This is the accessible text file for GAO report number GAO-05-622T 
entitled 'Community Development Block Grant Formula: Targeting 
Assistance to High-Need Communities Could Be Enhanced' which was 
released on April 26, 2005.

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.

Testimony:

Before the Subcommittee on Federalism and the Census, Committee on 
Government Reform, House of Representatives:

United States Government Accountability Office:

GAO:

For Release on Delivery Expected at 10:00 a.m. EDT:

Tuesday, April 26, 2005:

Community Development Block Grant Formula:

Targeting Assistance to High-Need Communities Could Be Enhanced:

Statement of Paul L. Posner, Managing Director:
Federal Budget Analysis and Intergovernmental Relations:

GAO-05-622T:

GAO Highlights:

Highlights of GAO-05-622T, a report to House Committee on Government 
Reform, Subcommittee on Federalism and the Census.

Why GAO Did This Study:

The subcommittee asked GAO to comment on the Department of Housing and 
Urban Development’s (HUD) 2005 report on the Community Development 
Block Grant (CDBG), “CDBG Formula Targeting to Community Development 
Need.” The CDBG program distributes funding to communities using two 
separate formulas that take into account poverty, older housing, 
community size, and other factors. That study evaluates the program’s 
funding formula from two perspectives: 1) to what extent do communities 
with similar needs receive similar CDBG funding, and 2) to what extent 
are program funds directed to communities with greater community 
development needs. The HUD report is particularly salient in light of 
the administration’s 2006 budget request which criticizes the program 
for not effectively targeting high-need communities. The subcommittee 
asked us to provide our views on the HUD study based on our experience 
and past assistance to various congressional committees on a wide 
variety of federal formula funding issues.

What GAO Found:

HUD’s report on the CDBG formula provides a thoughtful and 
sophisticated analysis of those elements of the formula that impede 
effective and equitable targeting of limited federal resources. Central 
to HUD’s analysis is an index of need that encompasses a wide variety 
of indicators related to poverty, housing infrastructure, and 
population growth and decline. While we would question some of the 
factors in their index, overall we believe it serves as a reasonable 
basis for evaluating CDBG targeting. 

The study identifies a number of causes that explain the poor 
performance of the current formula.
*	The use of two formulas rather than one is an important reason 
communities with similar needs do not receive similar funding. 
*	The use of population size as a need indicator significantly 
reduces the extent to which funding is directed to high-need 
communities.
*	Changing the poverty measure to one based on the poverty status 
of households rather than individuals would avoid large grants to 
communities with large student populations. 
*	An increasing number of communities have attained the minimum 
population size necessary to be eligible for formula funding and this 
has also reduced funding to communities with the highest needs. 

In addition to presenting formula options that address a number of 
these problems, HUD’s study also presents an option that would include 
per capita income in the formula. The inclusion of per capita income 
could be justified on the grounds that it directs more funding to 
communities with weaker economic capacity to meet needs from local 
resources. However, some of the effect of this factor is offset by 
introducing an additional factor--metropolitan per capita income. The 
metropolitan per capita income factor directs more rather than less 
funding to communities located in high-income metropolitan areas. This 
works at cross purposes with the local per capita income factor. 

GAO suggests that the subcommittee consider a needs-based criterion to 
determine eligibility and eliminate the grandfathering of eligibility 
into the formula before this approach is adopted as a means of 
improving the targeting performance of the program. 

[Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-05-622T].

To view the full product, including the scope
and methodology, click on the link above.
For more information, contact Paul L. Posner, (202) 512-9573, 
posnerp@gao.gov.

[End of Section]

Mr. Chairman and Members of the Subcommittee:

I am pleased to be here today to discuss policy considerations 
associated with fashioning a grant targeting policy and provide our 
observations on the Department of Housing and Urban Development's (HUD) 
report titled: "CDBG Formula Targeting to Community Development Need." 
In our recent report on 21st Century Challenges,[Footnote 1] we argue 
for the importance of a thorough assessment of federal programs and 
policies across the board due to long term fiscal challenges the nation 
currently faces. In that report we specifically recommend that programs 
such as the Community Development Block Grant (CDBG) be judged 
according to whether they target assistance to those with the greatest 
needs and the least capacity to meet them.

The CDBG program is a significant direct federal-to-local grant 
program. It supports a wide array of local community development 
activities that are primarily to benefit low-and moderate-income 
persons. Program funding is allocated to local communities using two 
statutory formulas that take into account various indicators of 
community development need. The HUD report observes that this formula 
provides widely different payments to recipients with similar needs and 
that funds going to the neediest communities have decreased over time 
on a per capita basis. The study then presents several alternative 
measures of community need that would systematically focus support on 
those communities with the greatest need. This subcommittee asked us to 
evaluate the HUD report.

The HUD study takes on even greater significance in light of the 
administration's proposal to consolidate 18 federal community and 
economic development programs, including CDBG, into a single block 
grant. The administration proposal would reduce overall funding by 30 
percent. Such a cut raises issues regarding the need to more sharply 
focus limited funding on those communities in greatest need. In this 
regard the administration's initiative criticizes the CDBG program as 
being poorly targeted, indicating that 38 percent of the funds go to 
eligible communities and states with poverty rates below the national 
average. To improve targeting, the administration proposal cites both 
need, specifically poverty, and economic capacity indicators such as 
unemployment and job loss as important indicators of the need for 
development funding. Criticisms of poor targeting raise fundamental 
questions about the relationship between formula design choices and 
federal policy goals.

Over the years we have evaluated and provided technical assistance on a 
number of formula grant programs. Consequently, we have a broad 
perspective on formula design issues. Today I will draw on our past 
work on a variety of grant programs to discuss several key issues that 
can contribute to good formula design. I will then provide our 
observations on HUD's evaluation of the current formula and the 
alternative targeting policies outlined in their report. Finally, I 
will offer some suggestions the subcommittee may wish to consider to 
better account for differences in local communities' economic 
capacities to meet local needs with local resources. We did not 
independently verify the reliability of the data used in HUD's report 
nor did we verify their analysis.

To briefly summarize our observations, I would first note that good 
formula design and grant targeting depend on a number of important 
policy choices. While the HUD study provides a thoughtful analysis of 
grant targeting based on improved measurement of program need, 
additional issues merit further consideration, including taking into 
account not only the need for community infrastructure improvement but 
also communities' economic capacities to address those needs. In 
addition, the subcommittee should consider revising eligibility 
criteria to encompass both needs and economic capacity.

As agreed with the subcommittee, I will not be commenting on issues 
related to the state program that provides funding for non-entitlement 
communities. I would be happy to discuss these issues during our 
question and answer period if time allows.

Grant Formula Design Embodies Several Policy Considerations:

Over the years we have reported on a wide variety of grant formula 
issues. During the 1970s and 1980s, we issued a number of reports on 
the funding formulas used to direct Revenue Sharing funds to local 
communities based on both their capacity and willingness to utilize 
local resources to address local needs. In anticipation of the 2000 
census, we examined the potential effect of the decennial census 
population undercount on the distribution of federal grant funds for 25 
large formula grant programs, including Medicaid. Over the years we 
have also assisted the Congress in revising the funding formulas under 
the Ryan White CARE Act, the Older Americans Act, Substance Abuse and 
Mental Health Block grants, and Title I education grants so that 
program funding would be more responsive to changes in program needs. 
This wide range of experience provides us with an in-depth 
understanding of the issues associated with the equitable and efficient 
targeting of federal grant dollars.

Based on our past experience, I would like to offer a number of 
observations on the design of grant funding formulas. First, grant 
formulas reflect an intergovernmental partnership that structures how 
costs are to be shared among the various levels of government. When 
federal resources represent a declining share of the cost of meeting 
national goals, a greater effort to target high-need communities is 
necessary if federal funding is to make a significant contribution to 
closing the fiscal gap between high-and low-need communities.

Second, targeting grant funding involves two key decisions: 1) 
determining which communities are eligible for assistance and 2) how to 
distribute funding among eligible communities. A clear statement of 
policy goals and objectives is essential as a guide for establishing 
grantee eligibility standards and identifying a manageable number of 
statistical indicators that can reliably direct formula funding to 
communities with the greatest need. Because the CDBG program has a wide 
variety of policy goals--the elimination of slums, historic 
preservation, and promoting more rational land use, among others--
identifying eligibility standards and a reasonable set of indicators to 
represent program need is especially challenging. For example, the CDBG 
program's goal of improving the physical infrastructure of economically 
distressed communities is reflected in several of the need indicators 
used in the program's formula, such as poverty and older housing. 
However, there are no indicators for historic preservation or rational 
land use.

In addition to program needs, consideration of fiscal equity or 
fairness suggests additional targeting factors beyond need indicators. 
Here there are two issues: 1) wide differences in communities' ability 
to meet local needs with local resources and 2) geographic differences 
in the cost of financing local development projects. Regarding local 
resources, high income communities generally have stronger tax bases 
from which to fund program needs without relying on federal assistance 
compared to lower income areas. Accordingly, the allocation of scarce 
resources might reflect variations in local funding capacity. In 
addition, the cost issue arises for areas faced with a high cost-of-
living since they would need to pay more for the workers who actually 
deliver services at the local level.

Performance indicators are sometimes considered as a targeting factor 
though they present challenges as well. Ideally, performance indicators 
would reflect only grantee performance and not program outcomes that 
result from factors local officials have little ability to control. For 
example, it makes little sense to reward a state that has substantially 
reduced welfare dependence because it enjoyed a particularly strong 
economy but did no better than other grantees in terms of efficiently 
managing its welfare programs. Accurate performance indicators are 
particularly difficult to develop, especially as they pertain to goals 
that may take literally decades to realize. As a consequence, they 
require an even higher degree of scrutiny than needs-based indicators 
before being incorporated into funding formulas.

For this reason a more common approach to promoting accountability is 
to require grantees to provide matching funds for projects funded under 
the program. Grantees are likely to be more vigilant in screening and 
funding individual projects if they must put a significant portion of 
their own resources at risk. While often difficult to enforce, at a 
minimum, such a requirement forces public discussion of how grant funds 
are to be employed.

Two Formulas Are Used to Target Program Funding:

Before I turn to discussing the HUD study and its findings, I would 
first like to provide a brief description of the eligibility standards 
and funding formulas now used to target CDBG funding. To obtain 
entitlement status, a city must be the principal city of a metropolitan 
statistical area, as designated by the Office of Management and Budget 
(OMB), or have a population of at least 50,000 residents. An urban 
county must have a population of at least 200,000 residents. The 
formulas used to distribute funding among eligible communities reflect 
several broad dimensions of need. Originally, CDBG funding was 
distributed to entitlement communities based on a simple three-factor 
formula that took into account:

* the number of residents (population),

* the number of residents living in poverty, and:

* the number of overcrowded housing units.

Beginning in fiscal year 1978, Congress added a second three-factor 
formula that included the following need indicators:

* the number of residents living in poverty,

* the number of older housing units, and:

* slow population growth or decline.

Under this dual formula approach, grantees receive the larger amount 
allocated by either the first formula, commonly referred to as formula 
A, or the second formula, commonly referred to as formula B. The use of 
two formulas, each with three factors, results in allotments exceeding 
the funds available for distribution. To avoid this outcome, all 
grantee allotments are proportionally reduced to conform to the amount 
available for distribution by formula.

Declining Budget Resources Underscore the Need for More Efficient 
Targeting of Available Funding:

Since the advent of the entitlement portion of the program, the number 
of participating communities has nearly doubled, increasing from 606 in 
fiscal year 1975 to more than 1,100 in fiscal year 2004. This trend can 
be expected to continue both because population will continue to grow 
and because new standards for designating metropolitan areas, as 
promulgated by OMB and utilized by the program, are also likely to 
increase the number of eligible communities.

Since 1978 program funding has declined to roughly half its peak of 
$10.2 billion when measured in purchasing power of today's dollars. 
When population growth is factored in, the decline in real per capita 
spending has declined by two-thirds, as illustrated in the accompanying 
figure.

Figure 1: Trends in CDBG Funding Per Capita 1975-2005:

[See PDF for image]

[End of figure]

The policy implication of these trends is that with more limited 
resources, narrowing the gap between high-and low-need communities can 
only be realized by concentrating this more limited funding on high-
need communities. This requires a new look at the program's eligibility 
standards and funding formulas.

Given the Program's Broadly Defined Purposes, HUD's Evaluation Criteria 
for Grant Targeting Appear Reasonable:

The HUD study relies on two generally accepted equity or fairness 
principles to evaluate the targeting of CDBG funding: 1) equals should 
be treated equally and 2) those with greater needs should receive more 
than those with lesser needs. The first principle is based on the idea 
that communities with similar needs should receive roughly similar per 
capita funding amounts. The second standard is based on the idea that 
to reduce the gap between high-and low-need communities, additional 
funding must be targeted to communities with greater needs. This 
criterion is especially pertinent because, as the HUD report observes, 
Congress designed a formula intended to allocate CDBG funds according 
to variations in community needs. However, determining the extent to 
which program funding is disproportionately allocated to communities 
with the highest needs involves value judgments that are the 
responsibility of policymakers rather than technicians and 
administrators. The HUD study measures the extent to which funding is 
targeted to high-need communities and leaves it to policymakers to 
decide the appropriate degree of needs-based targeting.

Before I address the conclusions reached in the HUD study, I first want 
to spend a couple of moments discussing the factors underlying the 
study's need criterion, since all conclusions rest upon its validity. 
One of the criticisms directed at the CDBG program in the 
administration's fiscal year 2006 budget proposal is that there is a 
"lack of clarity in the program's purpose," a statement which is 
supported by the long list of specific program objectives cited in 
HUD's report. Given the broad and diffuse goals established for the 
program, it is difficult to identify a few clear and succinct 
indicators of program need appropriate for this program. Though HUD's 
need criterion is not immune from criticism, it is, in our view, 
reasonable given the program's diverse objectives. HUD's criterion is 
strongly related to poverty and older housing occupied by low-income 
households and a number of other variables related to local poverty 
conditions such as education, crime, and racial segregation. These 
variables represent 80 percent of HUD's overall index of need. This, I 
feel, represents a reasonable approach for distinguishing between high-
and low-need communities.

Other indicators included in HUD's need criterion may be more 
questionable. For example, overcrowded housing, one of the elements in 
the current formula, may be more indicative of a strong local economy 
that reflects strong demand pressures in the local housing market 
rather than economic decline. In addition, low population densities and 
strong population growth, both reflected in HUD's need criterion, may 
be more indicative of strong rather than weak economic conditions. 
However, to the extent that these indicators may be problematic, they 
represent a comparatively small part of the overall need criterion. 
Consequently, even if these factors were eliminated from the need index 
it is unlikely that they would affect their main conclusions to any 
significant degree.

Many Features of CDBG Funding Formulas Limit Their Ability to 
Consistently Target High-Need Communities:

The HUD study reaches a number of valid conclusions regarding the 
targeting performance of the program's funding formulas. I will just 
mention their conclusions to echo the more detailed analysis presented 
in the HUD report:

* The primary reasons entitlement communities with similar community 
development needs receive wide differences in funding are 1) using two 
formulas rather than a single formula and 2) the factor that reflects 
older housing in formula B results in especially large disparities in 
funding among communities with similar needs because units occupied by 
higher income residents typically are not in need of rehabilitation at 
public expense.

* Formula A is most responsible for reducing the extent to which 
funding is targeted to high-need communities, because its reliance on 
general population precludes greater targeting based on community 
development needs.

* Changing the poverty measure to one based on the poverty status of 
households rather than individuals would avoid awarding large grants to 
low-need college towns.[Footnote 2]

While HUD Formula Options Improve Needs Targeting, Additional Options 
Should Also Be Explored before Deciding on a Particular Reform Strategy:

In our view, the HUD study has clearly identified the major elements 
that limit the current formula's ability to efficiently and effectively 
target funding to high-need communities, and it puts forward a number 
of formula alternatives that would strengthen the program in this 
regard. Proposals range from a comparatively modest reform to options 
that result in a more substantial redistribution of program funding.

The study describes two formula alternatives to improve grant targeting 
among entitlement communities that incorporate new need indicators. The 
first option, formula alternative one, introduces revised indicators of 
poverty, older housing units and slow population growth and decline, 
and places greater emphasis on the poverty indicators. It provides 
modest improvements by narrowing wide differences in funding received 
by communities with similar needs and it directs a larger portion of 
funding to high-need communities. The second option, alternative two, 
takes a somewhat more aggressive approach by eliminating the use of two 
formulas and replacing them with a single formula that includes a range 
of indicators related to need. It provides a substantial improvement in 
the program's ability to provide comparable funding for communities 
with comparable needs.

However, it is important to point out that neither the poverty 
indicator used in the current formula nor the alternative HUD proposes 
takes into account geographic differences in the cost-of-living. As a 
consequence, both the current formula and the two alternatives probably 
overstate needs in communities with relatively low cost-of-living and 
understate them in communities with a higher cost-of-living.

I would characterize the first two alternatives as making technical 
improvements, in that they utilize better indicators of need and 
eliminate the primary causes of wide differences in funding for 
communities with similar needs. In contrast, a third option, formula 
alternative three, introduces two additional factors--community per 
capita income and the per capita income of the wider metropolitan area 
in which the grantee is located. Community per capita income (PCI) is 
used to increase funding for low-income communities and reduce funding 
for higher income communities. The metropolitan PCI factor partly 
offsets the effect of community PCI by increasing funding for 
communities in high-income metropolitan areas. The net effect of both 
factors is that the two factors, to some extent, work at cross 
purposes. For example, if two communities located in different 
metropolitan areas had the same PCI, the community located in the 
metropolitan area with a lower area-wide income would receive less aid 
than the community located in the high-income metropolitan area.

The HUD report suggests using the two per capita income factors because 
they provide a means of directing more funding to high-need 
communities. However, they really are much more than a technical means 
of producing more targeting to high-need communities. And for that 
reason, I would like to talk about their introduction into the formula 
in a little more detail.

While these two factors do direct more funding to high-need 
communities, they also widen rather than narrow differences in funding 
among communities with similar needs, in effect, increasing the error 
rate if measured simply in terms of targeting need. The HUD report does 
not provide any discussion that would justify allowing funding 
differences to widen under this option. The policy question this raises 
is: Can these differences be justified by differences in funding 
capacity or cost differences?

Clearly, the introduction of per capita income can be justified on the 
grounds that it provides a means of taking into account the underlying 
economic strength of communities and their ability to fund local needs 
from local resources. I would also observe that doing so is consistent 
with the administration's Strengthening America's Community Initiative, 
which emphasizes indicators of economic conditions such as job loss and 
unemployment. However, introducing economic capacity also raises the 
question of to what extent should low income places be targeted? For 
example, should a community with half the average income be given a 
grant that is twice the average, or possibly even more? The HUD study 
provides one answer to this question. The subcommittee may wish to 
consider possibilities with either a greater or lesser effect.

The inclusion of the metropolitan PCI introduces more controversial 
issues as well. This factor, rather than targeting more funding to low-
income areas, does the opposite. It actually targets more funding to 
communities in higher income metropolitan areas. However, the rationale 
for doing so is not discussed in HUD's report. One possible reason for 
introducing metropolitan PCI as a factor is that it would take account 
of geographic differences in the cost-of-living. However, consensus 
within the research community has not yet been achieved regarding the 
magnitude of these cost differences. Technical experts are therefore 
unable to provide guidance regarding how these cost differences may be 
offset in a funding formula. As a consequence, there is no objective 
basis to determine if HUD's use of metropolitan per capita income is 
appropriate.

Concluding Observations:

In conclusion, the prospect of increasing budgetary stringency at the 
federal level appropriately prompts a reexamination of programs that 
respond to challenges faced by communities throughout the nation. The 
administration's proposal to restructure assistance for community 
development opens up important issues regarding how to focus such aid 
on the nation's more hard pressed areas.

For the most part, the HUD study does a very effective job of 
identifying the critical decisions regarding grant targeting for 
congressional consideration. However, additional formula options are 
not explored as part of the process of reaching a decision on how best 
to target CDBG funding. If program funding continues to decline in 
inflation-adjusted dollars, it may be appropriate to go beyond simply a 
needs-based targeting policy and consider alternatives to also take 
into account the underlying strength of local economies to meet those 
needs.

Finally, while the formula is a central instrument in targeting program 
funding, the criteria used to establish entitlement status could also 
play an important role in directing a larger share of program funding 
to communities with the greatest need. Rather than the current 
program's reliance on population size as the primary criterion, the 
subcommittee may also wish to consider either including a needs-based 
element in eligibility standards or establishing a minimum threshold 
allotment in order to qualify for entitlement status. Finally, the 
subcommittee may wish to reconsider the grandfathering provisions that 
allow communities that no longer meet eligibility standards to continue 
participating in the entitlement program.

In closing, I would like to emphasize that the targeting issues raised 
by the HUD report are important no matter what level of financial 
support Congress provides for community development activities. The 
prospect of reduced support for such efforts, as proposed by the 
administration, would make consideration of these targeting issues 
particularly salient. I would also note that GAO's report on 21st 
Century Challenges calls for a reexamination of federal policies and 
programs to respond to a growing fiscal imbalance. Central to such a 
reexamination is assessing how to better target federal assistance to 
those with the greatest need and the least capacity to meet those needs.

Mr. Chairman, this concludes my statement. I would be happy to answer 
any questions you or other members of the subcommittee may have. For 
future comments or questions regarding this testimony, please contact 
Paul L. Posner, Managing Director for Federal Budget Analysis and 
Intergovernmental Relations, at (202) 512-9573. Individuals making key 
contributions to this testimony included Jerry C. Fastrup, Michael 
Springer, Robert Dinkelmeyer, and Michelle Sager.

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 (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 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, Waste, and Abuse in Federal Programs:

Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov 
Automated answering system: (800) 424-5454 or (202) 512-7470:

Congressional Relations:

Gloria Jarmon, Managing Director, JarmonG@gao.gov (202) 512-4400 U.S. 
Government Accountability Office, 441 G Street NW, Room 7125 
Washington, D.C. 20548:

Public Affairs:

Jeff Nelligan, Managing Director, NelliganJ@gao.gov (202) 512-4800 U.S. 
Government Accountability Office, 441 G Street NW, Room 7149 
Washington, D.C. 20548:

FOOTNOTES

[1] GAO, 21st Century Challenges: Reexamining the Base of the Federal 
Government, GAO-05-325SP, February 2005.

[2] Data on persons in poverty are from the Bureau of the Census which 
includes off-campus college students, who often receive support from 
their families that is not recorded by Census.