Rural Housing: Changing the Definition of Rural Could Improve
Eligibility Determinations (03-DEC-04, GAO-05-110).
Rural America has become more diverse: technology and the spread
of suburbia have linked rural areas to urban areas, resulting in
diminished distinctions between the two. The Rural Housing
Service (RHS) applies statutory requirements for eligibility that
may not reflect changes in rural areas or best determine which
areas qualify for its housing programs. GAO's objectives included
assessments of how eligibility is defined for RHS programs and
how changes in the current eligibility requirements might impact
the RHS mission of meeting rural housing needs.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-05-110
ACCNO: A14156
TITLE: Rural Housing: Changing the Definition of Rural Could
Improve Eligibility Determinations
DATE: 12/03/2004
SUBJECT: Census
Comparative analysis
Eligibility criteria
Eligibility determinations
Population statistics
Rural housing programs
Arizona
Maryland
California
Ohio
Massachusettes
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO Product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
******************************************************************
GAO-05-110
United States Government Accountability Office
GAO
Report to the Chairman, Subcommittee on Housing and Community Opportunity,
Committee on Financial Services, House of Representatives
December 2004
RURAL HOUSING
Changing the Definition of Rural Could Improve Eligibility Determinations
a
GAO-05-110
December 2004
RURAL HOUSING
Changing the Definition of Rural Could Improve Eligibility Determinations
What GAO Found
RHS determines which areas will be eligible (that is, defined as rural)
for its programs by applying requirements in the Housing Act of 1949, as
amended. While the definition largely focuses on population-generally up
to 20,000- certain communities must also be "rural in character," not part
of metropolitan statistical areas (MSA-which is defined as a county or
counties associated with a city or urbanized area that has a population of
at least 50,000), or demonstrate a serious lack of mortgage credit for
lower-and moderate-income families. Also, a "grandfather" clause allows
communities with populations over 10,000 to retain eligibility if they
become part of an MSA-and still meet the "rural in character" criterion
and not exceed 25,000 in population.
These eligibility requirements resulted in dissimilar determinations for
what appeared to be similar areas. For example, in visits to five states
GAO found that applying the grandfather clause enabled certain communities
within MSAs to retain eligibility while other communities within the same
MSAs remained ineligible even though they met current "rural" and
population criteria. In addition, GAO analysis of nationwide data found
that RHS made more than 1,300 communities with populations of 10,000 or
below eligible that were within or contiguous to areas that had
populations of 50,000 or more.
GAO identified alternatives to retaining the MSA, grandfather, and credit
requirements in the Housing Act of 1949. Because MSAs contain both urban
and rural areas and have increased substantially in both size and number
in recent decades, they may not be good determinants of urban-rural
distinctions. According to the 2000 census, about half the nation's rural
population lives in MSAs. An alternative measure would be to use the
Census Bureau's urbanized areas and urban clusters, which are densitybased
measures that provide finer-scale information and could help RHS better
and more consistently make eligibility determinations for areas with
similar population and characteristics. By dropping the MSA requirement,
"grandfathering" also could be phased out, allowing RHS to make
determinations that focused on current conditions rather than prior
eligibility. In addition, a 1997 USDA report found that lack of credit in
rural areas was no longer a serious problem; rather, a lack of income and
ability to pay the mortgage were greater issues. Therefore, the
requirement to demonstrate a lack of credit no longer appears to be
relevant to meeting housing needs in rural America.
United States Government Accountability Office
Contents
Letter
Results in Brief
Background
While Eligibility for Rural Housing Programs Focuses on Population,
"Rural in Character" Requirement Is Open to Interpretation RHS Targets Areas for Its Programs through Funding, Application, and Strategic Planning Processes
A Majority of RHS Loans and Grants in Five States Went to Smaller Communities, but Our Review Found Varying Interpretations of Eligibility
Changes to Components of the Definition of Rural Could Address
Inconsistent Treatment of Similar Communities Conclusions Matters for Congressional Consideration Agency Comments and Our Evaluation
1 2 5
13
19
22
32 40 41 41
Appendixes
Appendix I: Scope and Methodology 44
Appendix II: USDA Comment Letter 47
Appendix III:
GAO Contacts and Staff Acknowledgements 49 GAO Contacts 49 Staff Acknowledgements 49
Tables Table 1: Table 2: Table 3:
Table 4:
USDA's Rural Development Programs Have Different
Eligibility Requirements
Communities Must Meet Different Requirements for
Eligibility Based on Their Populations
Fourteen Communities Designated Exempt by the
Congress in the 2004 Appropriations Legislation (P.L.
108-199), with Population Levels
Communities by Population Range in Five States That
Received RHS Housing Loans and Grants in Fiscal Years
1998-2004
9 14
15
23
Figures Figure 1: Congress Increased Population Limits 7 Figure 2: MSAs and Urbanized Areas/Urban Clusters Have Different Boundaries 12
Contents
Figure 3: Road Serving As Eligible Area Boundary outside Hagerstown, Maryland 18 Figure 4: City Line of Brookside, Ohio, Divides Eligible from Ineligible Area 19 Figure 5: Belpre, Ohio, Is Part of the Parkersburg, West Virginia-Ohio, Urbanized Area 25 Figure 6: Circleville and Washington Court House Relative to Columbus, Ohio MSA 27 Figure 7: Lamont and Taft Relative to Bakersfield, California, within the Bakersfield MSA 28 Figure 8: Both South Hadley and Belchertown Have Less Densely
Settled Areas within the Springfield MSA 29 Figure 9: South Hadley and Belchertown Rural Areas 30 Figure 10: Only a Small Portion of Casa Grande Is Densely
Populated 31 Figure 11: MSAs Encompassed More Territory in 2000 Than in 1970 34 Figure 12: Eliminating MSA Criterion Could Allow Circleville To Be Considered for Eligibility 35 Figure 13: Using Urban Clusters, Rural Portions of Both Belchertown and South Hadley Could Be Eligible 36 Figure 14: Population Density in and around Westminster, Maryland, Varies 37 Figure 15: Taft, California, Could Be Eligible under Density-based Criteria 39
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. However, 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.
A
United States Government Accountability Office
Washington, D.C. 20548
December 3, 2004
The Honorable Robert W. Ney
Chairman
Subcommittee on Housing and Community Opportunity
Committee on Financial Services
House of Representatives
Dear Mr. Chairman:
The federal government has provided housing assistance to eligible
residents of rural America since the 1930s. Housing assistance is still
available for such residents; however, the rural America of 2004 is different
from the rural America of the 1930s, and the population levels and other
criteria used to determine whether a community is rural and which areas
are eligible for housing assistance have evolved.1 As a result of changing
economic and demographic conditions, and greater proximity to urban
centers, many of the distinctions between rural and urban life have blurred.
Advances in transportation, computer technology, and
telecommunications, along with the spread of suburbia, have linked many
rural areas to urban areas. Farming is no longer the primary economic
activity of rural areas. However, as in urban areas, the need for decent,
safe, and affordable low-income housing remains strong in rural areas.
The Housing Act of 1949 authorized new rural lending programs to farmers,
which were administered by the Rural Housing Service's (RHS)
predecessor, the Farmers Home Administration, within the U.S.
Department of Agriculture (USDA). Over the decades, Congress changed
the requirements for rural housing program eligibility-for example, by
changing population limits-and rural housing programs have evolved to
serve low-and moderate-income people of all occupations. RHS now
facilitates homeownership, develops rental housing, and promotes
community development through loan and grant programs in rural
communities. Also, the eligibility requirements in the Housing Act of 1949
have been amended. The current definition of rural considers factors such
as whether an area is contained in a standard metropolitan statistical area
(MSA), is "rural in character," and "has a serious lack of mortgage credit for
1For the purpose of this report, we use area and community interchangeably to refer to any open country, place, town, village, or city.
lower and moderate-income families."2 Concerned about whether the current definition of "rural" used for RHS housing programs is still appropriate, you requested that we evaluate RHS's eligible service areas. The objectives of this report are to assess (1) how rural is defined and implemented for RHS housing programs; (2) how RHS targets areas for program delivery; (3) the populations of the rural areas served by RHS housing programs; and (4) how a change in the current definition of rural would likely impact the RHS mission of meeting rural housing needs.
