
[24] For 3 states, it was unclear from their action plans which method
was used to distribute funds within the state. Our review of states'
NSP action plans included all 50 states and the District of Columbia
but did not include Puerto Rico.
[25] A soft second loan is a second mortgage with payments that are
forgiven, deferred, or subsidized in some fashion, generally until
resale of the mortgaged property.
[26] As discussed in appendix I, we spoke with 18 grantees (2 states,
6 counties, and 10 cities).
[27] The two grantees that did not obligate their total grant amounts
by October 1 had obligated 94 percent and 97 percent, respectively.
[28] Harriet Newburger, Federal Reserve Bank of Philadelphia,
"Acquiring Privately Held REO Properties with Public Funds: The Case
of the Neighborhood Stabilization Program." REO & Vacant Properties:
Strategies for Neighborhood Stabilization, A Joint Publication of the
Federal Reserve Banks of Boston and Cleveland and the Federal Reserve
Board, 2010.
[29] HUD Secretary Announces National First Look Program To Help
Communities Stabilize Neighborhoods Hard-Hit By Foreclosure, September
1, 2010.
[30] 75 Fed. Reg. 41225 (July 15, 2010).
[31] The purpose of a FERA is to detect conditions that may adversely
affect the achievement of program objectives and to provide reasonable
assurance that program goals, including compliance with applicable
laws and regulations, will be met. The FERA is mandatory for any new
HUD program with a funding level totaling $10 million or more and for
certain substantially revised programs. HUD uses the FERA in
accordance with principles of risk assessment outlined in OMB Circular
A-123, to identify and analyze risks, from both internal and external
sources, which may affect the ability of the agency to meet objectives.
[32] The eight modules were: (1) Eligible Uses, Activities, and
Properties; (2) Pre-Acquisition Considerations; (3) Post-Acquisition
Considerations; (4) Disaster Recovery Grant Reporting System; (5)
Financing Issues; (6) Program Administration; (7) Land Banking and
Demolition; and (8) Eligible Use Scenarios. They are available at
[hyperlink,
http://www.hud.gov/offices/cpd/communitydevelopment/programs/neighborhoo
dspg/training/index.cfm].
[33] Aggregate purchases for NSP are defined as all of the properties
that an NSP grantee purchases with its NSP funds.
[34] We spoke with a number of grantees we contacted during our review
about the impact of HUD's definition changes on their NSP programs.
This work was done in coordination with another GAO team reviewing
issues of incomplete or "abandoned" foreclosures. See GAO, Mortgage
Foreclosures: Additional Mortgage Servicer Actions Could Help Reduce
the Frequency and Impact of Abandoned Foreclosures, GAO-11-93
(Washington, D.C.: Nov. 15, 2010).
[35] The term "abandoned" was originally defined as a property that
had been foreclosed upon and was vacant for at least 90 days. However,
HUD explained that this definition limited opportunities for acquiring
properties in a strategic and timely manner. For example, the
requirement that the property had to be vacant for at least 90 days
left out properties abandoned by owners, but where tenants were still
in place. Therefore, HUD expanded the definition of abandoned to
include vacant and nonvacant properties (a) for which no mortgage or
tax payments have been made by the property owner for at least 90
days, (b) for which a code enforcement inspection determined that the
property is not habitable and the owner has taken no corrective
actions within 90 days of the notification of the deficiencies, or (c)
subject to a court-ordered receivership or nuisance abatement related
to abandonment.
[36] Local technical assistance providers serviced the following
locations: Northern and Southern California, Northern and Southern
Florida, Illinois, Indiana, Ohio, Georgia, New England, and Michigan.
HUD news release: HUD No. 09-159 on August 26, 2009.
[37] The field offices are in Jacksonville and Miami, Florida;
Columbus, Ohio; Detroit, Michigan; San Francisco and Los Angeles,
California; and Las Vegas, Nevada. The Las Vegas field office reports
to the San Francisco regional office.
[38] The HUD OIG identified some inconsistency in the way HUD field
offices applied the special conditions. While some field offices
considered past performance only in the CDBG program, others also
considered grantees' performance in other HUD programs--for example,
HOME Investment Partnerships and the Supportive Housing Program. See
HUD Office of the Inspector General Audit Report 2010-CH-0001, March
29, 2010.
[39] HUD officials told us that HUD headquarters staff reverified
compliance with the checklist for the vast majority of grantees.
[40] HUD Notice CPD-09-04: Implementing Risk Analyses for Monitoring
Community Planning and Development Grant Programs in FY 2010 and 2011,
Issued 8-24-2009.
[41] As of the end of September 2010, the OIG had issued capacity
audits of 22 NSP 1 grantees and found that 12 grantees generally had
sufficient capacity, while 8 grantees needed to improve their capacity
to administer the program. Further, 2 grantees did not have the
capacity to effectively and efficiently administer NSP funding.
Examples of lack of capacity included inadequate staffing levels and
policies, procedures, and internal controls.