To meet these objectives, we interviewed officials from USDA, the Department of Housing and Urban Development (HUD), the Census Bureau (Census), and the Office of Management and Budget (OMB) for information about housing and other community programs for rural areas and to understand how rural is defined for these programs and how populations are counted and classified. We also reviewed the legislative history of the statutory definition of rural for the purposes of rural housing programs. We analyzed population data from Census and RHS program data from USDA to determine what areas receive RHS loans and grants, and the effects on those areas if the eligibility boundaries were to change. To obtain a better understanding of how rural areas vary across the country, we conducted case study visits in Arizona, California, Maryland, Massachusetts, and Ohio. We did not attempt to derive our own definition of "rural," but relied on USDA, Census, and OMB criteria for determining rural areas. Appendix I contains a full description of our scope and methodology.
We conducted our review from January 2004 through October 2004 in accordance with generally accepted government auditing standards.
Results in Brief
Section 520 of the Housing Act of 1949, as amended, defines rural for most RHS housing programs and, based on instructions promulgated by the national office, state and local (together, field) offices, determines the boundaries to delineate eligible areas from ineligible areas-a task field office officials acknowledged is time-consuming, based on judgment, and
2Section 520 of the Housing Act of 1949, as amended (42 U.S.C. 1490). An MSA is a statistical entity consisting of a county or counties associated with a core city that has a population of at least 50,000. Since the 1990 Census, OMB has not used the term "standard MSA."
can be problematic.3 The statutory definition generally identifies eligible rural areas as those with populations up to 20,000 and defines "rural" and "rural areas" as any open country or any place, town, village, or city that is not part of or associated with an urban area. Specifically, there are several population levels at which communities may be determined eligible, but as a community's population increases, the statute imposes additional requirements that include being "rural in character" (a concept that is not defined in the statute), having a serious lack of mortgage credit, or not being located within an MSA. Certain communities with populations up to 25,000 may be "grandfathered in" based on prior eligibility if they still met the "rural in character" and lack of credit criteria. Additionally, specific communities are automatically eligible for RHS funding, irrespective of their population, because they are so identified in the statute or in annual appropriations legislation. In implementing the statute, USDA's instructions give its field offices flexibility in determining which communities are eligible but require a re-evaluation of eligibility determinations every 3 to 5 years, depending on whether or not the communities are within an MSA. USDA's instructions also provide descriptions of "open country," which is defined in the statute, and instructs field offices on how to consider contiguous areas for eligibility purposes.4 Additionally, while the field offices may use population as the primary factor in determining eligibility, field office officials said that drawing the eligibility boundaries required an element of judgment because "rural in character" is open to interpretation-even with the overall national guidance on the statute and review of census populations, MSA standards, maps, aerial photographs, and visits to communities.
While Section 520 determines how communities become eligible, RHS relies on other processes to target areas of greatest need for program delivery: funding set asides, funding allocations, application reviews, and
3The definition of rural applies to most RHS housing programs. Two programs-farm labor housing loans and grants-are not restricted by any rural eligibility definition. Housing built for farm workers through these programs can be located anywhere in the country, including cities. The definition also can be waived in situations where an RHS loan was made on a house that currently is no longer in an eligible area but needs a repair loan to remove major health and safety hazards. While RHS also administers community facilities programs for rural areas, Congress imposed a different definition of rural to determine eligibility. This report focuses on RHS' housing programs.
4According to the RHS handbook, "contiguous areas" can be two or more towns, villages, cities, or places. Field offices can consider them separately for eligibility determinations "if they are not otherwise associated with each other, and their densely settled areas are not contiguous."
state-level strategic plans. RHS has regulations, guidance, and procedures in place to cover the programmatic requirements to target housing resources to specific places or special needs groups. However, in the states we visited, economic and demographic factors limited the agency's ability to make loans, and sometimes grants, in the specified areas or to the groups targeted. For example, Arizona state officials told us that the lack of qualified applicants, insufficient infrastructure, and the poor quality of the housing stock have kept RHS from meeting its goals for the colonias (rural communities located within 150 miles of the Mexican border that often lack basic infrastructure) and tribal lands.
Analysis of USDA's eligibility areas and program funding data from the five states we visited found that, from October 1998 through April 2004, RHS's housing loans and grants generally went to residents of communities with populations at or below 10,000. About 60 percent of the 29,000 housing loans and grants approved for these five states during that period went to communities with populations at or below 10,000. Another 25 percent went to communities with populations of 10,001 to 20,000, and almost 15 percent went to communities with populations over 20,000, which included higher-populated communities that received exemptions to the eligibility requirements. Our analysis of RHS eligible areas nationwide compared to Census data found approximately 1,300 examples where communities with populations at or below 10,000 were within or contiguous with urban areas that had populations of 50,000 or more. The statute states that eligible communities cannot be part of or associated with an urban area. Some field staff determinations of eligibility in these cases might be questionable as some of these communities, despite their low populations, might not be considered rural, and thus, eligible. We also found rural communities with populations exceeding 10,000 that were directly impacted by the MSA restriction and the grandfather clause. Because MSAs are county-based and may contain both urban and rural areas, the MSA restriction and the grandfathering of certain communities resulted in some communities being eligible while others with similar demographic profiles were ineligible.
Changes to the way rural is defined might allow RHS to treat communities with similar characteristics more consistently. The methods vary from changing population limits to dropping the "lack of credit" requirement. Decreasing population limits would reduce the number of areas eligible for the programs, potentially allowing RHS to concentrate services in fewer areas. In contrast, increasing population limits would increase the number of eligible areas, but potentially dilute RHS's ability to serve many of the rural areas that it targets. Another alternative, which might help RHS better
serve its clients, is to eliminate the MSA requirement. As OMB pointed out, many counties included in MSAs contain both urban and rural areas. Using an alternative measure, such as Census's urbanized areas and urban cluster classifications as a guide, could help RHS better draw boundaries around rural areas because density-based measures provide finer-scale information. Furthermore, if MSAs were removed from the eligibility criteria, RHS officials could make determinations for more communities based on population data and "rural" character. Additionally, eligible communities within MSAs would not need to be "grandfathered" based on previous eligibility, which essentially gives these communities an advantage over similar though ineligible towns located in MSAs. Finally, eliminating the "lack of credit" requirement would have little or no effect on targeting of priority areas or impact RHS eligibility determinations because eligibility for RHS programs are based on income levels.
To improve eligibility determinations in rural housing programs, this report contains three matters for congressional consideration. Congress may wish to consider (1) eliminating the MSA criterion and recommending that RHS use density measures as a basis for its eligibility decisions, (2) phasing out the practice of "grandfathering" communities that experienced changes in eligibility because of inclusion in an MSA, and (3) eliminating the "lack of credit" requirement.
In commenting on a draft of this report, USDA generally agreed with our matters for congressional consideration. USDA stated that our report articulates how the use of MSAs has resulted in disparate treatment of some communities. USDA stated that applying a density-based measure may have merit but that further study is needed to properly define such a measure for nationwide application. We concur with this position. In addition, the agency anticipates that the "lack of credit requirement" could be removed with no detriment to RHS housing programs.
Background
Federal housing assistance in rural America dates back to the 1930s when most residents of rural areas worked on farms, and rural areas were generally poorer than urban areas. Subsequently, Congress authorized separate housing assistance for rural areas and gave USDA responsibility for administering it. Rural housing programs are now part of USDA's rural
development mission area and RHS is responsible for rural housing and community facilities programs.5
Congress Has Changed the Definition of Rural to Reflect Population Changes
When the Housing Act of 1949 authorized rural lending programs in USDA, eligibility was limited to persons who lived in dwellings on land capable of producing at least $400 worth of agricultural products annually. In 1962, Congress expanded eligibility to include low-income elderly people living in rural areas. In 1965, Congress first defined "rural" and "rural areas" for the purpose of rural housing program eligibility as communities with populations below 2,500 or from 2,500 but not over 5,500 and "rural in character." Congress expanded the eligibility categories and population limits for rural housing programs several more times and imposed additional requirements.6 In 1970, Congress increased the population threshold to 10,000 as long as the community was "rural in character." Then, in 1974, Congress increased the population limit to 20,000 but stipulated that communities with populations from 10,001 to 20,000 could not be within an MSA and had to have a serious lack of mortgage credit for their lower-and moderate-income families. In 1983, Congress modified the statute by permitting communities to retain eligibility even though they had become part of an MSA. Congress has updated this "grandfather" clause several times to take into account updated census data. Figure 1 provides a timeline of changes to population limits and other eligibility requirements.
5RHS administers most federal rural housing programs. While HUD's programs are generally for urban residents and communities, HUD administers several housing programs targeted specifically to assist Native Americans and residents of colonias, which are rural, mostly unincorporated, communities within 150 miles of the U.S.-Mexican border that often lack basic infrastructure. Residents may also apply for other HUD programs, ranging from public housing to Federal Housing Administration (FHA) loans. However, RHS loans, which have lower interest rates and low or no down payment requirements, are generally more affordable than FHA loans. Rural residents who are veterans can also apply for loans guaranteed by the Department of Veterans Affairs.
6Eligibility for RHS's community facilities loans and grants programs is based on a different statutory definition. For the community facilities programs, "rural" means a city, town, or unincorporated area that has a population of not more than 20,000.
Figure 1: Congress Increased Population Limits
Source: GAO.
RHS Offers a Wide Array of Housing Services
RHS helps low-to moderate-income residents (including the elderly and farm laborers) of rural communities by providing loans and grants for single-family homes and apartments. Eligible rural residents may obtain
direct or guaranteed loans to purchase single-family homes.7 Additionally,
RHS offers a housing preservation grant program, which provides grants to
sponsoring organizations for the repair or rehabilitation of low-and very
7Direct housing loans enable low-or very low-income rural households
(less than 80 percent of area median income and less than 50 percent of
area median income, respectively) to
purchase, build, repair, renovate, or relocate houses. There is
no required down payment and loan terms can be for up to
38 years. Guaranteed loans are made by a bank or another private lender
rather than RHS; RHS guarantees
repayment if the borrower defaults. Applicants for guaranteed loans must have incomes below 115 percent
of area median incomes.
low-income housing. Very low-and low-income rural residents can also participate in a program in which they help in the construction of their
homes in order to reduce their cost of ownership. While many RHS programs promote homeownership in eligible rural areas, RHS rental
assistance programs also are available to a variety of
populations. Direct or
guaranteed loans also are available for the purchase or development of affordable rural rental housing for low-income people, the elderly, persons with disabilities, and farm laborers. Finally, RHS also offers rental
assistance for eligible households with incomes too low to pay the RHS subsidized rent from their own resources.
From
October 1, 1998, through September 30, 2004, RHS issued more than 434,000 direct and guaranteed housing loans and grants valued at $28.7
billion.8 All of these housing
programs, with the exception of the farm labor housing programs, are required to follow the statutory definition of
rural-as stated in Section 520 of the Housing Act of 1949, as amended-to
determine which communities are eligible. Housing funded by the farm
labor programs can be built anywhere in the country as long as it
is intended for farm laborers.
Definition of
Eligible Area USDA's rural development mission area administers a number of related for RHS Housing Programs community and business development programs that have eligibility Differs from Other USDA requirements that differ from RHS's. Most of the other definitions, also
statutorily imposed, cover areas with population limits ranging from lessRural Development than 2,500 to 50,000. Table 1 summarizes key rural development programs Programs
and lists population and other requirements used for determining eligibility.
8Actual federal outlays are much lower because guaranteed housing
loans are based on subsidy costs and projected losses that are less
than loan levels.
Table 1: USDA's Rural Development Programs Have Different Eligibility
Requirements
Rural Development
Agency Rural Development Program Area Eligibility Requirements
Rural Housing Service RHS Single-family Direct and Guaranteed Area with a
population of 10,000 or below and rural in character.
(RHS)
Loans, Single-familyHome Repair Grants, Multifamily Housing,Self Help
Housing Technical Assistance, Housing Preservation Grants, and Rural
Rental Housing
If not contained in a metropolitan statistical area (MSA) and has a
serious lack of mortgage credit with a population in excess of 10,000 but
not in excess of 20,000. If classified as rural prior to Oct. 1, 1990,
even if in an MSA, and (1) has a population in excess of 10,000 but not in
excess of 25,000 and (2) is rural in character. This designation to remain
in effect until receipt of 2010 census data.
RHS Farm Labor Housing Loans and There are no geographical or population
Grants limitations.
RHS Rental Assistance Payments With the exception of farm labor units
and Rural and developing sites for
Housing Site Loans farm labor housing, rural areas are as
defined in the first box
above.
RHS Community Facilities Direct Towns with populations up to 20,000
and people.
Guaranteed Loans and Grants
Business and Industry Any areas other than (1) a
Rural Direct and city or town with a
population greater
than 50,000, and (2) the
Business-Cooperative Guaranteed Loans, urbanized area contiguous
and
Services (RBS) Rural Business Enterprise adjacent to such a city or
Grants, Rural town.
Business Opportunity
Grants, and Section
9006 Renewable Energy
Systems and
Energy Efficiency Program
RBS Rural Cooperative Development Any areas other than (1) a city or
Grants town with a population greater
than 50,000, and (2) the urbanized
area contiguous and
adjacent to such a city or town.
RBS Rural Economic Development Area with a population of 2,500 or
Loans and below.
Grants
RBS All areas of a state not within the
Intermediary Relending Program outer boundary of any city with
a population of 25,000 or more,
according to the last census.
Rural Utilities Services (RUS)
Electric LoansAny area with a population up to 2,500. Once an area or town
has a loan, that area will always be eligible for subsequent loans.
RUS Telecommunications Infrastructure Loans Area with a population of
5,000 or below.
RUSDistance Learning and Telemedicine Loans Area with a population of
20,000 or below. and Grants
RUS Rural Broadband Loans and GuaranteesArea with a population of 20,000
or below and not located in a standard MSA.
(Continued From Previous Page)
Rural Development
Agency Rural Development Program Area Eligibility Requirements
Water and Environmental Programs (WEP)
Water and Waste Disposal Loans, Grants, and Guarantees,Emergency Community
Water AssistanceGrants,Empowerment Zones/Enterprise Communities,Rural
Economic Area Partnership,Colonias Grants,Native American Grants,Alaskan
Villages Grants,Solid Waste Management Grants, andTechnical Assistance and
Training Grants
Area with a population of 10,000 or below.
Source: USDA.
OMB and Census Provide Geographic Area Classifications
For the past 50 years, OMB, in consultation with Census and other experts inside and outside government, has defined MSAs to provide consistency in
collecting, tabulating, and publishing federal statistics for geographic areas
in the United States.9 An MSA is
a statistical entity consisting of a county or
counties associated with a core urbanized area of 50,000 or more people.
MSAs are not intended to be
urban-rural classifications, and MSAs can contain both urban and rural areas. OMB states that MSAs may not be suitable for use in program funding formulas and cautions against using them for nonstatistical purposes.
OMB uses Census's "urbanized areas" and "urban clusters" to help define MSAs and micropolitan statistical areas.10 Urbanized areas and urban clusters are statistical areas that do not necessarily follow political
boundaries. An urbanized area is a continuously built-up area with a
population of at least 50,000, comprising one or more places and adjacent
9After each decennial census, OMB uses census data to define new
MSAs. OMB issued the
most recent MSA list in 2003, based on the 2000 census. For details
on the new standards, see U.S. Government Accountability Office,
Metropolitan Statistical Areas: New Standards and Their Impact on Selected
Federal Programs, GAO-04-758, (Washington, D.C.: June 14, 2004).
10The micropolitan statistical area, was created for the
2000 census; it too is county-based and includes densely settled
area with a population of 10,000-50,000. The micropolitan
statistical area is built around urban clusters. Census defines a densely settled area as an
area with a population density of 1,000 people per square mile and includes
surrounding areas with a density of at least 500 people per square mile.
densely settled areas. An urban cluster consists of densely settled territory that has at least 2,500 people but fewer than 50,000 people, a definition which encompasses a variety of communities, some of which are eligible
for rural housing and development programs. Collectively, urbanized areas
and urban clusters are referred to as urban areas.
Census defines "rural area" by exclusion; that is, it views all areas that
it did not already identify as urbanized areas or urban clusters as "rural." Using this Census definition, based on the 2000 census, 59 million Americans (or
20 percent of the population) reside in rural areas, with slightly more than half of them residing within MSAs. Figure 2 shows MSAs and urbanized areas/urban clusters, using data derived from the 2000 census.
While Eligibility for Rural Housing Programs Focuses on Population, "Rural in
Character" Requirement Is Open to Interpretation
The Housing Act of 1949, as amended, defines rural for most RHS housing
programs largely based on
population limits but also includes other criteria
such as being "rural in character," not being part of an MSA, and having a serious lack of mortgage credit for lower-and moderate-income families. Additionally, Congress has exempted certain communities from
complying
with the eligibility requirements and, in its annual appropriations, designated the communities that are eligible for rural housing programs without having to meet the usual eligibility requirements. A USDA handbook provides guidance to state field office staff on implementing the definition. However, because "rural in character" is open to interpretation, RHS field officials acknowledged that determinations are based on judgment and can be problematic.
The Housing Act of
1949, as Amended, Defines Eligibility along Population Thresholds
While the Housing Act of 1949, as amended, generally identifies eligible rural communities as areas not part
of or associated with urban areas and having populations of 20,000 or below, it appends other requirements to areas with populations above 2,500. Communities with populations from
2,501 to 10,000 also must be "rural in character." Communities with
populations from 10,001 to 20,000 are not required to show that they are "rural in character" but they cannot be within MSAs and must have a serious lack of mortgage credit for lower-and moderate-income families.
Additionally, the statute's "grandfather" clause allows communities that
have experienced population growth or inclusion within an expanding
MSA to remain eligible for RHS programs (see table 2). The clause enables communities to remain eligible if they meet the following criteria:
o
Were classified as "rural" before October 1, 1990, but determined not to
be "rural" as a result of data received from or after the 1990 or 2000 censuses,11
o Can demonstrate that they are still "rural in character,"
11When the "grandfather" clause was first added to the Housing Act of 1949
in 1983, it allowed communities to retain eligibility if they were
"rural" prior to the receipt of data from
(or after) the 1980 census. However, grandfathered communities must still meet the "rural in character" and lack of credit requirements, and have a population not
in excess of 25,000 to retain eligibility.
o
Have a serious lack of mortgage credit for lower-and moderate-income residents,12 and
o Have a current population in excess of 10,000 but not in excess of
25,000.
Table 2: Communities Must Meet Different Requirements for Eligibility Based on
Their Populations Eligibility by population
"Rural" or "rural area" equates to open country or any place, town,
village, or city which is not part of or associated with an urban area
"Rural in character"
Not within a standard MSA
Has serious lack of mortgage credit for lower-and moderate-income families
2,500 or fewer residents X
2,501 to 10,000 X X
10,001 to 20,000 X X
Current grandfather clause:
community with population from
10,001 to 25,000, was classified as
"rural" or "rural area" before Oct. 1,
1990, but determined not "rural" or X
"rural area" from or after the 1990 or
2000 censuses; eligible until 2010
census.
Source: GAO analysis of Section 520 of the Housing Act of 1949.
Note: Since the 1990 census, OMB has not used the term "standard MSA."
Exemptions to the Eligibility Requirements Found in Statute and Annual
Appropriations Legislation
A number of communities are exempt from meeting the population requirements of the Housing Act of 1949, as amended. The act made Pajaro, California, Guadalupe, Arizona, and Plainview,
Texas, permanently eligible.
Altus, Oklahoma, is eligible until receipt of data from the 2010 census. In
addition, in annual appropriations legislation over the past several
years,
Congress has designated other communities as meeting the eligibility requirements. These congressional designations typically are valid for one
12Section 5.3 of a USDA Handbook (HB-1-3550) states "there
is a serious of lack of mortgage credit readily available to families throughout rural America at rates and terms comparable
to those offered by the agency. Therefore, Agency officials do
not need to determine if
there is a serious lack of mortgage credit available when determining whether an area is rural or
in reviewing rural area designations."
fiscal year. However, some
communities, such as Casa Grande, Arizona, have been congressionally designated for several years (see table 3).
Table 3: Fourteen Communities Designated Exempt by the Congress in the
2004 Appropriations Legislation (P.L. 108-199), with Population Levels
Community Designated in the FY 2004 Appropriations Population, Based on
Legislation (P.L. 108-199) 2000 Census
Lawrence County, Ohio 62,319
Havelock, North Carolina 22,442
Portsmouth, Ohio 20,909
Binghamton, New York 47,380
Vestal, New York 26,535
Ithaca, New York 29,287
Casa Grande, Arizona 25,224
Clarksdale, Mississippi 20,645
Coachella, California 22,724
Salinas, California 151,060
Watsonville, California 44,265
Hollister, California 34,413
Carolina, Puerto Rico 168,164
Kinston, North Carolina 23,688
Source: P.L. 108-199 and 2000 census data.
Handbook Provides Guidance to Field Staff on Implementing Definition
USDA's handbook, HB-1-3550, contains implementing guidance to help RHS field staff designate eligible areas, and defines the flexibility field offices have in implementing RHS housing programs. The handbook provides information on assessing open country, population, and contiguous areas,
but is
silent on what "rural in character" means. For instance, open country
includes areas that are "not part of or associated with an urban area" and
"separated by open space from any adjacent densely populated urban area." Open space could include undeveloped land or agricultural land, but not rivers, parks, or commercial developments. The handbook also
specifies that population figures used by field staff must come from Census. Additionally, it instructs that two or more communities "that are contiguous may be considered separately for a rural designation if they are not otherwise associated with each other, and their densely settled areas are not contiguous." As previously noted, the handbook also states that
there is a serious lack of mortgage credit throughout rural America and thus agency officials do not need to make this statutorily required determination for areas they have already defined as rural.
The national office leaves it up to the field offices to determine
which areas are eligible in the states for which they are responsible. In addition to
applying the national guidance on open country, population, and contiguous areas, field staff must ensure that any site eligible for a rural
housing loan or grant must
o be "modest,"
o meet minimum standards regarding water and wastewater systems, and
o meet street and access requirements, such as having direct access to a
maintained hard-surfaced road.
Field offices are required to review all areas under their jurisdiction every 3-5 years to identify areas that no longer qualify as rural. More specifically, for eligible communities within MSAs and in areas of rapid growth, field offices are required to make the review every 3 years. Field staff may review current maps and aerial photographs, tap local staff knowledge of the area, and drive through the communities to determine where to draw the eligibility boundaries. Eligibility maps are supposed to be updated after every review.
Determining What Is "Rural
in Character" and Drawing Eligibility Boundaries Require Element of Judgment
According to RHS headquarters officials, rural America is diverse and the concept of "rural in character" is broad and will vary by state and circumstance. Senior headquarters officials told us that, given diverse
geography and settlement patterns in rural areas across the country and the intent to implement RHS programs at the state and local level, they depend
on the knowledge of field staff in making eligibility
determinations. In fact, they expect rural to be interpreted differently across the states. The
national office also defers to the field offices in determining what densely settled areas are and the extent to which areas are contiguous with each other.
While the handbook does not specifically address what "rural in character"
means, field staff use the other components of the handbook to help determine "rural in character" for
their jurisdictions. Field staff told us that drawing the eligibility boundaries required an
element of judgment because
"rural in character" was open to interpretation-even with the national
explanations of "open country" and "contiguous areas" and review of census
populations, MSA standards, maps, aerial photographs, and
visits to the boundaries. In December 2002, the national office provided guidance
on features that could be used as boundaries, including streets, highways, streams, lakes, railroads, and political boundaries, and requested that the
field offices redraw all boundaries in order to allow headquarters to create an automated mapping system for the public to determine whether a property was within an eligible area. Even with this guidance, delineating
eligible areas from ineligible areas is a task field staff acknowledged is
time-consuming, requires judgment, and can be problematic. For example, one local office official told us that he believes that "drive-through" visits to
boundary areas largely are subjective. Another senior
field office official told us that making adjustments to eligible areas is not an exact science. The official believed that local supervisors have the best knowledge of
local conditions and the rural character of the areas and he trusted the
local supervisor's judgments in recommending boundary line changes.
However, even when local supervisors fully understand the local conditions and rural character of an area, finding a way to equitably decide
on a boundary line sometimes becomes problematic. For instance, field staff in Maryland told us that in response to the December 2002 national
guidance, they chose to no longer use natural boundaries such as rivers or
mountains as eligibility boundaries for Maryland communities. Maryland now only uses roads as
boundaries. Figure 3 shows a new boundary, a road outside Hagerstown, Maryland, that divides the eligible area on the left
from the ineligible area on the right. RHS local office officials told us that
the "road only" criteria forced them to find the nearest public road to a populated section of Hagerstown, which happens to go through farmland. The result is that apparently similar rural areas received different designations.
Figure 3: Road Serving As Eligible Area Boundary outside Hagerstown,
Maryland
Source: GAO.
Figure 4 shows an area near the city line in Brookside, Ohio, where
the city line divides the eligible from the ineligible area. While the Maryland example shows that two parts of a road through farmland received different designations, the Brookside example shows that using a political boundary also does not necessarily equate to a readily discernible urban-rural difference.
Figure 4: City Line of Brookside, Ohio, Divides Eligible from Ineligible Area
Source: GAO.
RHS Targets Areas for Its Programs through Funding, Application,
and Strategic Planning Processes
RHS determines eligible areas by following requirements of the Housing Act of 1949, as amended, but relies on other processes to target areas for program delivery: funding set asides and reserves, funding allocations, application reviews, and state-level strategic plans. RHS has regulations, guidance, and procedures in place to ensure that it meets programmatic requirements to target housing resources to specific places or groups, and low- and moderate-income residents of rural areas in general. However, in the states we visited, economic and demographic factors limited RHS's
ability to make loans in the specific areas or to the groups targeted.
RHS Targets Low-and Moderate-Income Areas with the Greatest Housing Needs
Although RHS must respond to statutory, regulatory, and other procedural
requirements to make its loans and grants in certain areas, it also uses
planning procedures to further direct funds to areas of greatest need-rural
areas with low-and moderate-income populations that lack decent, safe, and affordable housing. Funds are targeted through set asides, allocations, application reviews, and state-level strategic plans.
o
RHS national and state offices are required to set aside funds to ensure
that single-family and multifamily programs serve specific populations
or rural areas. Areas targeted by set-asides are generally underserved, and have low-and moderate-income residents who lack decent, safe,
and affordable housing. For instance, Section 509(f) of the Housing Act of 1949, as amended, requires set asides in RHS's single-family direct (Section 502) as well as single-family housing repair loans and grants
(Section 504) and multifamily rural rental housing (Section 515) programs for the top 100 underserved counties and colonias. Funds are
also reserved and earmarked in RHS's single- and multifamily housing
programs at both national and state levels for many targeted groups such as homeless
applicants, or for specific areas such as
empowerment zones, or purposes such as self-help housing.13
o
The RHS national office annually targets remaining funds that have not been set aside or reserved to states. Using a formula, RHS weights the
allocations toward states with areas that have higher percentages of substandard
housing and lower-income populations. States may
be directed or given the option to suballocate funds to district or county offices. When performing a suballocation, states must use the same allocation formula used by the national office.
o
Once states receive funding, they approve applications annually for single-family and multifamily housing loans and grants in the state.
While requirements differ slightly depending on the program, applicants
must meet very low-to moderate-income requirements. Further, properties purchased by all applicants must be in eligible rural areas as
determined by the statutory definition.
o
States also are required to develop strategic plans under the Federal Agricultural Improvement and Reform Act of
1996 (P.L. 104-127). Unlike other targeting methods, which occur annually, state strategic planning is done on a 5-year cycle. These plans target program resources towards
areas with greatest housing needs. The act stipulates that USDA must give priority to communities with the smallest populations and lowest
per capita income. State plans that we reviewed generally used census or other economic data to assess and identify counties with poor
housing conditions, high vacancy rates, and high concentrations of lowand
moderate-income populations. Critical housing needs are identified
through surveys, forums, or interviews with interest groups.
13The self-help housing program enables low-income
people to work together to build each other's homes. It combines
the single-family direct (502) loan program with a grant program for
nonprofits to oversee the construction.
Economic and Demographic Factors Often Limit Ability to Deliver
Products to Areas with Greatest Housing Need
Although many funding directives are built into the system and RHS
prioritizes program activity, other factors may determine whether RHS can actually
make loans or grants in desired areas. In the five states we visited,
we found that the states had difficulty making loans in areas targeted in their strategic plans as having the greatest housing needs. For instance,
o
Ohio targets its Appalachian region because it is home to some of the lowest-income and highest-poverty areas in the state. However,
state officials explained that they have difficulty closing loans in this region because of issues related to road systems, sparse populations, high unemployment, unusable housing stock, and geography. For example, industry officials told us that they would not consider building in many
of the communities along the Ohio River floodplain due to higher insurance costs.
o
Arizona's strategic plan targets colonias, tribal lands, La Paz County on
the California border, and Pinal County between Phoenix and Tucson. However, RHS local officials said that they have difficulty meeting their
targets in these areas, and that foreclosure rates in
these areas are often higher because they are dealing with low-income populations that have
poor credit and other money management challenges. For example,
Arizona field office officials told us that they have had problems serving the less-populated areas in the colonias and Indian tribal lands due to
the lack of qualified applicants, insufficient infrastructure such as
roads, and the poor quality of the existing housing stock that could potentially be purchased.
o California's strategic plan prioritizes and ranks counties based on a
number of factors including most severe housing condition, incomes,
and housing vacancy rates. Loan and grant delivery to tribal lands and the colonias is a priority. However,
according to field office officials, outreach efforts on tribal lands and the colonias have not been as
successful as desired.
o Massachusetts officials described a different
scenario. The average sales
price of homes in the state often precludes many applicants
from finding an affordable house. Massachusetts officials said that the problem
exists
even though applicants in the Section 502 single-family guaranteed housing program can qualify at 115 percent of an area's median income. As a result, in 1999-2003, Massachusetts did not use its total allocation for both single-family direct program loans in 3 of the 5 years and its
guaranteed program in 4 of the 5 years.
A Majority of RHS Loans and Grants in Five States Went to Smaller Communities,
but Our Review Found Varying Interpretations of Eligibility
Our analysis of USDA's eligibility areas and program funding data
in the five
states we visited shows that a majority of RHS loans and grants went to
residents of communities with populations of 10,000 or below, while communities with higher populations received correspondingly fewer loans and grants. Additionally, we found variations in how RHS interpreted and applied the eligibility requirements, which can be partly explained by demographic differences in the rural areas of the states we visited and partly by differing interpretations or judgments as to what is "rural in character" by field offices. As a result, the application of the "rural in
character" finding in some eligible areas was open to question. We also
found that, as the population of a community increased, the application of
the "rural" criterion became less clear. Finally, we found that the
application of the statute's grandfather clause resulted in similar communities receiving different determinations for eligibility.
More Than Half of RHS Loans and Grants Went to
Communities with Populations of 10,000 or Below
For the five states we visited, 16,930, or about 58 percent, of the almost
29,000 RHS housing
loans and grants funded between fiscal years 1998 and 2004 were made in communities with populations of 10,000 or below.
The 29,000 loans and grants were made under RHS's major programs, including single-family and multifamily direct and guaranteed housing programs. As shown in Table 4, the majority of loans and grants issued in the different
housing programs went to communities with populations at or below 10,000 and successively smaller percentages of loans and grants for each program were made to communities in the higher population tiers. RHS
also provided 7,277 loans and grants, or about 25 percent, to communities
with populations ranging from 10,001 to 20,000. RHS funded an additional
15 percent of the loans and grants to communities with populations over
20,000, including some with populations over 25,000 that had received exemptions.
Table 4: Communities by Population Range in Five States That Received RHS
Housing Loans and Grants in Fiscal Years 1998-2004
Multifamily
Single-family Single- Direct and Percent
family of
Direct Loans Percent Guaranteed Percent Guaranteed Percent Total of All
of All
Population and Grants of Loan Total Loans of Programs Programs
Total Total
10,000 and 10,222 57.2 6,524 60.4 184 55.3 16,930
below
10,001 to 4,751 26.6 2,450 22.7 76 22.8 7,277
20,000
20,001 to 1,337 7.5 931 8.6 26 7.8 2,294
25,000
Greater
than
25,000 1,195 6.7 843 7.8 35 10.5 2,073
Not 358 2.0 49 0.5 12 3.6 419
available
Totals 17,863 62% 10,797 37% 333 1% 28,993 100%
Source: GAO Analysis of RHS Data.
Note: Data as of April 30, 2004.
Some Eligibility
Determinations in Communities with Populations at or below 10,000 Are Open to Question
During our site visits and through our analysis of RHS eligibility areas nationwide, we identified
communities with populations at or below 10,000 that
were contiguous to ineligible, higher-population communities, but that field offices determined to be eligible for RHS housing programs. The statute says that communities considered for eligibility cannot be part of
or associated with an urban area and that those with populations from 2,501 to 10,000 residents must be "rural in character." As a result, it
is not clear why all of these communities were considered eligible by the field staff.
For example, field staff told us that Belpre, Ohio, an economically depressed Appalachian community, is eligible for RHS programs because it
meets both the population and "rural in character" requirements. However, Belpre
is contiguous to Parkersburg, West Virginia, which
has a population of more than 33,000.14 In addition, the 2000 census considers Belpre, Ohio, along with Parkersburg and Vienna, West Virginia, as part of an urbanized area because its total population is over 50,000. Although it is across the Ohio River from Parkersburg, bridges have connected Belpre and Parkersburg for decades. According to a Belpre city employee, many people from Belpre work in Parkersburg because there are more jobs there. Furthermore, most of Belpre has a population density of 1,000
14Parkersburg, West Virginia, is not an eligible area.
people or more per square mile, which Census uses to determine urban areas. For these reasons, it is unclear whether Belpre meets the eligibility
requirements. Figure 5 shows Belpre, Ohio, in relation to Parkersburg and Vienna, West Virginia, as well as area density levels by census tract.15
15Census tracts are small, relatively
permanent statistical subdivisions of a county or
statistically equivalent entity used
to provide a stable set of geographical units for presenting decennial census
data. The 2000 census is the first census to divide the entire
country into census tracts.
remaining communities that are at or below 2,500 in population are not required to be "rural in character," but are required to "not to be part of or associated with an urban area." The approximately 1,300 communities represent 6 percent of the communities with population of 10,000 or fewer nationwide.
MSA and Grandfathering Provisions Affect Eligibility of Rural Communities
Circleville and Washington Court House, Ohio
During our visits, we also identified communities made
eligible through the grandfather clause and other communities that were ineligible because of the MSA restriction in Section 520, even though these communities had similar demographic profiles. As discussed previously, the statute
governing eligibility for RHS housing programs states that communities with populations from 10,001
to 20,000 are not eligible if they are within
an MSA. In contrast, other communities with populations from 10,001 to
25,000 could still be eligible even if they are within an MSA because the statute allowed some communities to be "grandfathered in" if they were
eligible prior to October 1, 1990, and still met the "rural in character" and
lack of mortgage credit requirements. The following four examples
illustrate these effects:
Both Circleville and Washington Court House, Ohio, are located in the general vicinity of Columbus, the state capital. Based on the 2000 census, both communities had populations of about 13,500 and were about the same size-about 6.5 square miles. However, Circleville, which is part of
the Columbus MSA, is not eligible while Washington Court House, which is
just outside the Columbus MSA, is eligible. Because Circleville is in
an MSA and was not eligible prior to October 1990, it remains ineligible
regardless
of its current population and whether or not it is "rural in character" (see fig. 6).
Figure~6: Circleville and Washington Court House Relative to Columbus, Ohio MSA
Source: GAO analysis of Census data.
Lamont and Taft, California We identified two communities in the
Bakersfield, California, area that, according to field office officials,
had similar demographics. Both of these communities are in Kern County and
the county's borders have defined the Bakersfield MSA since 1960. However,
Lamont is eligible and Taft is not. Lamont has just over 13,000 residents
and is about 14 miles from downtown Bakersfield. According to a state
office official, Lamont was grandfathered because it was eligible prior to
1980 while the Taft area was not. But the Taft area, which now has a
population of about 13,700 residents in four contiguous areas, is more
isolated and is located about 37 miles from downtown Bakersfield (see fig.
7).16 Field office officials believe that their housing programs would
benefit Taft, an economically depressed area as evidenced by its slow
growth since 1980. They asked the national office to allow Taft to receive
RHS funding. However, a field office official told us that the national
office denied the petition stating that they had to follow the statute
regarding eligibility determinations and could not make exceptions.
16The Taft area includes Taft, Taft Heights, South Taft, and Ford City.
Figure 7: Lamont and Taft Relative to Bakersfield, California, within the
Bakersfield MSA
Source: GAO analysis of Census data.
The grandfather clause also directly affected the eligibility status of two communities in Massachusetts that are adjacent to each other. Both
Belchertown and South Hadley are in the Springfield MSA. While both towns have areas that could be considered "rural" and current populations
are in excess of 10,000 but not over 20,000, Belchertown was grandfathered based on previous eligibility. In contrast, South Hadley was never eligible
because its population was over 10,000 and it was already part of the
Springfield MSA when grandfathering first began. (see fig. 8).
Figure 8: Both South Hadley and Belchertown Have Less Densely Settled
Areas within the Springfield MSA
Sources: GAO analysis of Census and RHS data.
Note: Population density is shown only for South Hadley and Belchertown.
Figure 9: South Hadley and Belchertown Rural Areas
South Hadley, Massachusetts Belchertown, Massachusetts
Sources: GAO (left photo); Leslie Campbell (right photo).
Casa Grande, Arizona Casa Grande, Arizona, is located between Phoenix and Tucson and became
part of the Phoenix-Mesa MSA in 1993. Its eligibility for RHS housing programs was grandfathered after its population increased above 20,000 (the grandfather clause allowed eligibility to already eligible towns with populations up to 25,000). However, the 2000 census found that Casa Grande's population had increased slightly over the 25,000 limit. As a result, Casa Grande received a congressional exemption for each of the last 3 fiscal years (2002-2004) to allow it
to remain eligible for RHS programs. A
Casa Grande housing official told us that the population increase was, in part, attributable to the success they have had in adding over 300 homes and 1,500 residents to the area through the self-help housing program. Additionally,
Casa Grande covers about 48 square miles, and only one of its nine census tracts has a density of more than 500 people per square mile
(see fig. 10). As noted previously, Census uses density as a measure of
urbanization.
Sources: RHS and Census data.
Changes to Components of the Definition of Rural
Could Address Inconsistent Treatment of Similar Communities
Changes to eligibility requirements in the definition of rural-whether focused on population, the grandfather clause, or lack of credit
requirements-could affect whom and where RHS serves. An option that would involve only changing the population limits in the definition of rural would impact the number of eligible communities, but without additional
resources, the major change would be a shift in where resources are spent.
However, population limits alone do not determine whether communities in currently eligible areas are "rural in character" and "not part of or associated with urban areas." Consequently, we explore three options that could allow RHS to better determine eligibility: (1) using density measures
that better reflect where people live rather
than county-based MSA criteria,
(2) discontinuing the "grandfathering" of communities, and (3) eliminating
the lack of credit requirement.
Eliminating MSAs and Using Density-based Measures Could Better
Reflect Population Patterns
Changing the population limits in the definition of rural would impact the number of eligible communities. For instance, if eligibility were kept to communities with populations at or below
10,000, RHS might be able to
concentrate services in fewer eligible areas, potentially making loans or grants to more applicants in low-population communities. In contrast, increasing the population limit to 50,000, the upper limit for some of the other USDA
rural development programs, could potentially affect RHS's
ability to serve both existing and newly added eligible areas by increasing the size of the potential applicant pool. Without additional resources, RHS
would have to shift available resources to provide some service in all the added areas. However, population limits alone do not address whether communities in currently eligible areas are "rural in character" and "not
part of or associated with urban areas." As a result, we explored options
that focused on arriving at a better analysis of population patterns and eliminating requirements that are not key to a determination of rural.
If the definition were changed to delete the MSA requirement and allow
the use of a density-based measure such as urbanized areas and urban clusters
to help make eligibility determinations, RHS would have better information
on which to base its decisions. We believe that the MSA requirement is not a key element in identifying a rural community. According to OMB, MSAs do not equate to an urban-rural classification and counties included in
MSAs may contain both urban and rural territory. In addition, 2000 Census data show that 13 percent of the population of counties currently in MSAs lived in rural areas and that these rural residents living within MSAs included more than half the population that the Census defined as rural.
Also, updates and changes that OMB makes to MSAs can result in changes to
eligible rural areas that are
outside of RHS's control and may or may not support
RHS's mission of serving rural areas and residents.
In addition, the number of MSAs has increased since 1974 (when Congress increased the rural housing programs' population eligibility limit from 10,000 to 20,000 for
communities outside of MSAs that lacked mortgage credit). Figure 11
shows that since the 1970 census, OMB added 443 counties (net) to MSAs.
a relatively fine scale.17 We note that density alone cannot
measure differences between urban and rural.
However, mapping changes in density might help RHS officials better
pinpoint the areas in which the boundaries dividing eligible from ineligible
areas might be drawn. For example, by eliminating the MSA criterion, RHS could review the eligibility of Washington Court House and Circleville,
Ohio, based on population and rural character criteria. Additionally,
using density-based mapping could help RHS draw boundaries around these
communities, which although Census-designated as "urban clusters," still
meet rural housing program population requirements (see fig. 12).
Source: GAO analysis of Census data.
Similarly, replacing MSA
criteria with density-based classifications such as urbanized areas and urban clusters could help resolve issues illustrated by
17Block groups
generally contain 600-3,000 people, with an optimum size of 1,500 people.
the two communities in Massachusetts that we discussed earlier. Currently, all of Belchertown is eligible for RHS housing programs, but all of South Hadley, which is near Belchertown, is not. The limits of each town
encompass both more-and less-densely settled areas-that is, areas that
could be judged as urban or "rural in character" (see fig. 13). Using density as a guide for determining eligibility, RHS could better judge where to draw a boundary that would differentiate the rural from the urban areas in both jurisdictions.
Sources: GAO analysis of Census and RHS data.
In another example, in April 1990, an RHS field office made Westminster,
Maryland, ineligible because the town was "fast becoming urban." Westminster's population increased from about 9,000 residents in 1980 to
about 13,000 in 1990 and 17,000 in 2000. Westminster is included in the Baltimore MSA. As shown in figure 14, portions of the ineligible area are
densely populated. However, although Westminster is in a growing area, it
is not contiguous to highly-populated communities and is about 36 miles from downtown Baltimore, a situation that could be more clearly represented using density-based measures.
Sources: GAO analysis of Census and RHS data.
Although urbanized areas and urban clusters better reflect the urban-rural
continuum than county-based MSAs, the boundaries often appear jagged or
random. Because the boundaries can appear convoluted, it would not be prudent to use the exact urbanized area/urban cluster lines for eligibility boundaries. It would make more sense to use the density information provided by the clusters as input for eligibility determinations. For example, an RHS sister agency, the Rural Business Cooperative Service,
interprets language from the 2002 farm bill to apply exact urbanized area
boundaries in eligibility determinations for its business and industry loan
program
(which helps develop rural businesses).18 However, officials in one state we visited told us that they believe that using exact urbanized area boundaries prohibits them from making loans in rural town business districts. These and RHS officials in other states we visited agreed that
using the density information would be helpful in their eligibility determinations, but that they would not want to be held to using density as
the sole eligibility factor.
USDA's Economic Research Service (ERS) has developed a number of density-based approaches that can be used to measure "rurality," such as
commuting zones and labor market area classifications. ERS developed these approaches to go beyond the county level to better capture the economic and
social diversity of rural areas. For example,
ERS reports that its rural-urban settlement continuum
model can more precisely reflect the diversity of rural settlement patterns because it is based on about 62,000 census tracts, rather than about 3,000 counties, and also considers
commuting flows to and from metropolitan core (urbanized) areas, population density, and population growth during the previous decade.
ERS believes that the "census-tract continuum" provides a more precise territorial
delineation than do the county-level data, particularly in the
large western counties where MSA boundaries stretch far beyond the actual urban core.19
Grandfathering Attempts to Address Eligibility Issues
Arising from MSA Requirement
Congress first wrote "grandfathering" language into the statute in 1983, 9 years after the MSA restriction became part
of the statutory definition. The provision enabled a number of communities located in MSAs to retain their
eligibility. Because census data show that 13 percent of MSA
residents reside in rural areas, it is likely that RHS will continue to grandfather certain areas as long as exclusion from an MSA remains an eligibility
criterion. If
the MSA requirement were eliminated and density-based measures such as urbanized areas and urban clusters were used as a basis
for eligibility determinations, the need
to "grandfather" communities would lessen since eligible communities would then better reflect the rural character of the areas. Eligibility decisions also would be made based on
18Farm Security and Rural Investment Act of 2002 (P.L. 107-171).
19Cromartie, John B. and Swanson, Linda L., Census Tracts More Precisely
Define Rural Populations and Areas, Rural Development
Perspectives, vol. 11, no. 3, Economic Research
Service (Washington, D.C.: June 1996).
the most recent census data rather than data dating as far back as 1980.
Moreover, eliminating the MSA requirement would allow more consistent treatment of communities with similar demographics, some of which meet population
requirements and are rural in character, but are ineligible
solely because of their inclusion in MSAs.
The most illustrative example we found during our visits of the potential
effect of using urbanized areas and urban clusters and discontinuing
grandfathering is the Bakersfield/Lamont/Taft, California situation. The left
side of figure 15 shows the Bakersfield MSA, which encompasses a county that is more than 100 miles wide in parts, and is basically rural outside the environs of Bakersfield. Lamont was grandfathered because it lost
eligibility when its population went above 10,000 at the 1980 census. Taft's
population was already over 10,000 prior to the 1980 census, so Taft was
not eligible for grandfathering. The right side of the figure shows what
would happen if MSAs and the grandfather clause were removed from the equation. Taft would be in its own urban cluster outside of the Bakersfield
urbanized area, which happens to include Lamont. Based on our visit, we
believe this scenario, where the more rural community would be the one eligible, is more in line with the overall purpose of the legislation than the current situation.
Source: GAO analysis of Census data.
Eliminating Lack of Credit Requirement Would Have Little Effect on
Eligibility Determinations
The statute imposes a requirement to demonstrate a serious lack of mortgage credit for lower-and moderate-income families in communities
with populations from 10,001 to 25,000. RHS has developed a policy stating that
a serious lack of mortgage
credit at comparable rates and terms to that offered by RHS exists in
all rural areas. In effect, RHS's blanket policy frees
field staff to focus
on the other requirements in making eligibility determinations for communities with populations from 10,001 to 25,000. However, a 1997 study by USDA's
Economic Research Service concluded that credit problems in rural areas are primarily limited to sparsely
populated or remote rural areas.20 Such communities generally do not fall
into the 10,001-25,000 range cited in the act.
Lack of income, rather than lack of credit availability, appears to be the primary credit problem in much of rural America. Many of the RHS officials
and industry experts with whom we spoke said that they see the problem more as lack of income than lack of credit. Moreover, eligibility
requirements for RHS programs are based on income levels. For example,
the single-family direct program is limited to low-income households
defined as those making 80 percent or less of area median income. The single-family
guaranteed program is limited to moderate-income families,
defined as those making 115 percent or less of area median income. If
income rather than credit were considered the major issue, then the lack of credit requirement would not
be central to determining eligibility as RHS program
requirements already incorporate income levels. Also, RHS uses targeting to further prioritize loan and grant making to areas based on income and other factors that determine the need for affordable housing.
Conclusions Changing some of the requirements that define eligibility for RHS housing
programs would allow the agency to better serve rural areas. Specifically,
certain elements of the definition of rural in Section 520 of
the Housing Act of 1949, as amended, are not key to making determinations of what areas are "rural," and in some cases may cause similar communities to receive dissimilar treatment. One such option would be to eliminate the MSA requirement. As OMB acknowledged, many counties included in MSAs
contain both urban and rural areas. In addition, according to the Census
Bureau, over 50 percent of the nation's rural population resides in rural
20U.S. Department of Agriculture, Economic Research Service, Rural Economic Division,
Credit in Rural America, Agricultural Economic Report No. 749 (Washington,
D.C: 1997).
areas within MSAs. Using alternative measures, such as the urbanized area
and urban cluster classifications, as a guide could help RHS better draw boundaries around rural areas because density based measures provide finer-scale information. Furthermore, if MSAs were removed from the
eligibility criteria, the grandfather clause could also be phased out to allow more eligibility considerations to be based on a community's current
characteristics. Our visits to five states have shown instances where the combination of the MSA requirement and grandfathering granted eligibility to communities that appear more closely linked with urban areas than similar, but ineligible, communities. Finally, according to USDA and mortgage industry experts, most of rural America does not appear to lack credit. If
credit availability issues are not prevalent in rural areas,
particularly those areas at the population tier specified in the statute, then the requirement appears to
have little or no relevance to eligibility for rural housing programs.
Matters for To better ensure that the definition of rural results in more consistent
application of eligibility for RHS housing programs, Congress may wish to
Congressional consider Consideration
o
eliminating the MSA criterion and recommending that RHS use density-based measures as a basis for its eligibility decisions,
o phasing out the practice of "grandfathering" communities that
experienced changes in eligibility because of inclusion in an MSA, and
o eliminating the "lack of credit" requirement.
Agency Comments and Our Evaluation
We provided drafts of this report to USDA, OMB, and the Census Bureau for review and comments.
USDA provided written comments that are discussed below and presented in Appendix II, and the Census Bureau provided technical comments that
were incorporated where appropriate. The Census Bureau's technical comments included the observation that ERS has developed a set of
commuting areas that reflect both population density and commuting data
criteria. We added discussion of ERS's efforts to develop alternative
measurement approaches that add information beyond that captured at
the county level, such as commuting zones, labor market areas, and the
rural-urban settlement continuum and how these approaches could be used
as a basis to judge the "rural character" of individual communities.
USDA stated that our report articulates how the use of MSAs has resulted in disparate treatment of some communities. The department also noted that applying a density-based measure might have merit but that further study is needed to properly define such a measure for nationwide application. USDA
cautioned that since rural areas were diverse, unintended impacts could possibly result because density-based measures
might not always identify areas that were "rural in character."
We agree that further study is needed to develop a density-based approach to identify areas that are "rural in character." However, USDA's ERS has developed measurement approaches that reflect both population density and commuting data criteria, and these density-based approaches provide a basis to better identify and describe rural areas.
USDA also added that Rural Development would implement any modifications to the statutorily based "grandfathering" provisions that
Congress directs. Finally, USDA stated that Rural Development anticipates
that
the "lack of credit requirement" could be removed with no detriment to RHS housing programs.
As agreed with your office, unless you publicly announce the contents of
this report earlier, we plan no further distribution until 30 days from its
issuance date. At that time, we will send copies of this report to the
Ranking Minority Member for the House Subcommittee on Housing and Community Opportunity and the Chairmen and Ranking Minority Members
for the House Committee on Financial Services, Senate Committee on Banking, Housing, and Urban Affairs, House Committee on Agriculture, Senate Committee on Agriculture, Nutrition and Forestry, and other interested congressional committees. We will send copies to the Secretary
of the Department of Agriculture, the Director of the Office of Management
and Budget, and the Director of the Bureau of the Census. We will also provide copies to others upon request. In addition, this report will be available at
no charge on the GAO Web site at http://www.gao.gov.
If you or your staff have any questions concerning this report, please contact me at (202) 512-4325 or at
[email protected], or Andy Finkel at (202) 512-6765 or at [email protected].
Key contributors to this report are listed in Appendix III.
Sincerely yours,
William B. Shear Director, Financial Markets and Community Investment
Appendix I
Scope and Methodology
To determine how the Rural Housing Service (RHS) defines rural and implements its programs,
we interviewed officials from the U.S.
Department of Agriculture (USDA), Department of Housing and Urban Development (HUD), the Census Bureau (Census), and the Office of
Management and Budget (OMB) to obtain their views on the federal
government's role in providing housing and other community programs to
rural areas. In addition, we also met with representatives of private nonprofit housing organizations, including trade and advocacy organizations, to ascertain their views on the target populations and
locations for RHS housing programs. To obtain an understanding of how
rural areas vary across the country and how
RHS defines rural in different parts of the country, we visited five states-Arizona, California, Maryland, Massachusetts, and Ohio-and in each state met with officials from RHS Rural Development field offices, and representatives of state agencies and local organizations. We chose these states because they face different
challenges serving rural residents. Collectively, they administer all the available rural housing programs; depict a variety of urban, suburban, and rural populations; contain both affluent and underserved areas, including
tribal lands and the colonias; and represent geographically and demographically dispersed areas of the country. Furthermore, including Massachusetts allowed us to evaluate some of the unique issues involved with town-based rather than county-based local governing bodies. Finally,
we researched the legislative history of the statutory definition (Section
520 of the Housing Act of 1949 and its amendments) for all rural housing programs and analyzed Census and OMB classifications of geographic areas, based on population levels and density-particularly MSAs-to obtain an understanding of the applicability of these classifications to the current definition.
To determine how RHS targets areas for program delivery, we reviewed regulations and handbooks as well as state funding allocations and obligations. We also reviewed state strategic plans and discussed targeting goals and the ability to deliver resources to high-needs areas with appropriate RHS headquarters staff and officials in the five states we visited. We analyzed state loan and utilization data where available and compared the data
with the targeting goals for the five states we visited. In
addition, we spoke with representatives of several national housing organizations to obtain their views on whether RHS was targeting
resources efficiently.
To evaluate the populations of the rural areas served by RHS housing programs, we used information obtained from databases maintained by
Appendix I Scope and Methodology
Information Resources Management in St. Louis and focused our analysis on
the five states we visited. We used information extracted from the
Guaranteed Loan System (GLS), Dedicated Loan Origination and Servicing
System (DLOS), and Multi-Family Housing Information System (MFIS)
databases. Each extract from a database contained overall funding
information for USDA loans and grants in the department's single-family
direct, single-family guaranteed, and multifamily housing loan programs,
from October 1998 through April 2004.
To determine the characteristics of locations and populations within the
five states that receive RHS loans and grants, we used the most recent
funding information from the three databases in conjunction with 2000
census data. We analyzed the extent to which different population ranges
received RHS housing assistance under the aforementioned programs for the
5- 1/2 -year period. For our analysis, we divided each location's
population into one of five numerical ranges, and for each range, computed
the total number of loans and grants and the total dollar amount of such
assistance. This analysis enabled us to compare and contrast the extent to
which different locations within the states received RHS housing
assistance.
Although we found the St. Louis data to be generally adequate to perform
our analysis, the vast majority of these files did not contain population
data for the communities that received RHS assistance. We used the 2000
census to identify population data for the 3,014 unique communities in the
five states that received 28,993 loans and grants.1 In assessing the
reliability of relevant 2000 decennial U.S. Census Bureau data, we
reviewed information available online from the U.S. Census Bureau Web site
on its data quality assurance processes. On the basis of the results of
our document review, we concluded that the relevant census data we used
were reliable for our purposes for this analysis. Because the census data
did not contain population information for 403 (13.4 percent) of the
unique communities, we used other sources to obtain this information,
ranging from checking a community's Web site (if one was available) to
calling a community's administrative offices. USDA state officials also
helped by supplying population data in some cases and confirming
population data we received from others sources in other cases. These
actions enabled us to document
1Unique communities are places, towns, villages, or city names that we
identified in one or more of the 3 databases.
Appendix I Scope and Methodology
populations for all but 8 of the 3,014 unique communities identified in the RHS databases. We could not identify population data for 419 of the 28,993 loans and grants because the databases did not include adequate information (street address, name of town, and ZIP Codes) to identify a corresponding community. To determine the extent of loan and grant activity for the previous 6 years on a nationwide basis, we contacted officials at the St. Louis Finance Office and obtained financial reports for this period.
To ensure that RHS databases were reliable, we met with USDA officials in
St. Louis to discuss the accuracy and reliability of this information. We also performed internal electronic and manual checks to determine (1) the
extent to which the data were complete and accurate, (2) the reasonableness
of the values and
information in the data fields, and (3) the existence of data limitations such as missing information, incorrect data
types, and the existence of other unusable data that would undermine the
accuracy and usability of the data. Based upon this reliability assessment, we concluded that the data were reliable for purposes of this report. The data obtained from RHS are as of April 30, 2004.
To understand how a potential change in the definition of rural might
impact RHS' mission, we reviewed alternative definitions of and classifications used to define rural, such as those used by Census and OMB,
and several classification schemes developed by USDA's Economic
Research Service. We also developed case studies in the five states we visited. We interviewed officials at state and local RHS offices, and spoke
with representatives of local community development organizations to obtain their views on the current definition of rural and the challenges of
implementing programs. We also visited numerous eligible and ineligible
rural communities in the various states to assess the impact of the definition on seemingly similar communities
(based on demographic profiles).
We conducted our review from January 2004 through October 2004 in accordance with generally accepted government auditing standards.
Appendix II
USDA Comment Letter
Appendix II USDA Comment Letter
Appendix III
GAO Contacts and Staff Acknowledgements
GAO Contacts William B. Shear, (202) 512-4325, [email protected]
Andy Finkel, (202) 512-6765, [email protected]
Staff In addition to those named above, Diane Brooks, Martha Chow,
Mark Egger, Diana Gilman, Christine Kuduk, Richard LaMore, John
Acknowledgements McGrail, John Mingus, Barbara Roesmann, and Thomas Taydus made key contributions to this report.
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
The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site (www.gao.gov). Each weekday, GAO postsGAO Reports and newly released reports, testimony, and correspondence on its Web site. To
Testimony have GAO e-mail you a list of newly posted
products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."
Order by Mail or Phone The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders
should be sent to:
U.S. Government Accountability Office 441 G Street NW, Room LM
Washington, D.C. 20548
To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061
To Report Fraud, Contact:
Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm
E-mail: [email protected] Federal Programs
Automated answering system: (800) 424-5454 or (202) 512-7470
Congressional Gloria Jarmon, Managing Director, [email protected] (202) 512-4400
U.S. Government Accountability Office, 441 G Street NW, Room 7125
Relations Washington, D.C. 20548
Public Affairs Susan Becker, Acting Manager,
[email protected] (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
*** End of document. ***