[42] In total, the four field offices are responsible for overseeing
86 NSP 1 grantees, 50 of which they monitored on-site. We reviewed the
monitoring results for the 40 grantees for which the field offices had
completed "monitoring letters" as of September 15, 2010. Sixty days
after completing an on-site visit, HUD field staff send a monitoring
letter to the grantee discussing, among other things, the field
staff's conclusions, any monitoring findings, and any corrective
actions required.
[43] Monitoring letters containing findings specify corrective actions
the grantee should take and the time frame in which the grantee should
respond.
[44] See appendix I for a detailed description of our methodology.
[45] By focusing on grantees with completed activities, our sample
intentionally overrepresented those grantees that had made the most
progress in implementing NSP 1 at the time of our review. Also, while
our methodology for selecting properties at each grantee focused on
properties with the most completed activities, some planned activities
for some of the properties in our sample had yet to be completed. We
did not include those planned activities in our compliance review.
[46] For one property, however, the grantee did not receive final
supervisory approval of an otherwise complete environmental review
until after the property had been acquired.
[47] As previously noted, under the CDBG program, grantees' activities
must meet one of three national objectives, but under NSP grantees'
must meet one national objective--they must benefit low-, moderate-,
and middle-income persons.
[48] The activity type menu in DRGR includes a number of variations of
acquisition, rehabilitation, and other major activity types. For ease
of presentation, we generally refer only to the major activity type
(e.g., "acquisition" or "rehabilitation").
[49] Outputs are the products and services delivered by a program.
[50] HUD headquarters staff set policy for managing and monitoring NSP
1 activities, and CPD field staff conduct the actual oversight of
individual grantees under their purview.
[51] The scorecard contains data on key housing market indicators and
includes performance metrics for the Administration's housing recovery
efforts. Data are presented by HUD and the Department of the Treasury
in "The Obama Administration's Efforts to Stabilize the Housing Market
and Help American Homeowners" brochure, October 2010.
[52] HUD data indicate that the number of rehabilitated units is
greater than the number of newly constructed units.
[53] We reviewed action plans and QPRs for the second quarter of
calendar year 2010 for the 18 grantees we interviewed or visited on-
site. (See appendix I for a list of these grantees and our selection
criteria.)
[54] To illustrate, for a grantee that used a "rehabilitation"
activity to report outputs for the acquisition and rehabilitation of
five housing units, DRGR would show no properties under "acquisition"
(understating the actual number of acquisitions by five). For a
grantee that set up both "acquisition" and "rehabilitation" activities
for acquiring and rehabilitating five units, DRGR would show five
units under each activity. Consequently, totaling across these
activities would result in double counting.
[55] See GAO, Standards for Internal Control in the Federal
Government, [hyperlink,
http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.:
November 1999).
[56] Outcomes describe the intended result of carrying out a program
or activity.
[57] P.L. 110-289. The American Recovery and Reinvestment Act of 2009
(P.L. 111-5) provided an additional $2 billion in NSP funds (referred
to as NSP 2) and changed several aspects of the program. The Dodd-
Frank Wall Street Reform and Consumer Protection Act (P.L. 111-203),
enacted in July 2010, provided an additional $1 billion for the
program (referred to as NSP 3). This report focuses on NSP 1.
[58] That is, we considered grantees to have met the obligation
requirement as long as they obligated grant funds and program income
in an aggregate amount equal to their NSP 1 allocation. HUD officials
said they took this approach so as not to penalize grantees that
progressed faster in implementing NSP 1, potentially resulting in
larger amounts of program income.
[59] In total, the four field offices are responsible for overseeing
86 NSP 1 grantees, 50 of which they monitored on-site. Sixty days
after completing an on-site visit, HUD field staff send a monitoring
letter to the grantee discussing, among other things, the field
staff's conclusions, any monitoring findings, and any corrective
actions required.
[End of section]
GAO's Mission:
The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting
its constitutional responsibilities and to help improve the performance
and accountability of the federal government for the American people.
GAO examines the use of public funds; evaluates federal programs and
policies; and provides analyses, recommendations, and other assistance
to help Congress make informed oversight, policy, and funding
decisions. GAO's commitment to good government is reflected in its core
values of accountability, integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each
weekday, GAO posts newly released reports, testimony, and
correspondence on its Web site. To have GAO e-mail you a list of newly
posted products every afternoon, go to [hyperlink, http://www.gao.gov]
and select "E-mail Updates."
Order by Phone:
The price of each GAO publication reflects GAO‘s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO‘s Web site,
[hyperlink, http://www.gao.gov/ordering.htm].
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional
information.
To Report Fraud, Waste, and Abuse in Federal Programs:
Contact:
Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]:
E-mail: fraudnet@gao.gov:
Automated answering system: (800) 424-5454 or (202) 512-7470:
Congressional Relations:
Ralph Dawn, Managing Director, dawnr@gao.gov:
(202) 512-4400:
U.S. Government Accountability Office:
441 G Street NW, Room 7125:
Washington, D.C. 20548:
Public Affairs:
Chuck Young, Managing Director, youngc1@gao.gov:
(202) 512-4800:
U.S. Government Accountability Office:
441 G Street NW, Room 7149:
Washington, D.C. 20548:
