This is the accessible text file for GAO report number GAO-10-240R 
entitled 'Poverty Determination in U.S. Insular Areas' which was 
released on November 10, 2009. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as part 
of a longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 

Washington, DC 20548: 

November 10, 2009: 

The Honorable Jeff Bingaman:
Chairman:
Lisa Murkowski:
Ranking Member:
Committee on Energy and Natural Resources:
United States Senate: 

The Honorable Madeleine Z. Bordallo:
Chairwoman:
Henry E. Brown, Jr.
Ranking Member:
Subcommittee on Insular Affairs, Oceans and Wildlife:
Committee on Natural Resources:
House of Representatives: 

Subject: Poverty Determination in U.S. Insular Areas: 

Owing to high levels of poverty, American Samoa, the Commonwealth of 
the Northern Mariana Islands (CNMI), Guam, the Commonwealth of Puerto 
Rico, and the U.S. Virgin Islands (USVI) rely heavily on need-based 
federal programs to provide basic services.[Footnote 1] Two federal 
agencies publish measures used by some federal programs to determine 
poverty status and allocate need-based assistance: the Census Bureau 
(Census) publishes poverty thresholds--dollar-value benchmarks for 
determining poverty status--and the Department of Health and Human 
Services (HHS) provides poverty guidelines, which are derived from the 
poverty thresholds. The approaches used to determine these poverty 
measures affect, respectively, poverty population statistics and income 
eligibility of individuals and families for certain need-based federal 
assistance. The poverty thresholds apply nationwide and in the insular 
areas, with no geographic variation, while separate poverty guidelines 
for Alaska and Hawaii, but not for the insular areas, have been 
provided since 1970. 

In response to your request, we (1) examined how the Census poverty 
thresholds and HHS poverty guidelines are determined for the insular 
areas. In addition, we (2) considered the possibility of providing 
poverty thresholds and guidelines specific to the insular areas and 
identified the implications of extending to the insular areas the 
approach originally used to determine the Alaska and Hawaii guidelines. 
(We also presented this information in a recent briefing to 
congressional committee staff; see encl. I for an updated version of 
the briefing slides.) 

To address these objectives, we reviewed relevant literature on poverty 
determination in the United States and the insular areas and 
interviewed current and former agency officials. We studied the methods 
that were used to develop the Census poverty thresholds for the 50 
states and Washington, D.C., and the HHS poverty guidelines for the 
contiguous states and Washington, D.C. We also reviewed the reasoning 
applied in 1970 in establishing separate poverty guidelines for Alaska 
and Hawaii. We considered this approach because these are the only 
poverty guidelines that have been established specific to any 
geographic areas. The Alaska and Hawaii poverty guidelines are based on 
the cost-of-living differential between these two locations and 
Washington, D.C., as applied by the Civil Service Commission (CSC)--a 
predecessor agency to the Office of Personnel Management (OPM)--in 
making cost-of-living pay adjustments (COLA) for federal white-collar 
employees in Alaska, Hawaii, the CNMI, Guam, Puerto Rico, and 
USVI.[Footnote 2] These adjustments are known as nonforeign area COLAs 
and are in the process of being phased out and replaced by locality 
pay.[Footnote 3] We conducted our work from January to November 2009 in 
accordance with all sections of GAO's Quality Assurance Framework that 
are relevant to our objectives. This framework requires that we plan 
and perform the engagement to obtain sufficient and appropriate 
evidence to meet our stated objectives and to discuss any limitations 
in our work. We believe that the information and data obtained, and the 
analysis conducted, provide a reasonable basis for any findings and 
conclusions in this report. 

Background: 

Census poverty thresholds. The poverty thresholds were established in 
the 1960s as dollar-value benchmarks for measuring poverty:[Footnote 4] 
if a family's income is less than the assigned threshold, the family 
and each of its members is considered to be in poverty.[Footnote 5] 
Since 1968, Census has annually published the thresholds for use in 
generating statistics such as national, regional, and state estimates 
of Americans in poverty; in 1969, the Bureau of the Budget (the 
predecessor office to the Office of Management and Budget) issued a 
directive designating the thresholds published by Census as the federal 
government's official definition of poverty for statistical 
purposes.[Footnote 6] Some federal programs use such statistics in fund-
distribution formulas involving poverty to distribute program funds 
among states and other jurisdictions. The thresholds, which vary by 
family size and age group,[Footnote 7] apply throughout the 50 states 
and Washington, D.C., with no geographic variation. The thresholds for 
the base year 1963 were based on the January 1964 dollar costs of the 
U.S. Department of Agriculture's (USDA) Economy Food Plan for families 
of different sizes, multiplied by a factor of three to reflect--based 
on the 1955 Household Food Consumption Survey--the share of food in 
total expenditures. Census updates the thresholds for inflation 
annually, using the Bureau of Labor Statistics' Consumer Price Index 
(CPI).[Footnote 8] Although the methodology used to determine the 
thresholds has been subject to debate--for example, regarding whether 
the thresholds should be adjusted for geographic variations in cost of 
living--it has remained largely unchanged, except for minor technical 
adjustments in 1969 and 1981.[Footnote 9] 

HHS poverty guidelines. The poverty guidelines--a simplified version of 
the Census poverty thresholds--were also established in the 1960s and 
are used by certain federal programs, such as Job Corps and Head Start, 
in determining the income eligibility of individuals and families for 
need-based assistance.[Footnote 10] Like the thresholds, the guidelines 
reflect variations in family size but, unlike the thresholds, do not 
reflect variations in the age group of the family members. Each year, 
HHS issues guidelines for the 48 contiguous states and Washington, 
D.C.[Footnote 11] Since 1970, separate guidelines have been issued for 
Alaska and Hawaii that are higher than the national guidelines, based 
on the federal government's nonforeign area COLAs in 1970 for federal 
employees' salaries in those states.[Footnote 12] 

Census Uses the Same Poverty Thresholds for the Insular Areas as for 
the States, but HHS Does Not Issue Poverty Guidelines for the Insular 
Areas: 

Census applies the same thresholds for the insular areas as it does for 
the 50 states, without adjustment for geographic variations in cost of 
living. The thresholds constructed in the early 1960s did not include 
data specific to the insular areas; the two data sources for the 
thresholds, the 1955 Household Food Consumption Survey and the January 
1964 cost of the Economy Food Plan, did not cover these areas. In 
addition, the Bureau of Labor Statistics' CPI, which Census uses in 
adjusting the national thresholds for inflation each year, does not 
cover the insular areas.[Footnote 13] 

Although HHS issues poverty guidelines for the contiguous states and 
Washington, D.C., as well as separate guidelines for Alaska and Hawaii, 
HHS has not issued any guidelines for the insular areas. According to 
HHS, in cases in which a federal program using the poverty guidelines 
serves any of the insular areas, the federal office that administers 
the program is generally responsible for deciding whether to use the 
contiguous-states-and-D.C. guidelines for those jurisdictions or follow 
some other procedure.[Footnote 14] 

Lack of Data Prevents Construction of Poverty Thresholds for the 
Insular Areas, but Poverty Guidelines Could Be Constructed for These 
Areas: 

Census poverty thresholds specific to the insular areas cannot be 
constructed from available data. Because Census lacks certain insular 
area information--on the 1955 share of income spent on food, the cost 
of the January 1964 Economy Food Plan, and a record of CPIs for the 
insular areas--inflation-adjusted poverty thresholds for the insular 
areas cannot be constructed with the methodology used to construct the 
original thresholds.[Footnote 15] If these data were available, it is 
unclear whether the new insular area thresholds would be higher or 
lower than the national thresholds. However, an HHS official told us 
that applying the methodology used for the original thresholds to the 
insular areas would most likely produce thresholds lower than the 
national thresholds, owing to the higher share of food in insular 
areas' total family expenditures (HHS also made this observation in its 
written comments regarding a draft of this report; see encl. III). 
Increases or decreases in the Census thresholds for the insular areas 
could, by raising or lowering estimates of the incidence of poverty, 
have implications for federal programs that use fund-distribution 
formulas involving poverty. 

HHS poverty guidelines specific to the insular areas, reflecting 
geographic differences in the cost of living, could be constructed by 
applying the rationale used for Alaska and Hawaii in 1970.[Footnote 16] 
Using this approach, based on OPM's nonforeign area COLAs, would 
produce guidelines for the CNMI, Guam, and USVI that are 25 percent 
higher, and for Puerto Rico 14 percent higher, than the guidelines for 
the contiguous states and Washington, D.C. Because no nonforeign area 
COLA has been defined for American Samoa, this approach could not be 
used to compute guidelines for that area. (See encl. I, table 1, for 
indexes derived from the nonforeign area COLAs for Alaska, Hawaii, and 
the insular areas.) This approach would not take into account any 
differences in consumption patterns between federal employees and 
insular area poor populations.[Footnote 17] 

The implications of setting higher HHS poverty guidelines for the 
insular areas vary, depending on the design of the federal program and 
the program's reliance on the guidelines. 

* Higher HHS guidelines could affect federal programs that distribute 
need-based assistance directly to families and individuals in the 
insular areas, such as the Department of Agriculture's National School 
Lunch Program and Supplemental Nutrition Assistance Program (formerly 
known as the Food Stamp Program). However, we did not examine the 
potential impact on individual income eligibility or the cost effects 
that might result from extending the approach used for providing 
poverty guidelines for Alaska and Hawaii to the insular areas. 

* Higher HHS guidelines would not affect programs that use fund- 
distribution formulas for allocating funds to the insular areas as lump 
sums, such as HHS's Maternal and Child Health Block Grant, Community 
Service Block Grant, Social Services Block Grant, and Childcare and 
Development Fund.[Footnote 18] However, higher guidelines could affect 
such programs' distribution of any funds to beneficiaries, if those 
funds are distributed to families and individuals whose income 
eligibility is based on the guidelines. 

* Higher HHS guidelines also would not affect programs that cap federal 
spending in the insular areas, such as Medicaid, which is subject to 
annual funding limits.[Footnote 19] However, if an insular area bases 
the eligibility of families and individuals for such programs on HHS 
poverty guidelines, then higher guidelines could affect the area's use 
of capped federal funds--for example, by increasing the number of 
people who are eligible for the programs. 

Agency Comments and Our Evaluation: 

HHS and Census provided written comments regarding a draft of this 
report, which are presented with our responses in enclosures III and 
IV, respectively. HHS also provided technical comments, which we 
incorporated as appropriate. Following are summaries of HHS's and 
Census's written comments and our responses. 

HHS's written comments address three main points related to our report. 

* HHS suggests that OPM's nonforeign area COLAs are of insufficient 
statistical quality to use as the basis for adjusting the poverty 
guidelines for geographic cost-of-living differences. Citing as 
evidence three studies that consider alternative strategies for 
measuring poverty in the United States, HHS observes that none of these 
studies refer to the nonforeign area COLAs as a possible data 
source.[Footnote 20] However, as we note in our response to HHS's 
comments in enclosure III, all three studies are focused on poverty 
measurement nationwide and hence could not have considered the 
nonforeign area COLAs as a data source, because these COLAs do not 
cover the 48 contiguous states. 

HHS also states that it does not believe that it would be appropriate 
to use the current OPM nonforeign area COLA data to adjust the poverty 
guidelines for the insular areas. We considered these data because they 
were used in 1970 to construct the Alaska and Hawaii poverty guidelines 
and because these are the only poverty guidelines that have been 
established specific to any geographic areas. However, we are not 
making a recommendation for the use of this approach. 

* HHS states that it adapts and updates, rather than determines, the 
Alaska and Hawaii poverty guidelines. HHS notes that it inherited the 
Alaska and Hawaii poverty guidelines series in 1982 from the Office of 
Economic Opportunity, which established these guidelines in 1970. HHS 
also notes that it has not consulted OPM documents to derive cost-of- 
living data to calculate the guidelines. We changed the wording in the 
draft to clarify this point. 

* HHS states that if poverty thresholds for the insular areas could be 
constructed with the methodology used to construct the original 
national thresholds, the insular area thresholds would likely be lower 
than the national thresholds. HHS reasons that because median incomes 
in the insular areas in 1959 were significantly lower than those in the 
states, the share of family food consumption as a fraction of insular 
area family expenditures--the basis of the original Census methodology-
-would be higher, leading to a lower poverty threshold. HHS also 
presents two alternative methods for estimating poverty thresholds for 
the insular areas: (1) as a percentage of median family income and (2) 
based on the responsiveness of poverty thresholds to changes in 
inflation-adjusted income over time. HHS states that using these two 
methods would produce poverty thresholds for different insular areas 
ranging between 14 and 86 percent of the continental U.S. poverty 
thresholds. HHS says that it does not advocate lower guidelines for the 
insular areas but finds it inequitable to have higher guidelines for 
those areas because of the lower median income in those areas than that 
of the 48 states. 

We note that applying these alternative methods would result in 
different levels of access to federal programs for people with the same 
income living in two areas with different median incomes. For example, 
a family with an income of $10,000 in one of the insular areas would 
qualify for less federal assistance than a family with the same income 
in another U.S. jurisdiction. 

Census raises two main issues in its written comments: 

* Census urges us to further study the poverty guidelines in the 
insular areas before making any recommendations that would increase 
those guidelines by a significant amount. We note that we have not made 
any recommendations in this report. 

* Census suggests that we compare OPM's foreign area COLAs with insular 
area housing-cost differentials based on Census data and other sources. 
We did not consider alternative approaches for geographic COLAs of the 
poverty guidelines, such as the use of housing data; instead we focused 
our analysis on the approach used in 1970 to construct the Alaska and 
Hawaii poverty guidelines, based on OPM's process for computing 
nonforeign area COLAs. We note OPM's process includes a housing cost 
component that it adjusts for differences in the quality of housing, 
using statistical analysis; OPM then aggregates housing and other types 
of expenditures in constructing its cost-of-living index. 

We are sending copies of this report to interested congressional 
committees, as well as to HHS, Census, and the Office of Insular 
Affairs at the Department of the Interior. In addition, the report will 
be available at no charge on GAO's Web site at [hyperlink, 
http://www.gao.gov]. If you or your staff have any questions about this 
report, please contact me at (202) 512-3149 or gootnickd@gao.gov. 
Contact points for our Offices of Congressional Relations and Public 
Affairs may be found on the last page of this report. GAO staff who 
made major contributions to this report are listed in enclosure V. 

Signed by: 

David Gootnick: 

Director, International Affairs and Trade: 

Enclosures: 

[End of section] 

Enclosure I: Poverty Determination in U.S. Insular Areas: 

Briefing to Congressional Staff: 
May 28, 2009: 

Poverty Determination: 

Two federal agencies publish versions of the measure used to determine 
poverty in the United States: 

* Census publishes poverty thresholds to determine the number of people 
in poverty. Besides being released to the media each year, these 
figures are used for certain need-based federal programs in formulas 
for allocating program funds among states and other jurisdictions. 

*  The Department of Health and Human Services (HHS) provides poverty 
guidelines, which are used to determine households’ and individuals’ 
eligibility for certain need-based federal programs.

Census Bureau Poverty Thresholds: 

The poverty thresholds—dollar-value benchmarks for determining who is 
in poverty—vary by family size and by the ages of certain family 
members. 

* The thresholds were developed by Mollie Orshansky of the Social 
Security Administration. They are based on the January 1964 cost of the 
Economy Food Plan for families of different sizes, multiplied by a 
factor of three to account for other living expenses; somewhat 
different procedures were followed for one- and two-person units. The 
factor of three was based on data from the 1955 Household Food 
Consumption Survey. 

* If a family’s income—including all before-tax cash income, except 
capital gains, and excluding all noncash benefits, such as food stamps—
is less than the threshold, Census considers the family and each of its 
members to be in poverty. 

The thresholds are applied as benchmarks throughout the United States 
with no geographic variation.

Although the Orshansky methodology for developing the thresholds has 
been subject to debate, it has remained unchanged except for minor 
technical adjustments in 1969 and 1981. 

Census updates the thresholds for inflation yearly, using the Bureau of 
Labor Statistics’ Consumer Price Index.

The thresholds are used to generate poverty populations statistics, 
which are released to the media each year and are also used in formulas 
to allocate certain federal program funds among states and other 
jurisdictions.

HHS Poverty Guidelines: 

The HHS poverty guidelines are derived from the Census Bureau poverty 
thresholds but, unlike the thresholds, do not consider the number of 
family members who are children under 18 years of age or whether 
certain family units include individuals over 65. 

HHS publishes a set of poverty guidelines each year for the 48 
contiguous states and Washington, D.C.. 

Since 1970, HHS has also published separate guidelines for Alaska and 
Hawaii that were 25 and 15 percent higher, reflecting federal cost-of-
living allowances (COLA) for salaries of federal employees in those 
states at that time. 

Certain federal programs use the poverty guidelines to determine 
individuals’ and families’ income eligibility for need-based 
assistance: the higher the guideline amount, the greater the number of 
people who are potentially eligible for assistance.[A] 

[A] Some of the need-based programs using the poverty guidelines 
actually use percentages of the guidelines higher than 100 percent. 
Some other need-based programs use dollar figures for eligibility that 
are unrelated to the poverty guidelines. 

U.S. Insular Areas: 

American Samoa, the Commonwealth of the Northern Mariana Islands 
(CNMI), Guam, Puerto Rico, and the U.S. Virgin Islands (USVI) rely 
heavily on need-based federal programs to provide basic services. 

In 1999, the percentage of individuals below the continental-U.S. 
poverty thresholds ranged from 23 in Guam—nearly twice as high as in 
the continental United States—to 61 in American Samoa. 

The approach used to determine poverty affects poverty statistics and 
income eligibility of individuals and families for certain need- based 
federal assistance in the insular areas. 

Research Questions: 

How are Census Bureau poverty thresholds and HHS poverty guidelines 
determined for the insular areas? 

Could poverty thresholds and guidelines specific to the insular areas 
be provided, and what would be the implications of extending to the 
insular areas the approach originally used to determine the Alaska and 
Hawaii guidelines?

Census uses the same poverty thresholds for the insular areas that it 
uses for the 50 states. 

Census issues the same thresholds for the insular areas as it does for 
the 50 states, without adjustment for geographic variations. 

Data on the food/income share were not collected by the 1955 Household 
Food Consumption Survey for the insular areas. 

Data on the January 1964 cost of the Economy Food Plan – the other data 
element on which the thresholds were based – were not collected for the 
insular areas in 1964. 

Although Census uses the CPI to adjust the national thresholds for 
inflation each year, the Bureau of Labor Statistics does not collect 
price data to construct insular areas’ CPI. 

HHS does not issue poverty guidelines for the insular areas. 

HHS has never issued guidelines for the insular areas. 

According to HHS, in cases in which a federal program using the poverty 
guidelines serves any of those jurisdictions, the federal office that 
administers the program is generally responsible for deciding whether 
to use the contiguous-states-and-D.C. guidelines for those 
jurisdictions or follow some other procedure. 

Census Bureau poverty thresholds specific to the insular areas cannot 
be constructed from available data. 

Lacking information for the insular areas on the 1955 share of income 
spent on food, the January 1964 cost of the Economy Food Plan, and the 
CPI, inflation- adjusted poverty thresholds specific to the insular 
areas cannot be constructed using Orshansky methodology. 

Without the 1964 cost of the Economy Food Plan or the share of income 
spent on food in the insular areas, it is unclear whether thresholds 
for the insular areas, if constructed, would be higher or lower than 
the national thresholds. However, an official from HHS told us that the 
most likely result if the Orshansky methodology could be applied to the 
insular areas would be thresholds lower than the national thresholds 
due to the higher share of food in insular areas’ total family 
expenditures. 

Higher or lower thresholds for the insular areas would raise or lower 
estimates of poverty incidence, affecting allocations of need-based 
assistance from federal programs that use poverty statistics in 
distribution formulas. 

The approach used for providing HHS poverty guidelines for Alaska and 
Hawaii in 1970 could be extended to the insular areas. 

HHS poverty guidelines specific to the insular areas, reflecting 
geographic differences in the cost–of–living, could be constructed by 
applying the rationale used for Alaska and Hawaii in 1970. 

Using this approach would produce higher guidelines—25 percent higher 
for CNMI, Guam, and USVI and 14 percent higher for Puerto Rico—than the 
guidelines for the contiguous states and Washington, D.C. Because no 
COLA is defined for American Samoa, this approach could not be used to 
compute guidelines for that area. 

This approach would not take into account any differences in 
consumption patterns between federal employees and insular area poor 
populations. 

See table 1.

Table 1: Index Derived from OPM Cost- of-Living Allowances: 

Locality: American Samoa; 
Index derived from OPM nonforeign area COLA: N/A. 

Locality: CNMI; 
Index derived from OPM nonforeign area COLA: 125. 

Locality: Guam; 
Index derived from OPM nonforeign area COLA: 125. 

Locality: Puerto Rico; 
Index derived from OPM nonforeign area COLA: 114. 

Locality: USVI; 
Index derived from OPM nonforeign area COLA: 125. 

Locality: Alaska; 
Index derived from OPM nonforeign area COLA: 124. 

Locality: Hawaii; 
Index derived from OPM nonforeign area COLA: 123. 

Locality: Washington D.C.; 
Index derived from OPM nonforeign area COLA: 100. 

Source: OPM. 

Note: COLA rates shown were published in the Federal Register by OPM on 
February 20, 2009 amending subpart B of 5CFR part 591. Rates shown for 
Alaska and Hawaii are simple unweighted averages of statewide COLAs. No 
COLA is listed for American Samoa; however, OPM defines a post 
differential rate for American Samoa of 25 percent. The post 
differential is based on extraordinarily difficult living conditions, 
excessive physical hardship, or notably unhealthful conditions. The 
base is Washington, D.C. Under 5 USC § 5941 nonforeign area COLAs are 
limited to 25 percent of basic pay. 

[End of table] 

Higher HHS poverty guidelines would affect some federal programs. 

Higher HHS guidelines could increase the number of individuals and 
families eligible for need-based assistance from federal programs that 
use the contiguous-states-and-D.C. guidelines as a criterion. (See app. 
I for examples of such programs.) 

Higher HHS guidelines would not affect: 

* federal programs that distribute need-based assistance according to 
formulas involving poverty statistics based on Census Bureau poverty 
thresholds, or: 

* federal programs that cap the distribution of need-based assistance 
to the insular areas according to statutorily defined percentages, such 
as Medicaid. 

Methodology: 

We reviewed the relevant literature on poverty determination in the 
United States and the insular areas, interviewed agency officials, and 
analyzed OPM’s methods used for cost-of-living pay adjustments. 

We did not examine the potential impact of extending to the insular 
areas the approach originally used to determine the Alaska and Hawaii 
guidelines on individual income eligibility and did not evaluate the 
cost effects of this approach. 

We conducted our work in accordance with GAO's Quality Assurance 
Framework. 

Appendix I: Select Federal Programs Using HHS Poverty Guidelines to 
Determine Eligibility for Need- Based Assistance in Insular Areas: 

National School Lunch Program (and Commodity School Program); 

School Breakfast Program; 

Special Milk Program for Children; 

Child and Adult Care Food Program; 

Summer Food Service Program; 

Special Supplemental Nutrition Program for Women, Infants, and 
Children; 

Supplemental Nutrition Assistance Program (Food Stamp Program)[A] 

[A] Uses poverty guidelines to determine income eligibility for need-
based assistance in Guam and USVI only. 

[End of section] 

Enclosure II: Briefing for Congressional Requesters: 

Insular Areas' Poverty Rates in 1999: 

Guam; 
Poverty rate (%): 23.0. 

American Samoa; 
Poverty rate (%): 61.0. 

USVI; 
Poverty rate (%): 32.5. 

CNMI; 
Poverty rate (%): 46.0. 

Puerto Rico; 
Poverty rate (%): 48.2. 

Source: 2000 Census of Population and Housing, U.S. Census Bureau. 

[End of table] 

Comments from the Department of Health and Human Services: 

Note: GAO comments supplementing those in the report text appear at the 
end the this enclosure. 

Department Of Health & Human Services: 
Office Of The Secretary: 

Assistant Secretary for Legislation: 
Washington, DC 20201: 

October 8, 2009: 

David Gootnick: 
Director, International Affairs and Trade: 
U.S. Government Accountability Office: 
441 G Street N.W.: 
Washington, DC 20548: 

Dear Mr. Gootnick:

Enclosed are comments on the U.S. Government Accountability Office's 
(GAO) report entitled: Poverty Determination in U.S. Insular Areas 
(GAO- 09-1041R). The Department appreciates the opportunity to review 
this report before its publication. 

Sincerely,

Signed by: 

Andrea Palm: 
Acting Assistant Secretary for Legislation: 

General Comments From The Department Of Health And Human  Services 
(MIS) On The Government Accountability Office's (GAO)  Draft Report 
Entitled: Poverty Determination In U.S. Insular Areas (GAO-09-1041R): 

The Office of the Assistant Secretary for Planning and Evaluation 
(ASPE) appreciates the opportunity to review and comment on the 
Government Accountability Office's (GAO) draft report, which examines 
the Census Bureau's poverty thresholds and the Department of Health and 
Human Services (HHS) poverty guidelines and how they are applied to the 
insular areas. The draft report also describes one specific procedure 
that could potentially be used to develop thresholds and guidelines 
specific to the insular areas—a procedure involving federal employee 
cost-of-living allowance figures. ASPE's comments on the draft report 
address three main topics: (1) the quality of the federal employee cost-
of-living allowance figures as a possible data base for adjusting the 
federal poverty measure; (2) the use of the federal salary figures to 
develop poverty guidelines for Alaska and Hawaii in 1966-1970; and (3) 
the probability that poverty thresholds for the insular areas 
calculated on the same basis as Mollie Orshansky's poverty thresholds 
for the continental U.S. would be lower than those for the continental 
U.S. 

Federal Employee Cost-of-Living Allowances—Not a Data Base of 
Acceptable Quality: 

Over the decades, there have been several occasions on which poverty 
measurement professionals have indicated that they believe that the 
federal employee cost-of-living allowance figures are not a data base 
of sufficiently good statistical quality to be used in adjusting the 
federal poverty measure. The Office of Economic Opportunity (0E0) 
employees familiar with the development of the Alaska/Hawaii poverty 
guidelines subsequently concluded that the federal employee cost-of-
living allowances were among the data bases that were not of 
sufficiently good quality to be used in adjusting the poverty measure. 

See comment 1. 

0E0 began issuing poverty guidelines in December 1965. The first 
poverty guidelines were the same for all 50 states (and the District of 
Columbia). In this administrative context, 0E0 decided during the 1966-
1970 period to institute separate poverty guidelines for Alaska and 
Hawaii. The new guidelines were 25 percent (Alaska) and 15 percent 
(Hawaii) higher than the original guidelines.[Footnote 22] 0E0 never 
published any explanation of how or from what source these percentage 
differentials were derived. However, a later unpublished memorandum 
from three 0E0 employees indicated that 0E0 derived these percentage 
differentials from the Civil Service Commission's (CSC) federal 
employee pay (geographic) cost-of-living allowance differentials 
for Alaska and Hawaii.

In 1976, a federal interagency Poverty Studies Task Force published a 
major study on poverty measurement, The Measure of Poverty.[Footnote 
23] The Poverty Studies Task Force included two of the 0E0 employees 
who knew that 0E0 had used the CSC federal salary figures to derive the 
Alaska and Hawaii guidelines. Because of considerable interest in the 
general issue of geographic cost-of-living differences, this study 
devoted considerable effort to examining this issue. "The conclusion of 
this extensive analysis is that although there may be geographic 
differences in the cost of living, there is no known way to make 
satisfactory geographic adjustments to the poverty cutoffs" (p. 82). 
"None of [the] existing data sources [examined] is of sufficient size 
and quality to support the kind of detailed analysis that Would be 
required to establish valid geographic equivalences [for poverty 
measurement]" (p. 49). 

Although it has been four decades since 0E0 used the CSC figures to 
institute separate guideline figures for Alaska and Hawaii, the Census 
Bureau has never used the CSC figures to institute separate poverty 
threshold figures for those states. 

In addition, in 1995 the Panel on Poverty and Family Assistance 
appointed by the National Research Council published a 501-page report, 
Measuring Poverty: A New Approach[Footnote 24], with a proposed new 
approach for developing an official U.S. poverty measure. The report 
included an extensive discussion (pp. 182-203) on how the proposed new 
poverty measure should be adjusted for geographic cost-of-living 
differences. This discussion did not mention the federal employee cost-
of-living allowance figures as a possible acceptable data source for 
adjusting the poverty measure for geographic cost-of-living 
differences. 

Furthermore, also in 1995 the U.S. General Accounting Office (as it was 
then called) issued a report, Poverty Measurement: Adjusting for 
Geographic Cost-of-Living Difference (GAO/GGD95-64, March 
1995)[Footnote 25]. For the report, GAO identified twelve methodologies 
that could potentially be used in adjusting poverty measures for 
geographic cost-of-living differences, and presented them to fifteen 
experts for their review. None of the twelve methodologies used or 
referred to the federal employee cost-of-living allowance figures. 

ASPE does not believe that it would be appropriate to use the current 
Office of Personnel Management (OPM—the successor to CSC) version of 
those figures to adjust either version of the federal poverty measure 
for the insular areas. 

See comment 1. 

Which Agency Used Federal Salary Figures to Develop Separate Poverty 
Guidelines for Alaska and Hawaii? 

The second paragraph and several other statements throughout the draft 
report implicitly refer to adjusting poverty guidelines by the current 
OPM federal employee cost-of-living allowance figures as "...the 
approach that HHS currently uses to determine the Alaska and Hawaii 
[poverty] guidelines." Strictly speaking, that reference is not 
accurate. 

As noted above, 0E0 used federal employee cost-of-living allowance 
figures to institute separate poverty guideline figures for Alaska and 
Hawaii during the 1966-1970 period. 0E0 and its successor agency, the 
Community Services Administration (CSA), continued to update and 
publish the poverty guidelines—including the higher Alaska and Hawaii 
guidelines—each year until 1981[Footnote 26] without any explanation of 
the higher Alaska and Hawaii guidelines nor any indication that federal 
employee cost-of-living allowances were considered again each year in 
calculating the Alaska and Hawaii differentials. 

When HHS was assigned responsibility for updating the poverty 
guidelines in 1982[Footnote 27], the Department calculated Alaska and 
Hawaii guidelines using the 25 percent and 15 percent differentials 
which it "grandfathered in" from the OEO/CSA guideline series. HHS did 
not consult OPM documents to get figures to calculate Alaska and Hawaii 
guidelines, and has continued to use the "grandfathered-in" 
differentials to calculate Alaska and Hawaii guidelines since 1982. As 
a result, the proposal in the GAO draft report to use OPM federal 
salary figures to adjust poverty guidelines for insular areas is most 
accurately described as an adaptation and update (using today's OPM 
figures rather than 1960s CSC figures) of what 0E0 did to calculate 
Alaska and Hawaii guidelines.

See comment 2. 

Poverty Thresholds for Insular Areas Consistent with Orshansky's 
Continental-U.S. Thresholds Would Probably Be Lower, Not Higher: 

ASPE believes that if it were possible to construct poverty lines for 
the insular areas that were conceptually consistent with the Orshansky 
poverty thresholds, those insular area poverty lines would probably be 
lower than the poverty thresholds for the continental U.S., not higher. 

See comment 3. 

In order to calculate poverty thresholds for the insular areas using 
the same methodology that Mollie Orshansky used to calculate poverty 
thresholds for the continental U.S., it would be necessary to have 
economy food plans developed for those areas in 1961, and Household 
Food Consumption Surveys for those areas taken in 1955. Such food plans 
and such surveys do not exist. However, it is still possible to say 
something about what the results of such calculations would probably 
be. As can be seen in the table on the next page, median family incomes 
for the insular areas for 1959 (the year closest to the dates of the 
economy food plan and the Household Food Consumption Survey used) were 
significantly lower than median family income for the continental U.S. 
It is known from Engel's Law[Footnote 28] that for lower incomes, the 
proportion of income spent on food is higher than for higher incomes. 
For the continental U.S., the 1955 Household Food Consumption Survey 
found that the average family spent about one third of its after-tax 
money income on food. By Engel's Law, the corresponding food shares for 
the insular areas for (approximately) 1955 would have been 
significantly higher than one third. If one assumes (for illustrative 
purposes) that the 1955 food share for one of the insular areas might 
have been one half, then the corresponding "multiplier" (the figure by 
which one multiplies the cost of a food plan to get a poverty 
threshold) for that area would have been two, as compared with the 
multiplier of three for the continental U.S. This means that the 
poverty threshold for that insular area would have been significantly 
lower than the threshold for the continental U.S.. 

Median Family Incomes for the U.S. and Insular Areas-1959[Footnote 
29]:  

(Alaska): $7,305); 
(Hawaii): (6,366); 
United States  (50 states and D.C.) 5,660; 
Guam: 4,549; 
U.S. Virgin Islands: 2,243; 
Puerto Rico: 1,268; 
American Samoa: 770; 
Northern Mariana Islands: N/A. 

These median family income figures indicate that it is questionable to 
assume that it is appropriate to simply take the ad hoc procedure for 
developing guidelines for Alaska and Hawaii, and apply it to the 
insular areas. Rather than being economically similar to the insular 
areas, Alaska and Hawaii had 1959 incomes roughly three to nine times 
as high as most of the insular areas; while Alaska and Hawaii had 
incomes higher than the U.S. median, the insular areas had incomes 
significantly lower than the U.S. median. 

One method of developing a rough estimate of poverty lines for the 
insular areas would be to simply assume that poverty lines should be 
directly proportional to median family income figures—for instance, if 
an area's median family income was half of the corresponding figure for 
the U.S. as a whole, then the poverty line for that area would be 50 
percent of the continental- U.S. poverty line. 

See comment 3. 

Another method of developing insular-area poverty lines from income 
figures would be to make use of findings concerning the income 
elasticity of the poverty line.[Footnote 30] For the U.S., there is 
extensive evidence—from both "expert"-devised minimum budgets and 
"subjective" low-income figures derived from Gallup Poll responses—that 
successive poverty lines developed as absolute poverty lines tend to 
rise in real terms as the real income of the general population rises. 

See comment 3. 

Rough-estimate poverty lines for different insular areas based on these 
two methods range between 14 percent and 86 percent of the continental-
U.S. poverty line. (These results are in sharp contrast to the poverty 
guideline figures for insular areas in the enclosure to the GAO draft 
report based on the OPM federal salary figures; those guidelines 
figures are between 114 and 125 percent of the continental-U.S. poverty 
line.) 

See comment 3. 

Based on this analysis, ASPE finds no evidence to support the proposal 
in the GAO draft report to have poverty guidelines for insular areas 
between 14 and 25 percent higher than the 48-state poverty guidelines. 
ASPE is not proposing that poverty guidelines for the insular areas 
should actually be lowered from the 48-state guideline level. At the 
same time, it would seem inequitable to give the insular areas higher 
poverty guidelines than those for the 48 states when the standards of 
living of the various insular areas are generally so much lower than 
that of the 48 states. 

See comment 3. 

[End of section] 

Enclosure III: GAO Comments: 

The following are GAO's responses to HHS's letter, dated October 8, 
2009. 

1. HHS states that the federal employee nonforeign area COLAs published 
by OPM are not of sufficient statistical quality for use in adjusting 
the federal poverty guidelines. HHS cites as evidence the fact that 
OPM's database for the nonforeign area COLAs was not recommended in 
studies by the Poverty Studies Task Force in 1976, the Panel on Poverty 
and Family Assistance in 1995, and GAO in 1995 as a potential data 
source for adjusting the poverty measure for geographic cost-of-living 
differences. However, the three studies that HHS cites explored the 
feasibility of adjusting the federal poverty measure for geographic 
differences in the 50 states. The nonforeign area COLAs' application 
has been restricted to the insular areas, Alaska, and Hawaii; these 
COLAs have not been developed as a database to be used throughout the 
United States. 

HHS also states that it does not believe that it would be appropriate 
to use the current OPM nonforeign area COLA data to adjust the poverty 
guidelines for the insular areas. We considered these data because they 
were used to construct the Alaska and Hawaii poverty guidelines in 1970 
and because these are the only poverty guidelines that have been 
established specific to any geographic areas. However, we are not 
making a recommendation for the use of this approach. 

2. HHS comments that it adapts and updates, rather than "determines"--
as the draft of our report stated--the Alaska and Hawaii poverty 
guidelines. HHS notes that it inherited the Alaska and Hawaii guideline 
series in 1982 from OEO, which originally instituted these guidelines 
in 1970. We changed the wording of our report to reflect this fact. HHS 
also states that it has not consulted OPM documents to get figures to 
calculate the guidelines for Alaska and Hawaii. Our report acknowledges 
this point in footnote 16. 

3. HHS moves beyond discussing whether there are differences in the 
cost of living between insular areas and the 48 contiguous states and 
focuses on alternative approaches for setting poverty thresholds. HHS 
observes that if historical data did exist for the insular areas and 
the methodology used to construct the original thresholds were applied, 
poverty thresholds for those areas would likely be lower, because 
median incomes in the insular areas historically have been 
significantly lower than those in the states and, as a result, the 
share of family food consumption as a fraction of insular area family 
expenditures would be higher.[  31] We note that although the 1955 food 
share in the insular areas might have been higher than in the 
contiguous states, it is also likely that if there were 1964 food plans 
for the areas, the cost of these food plans would also be higher 
because food is imported in those areas. Therefore, the higher cost of 
the food plans could diminish the effect of the higher food share in 
the insular areas, rendering uncertain the overall level of the insular 
areas' specific poverty thresholds. 

HHS also provides rough estimates of poverty thresholds based on two 
alternative methods. First, it estimates poverty thresholds as a 
percentage of median family income. We note that using this approach 
would lead to lower poverty thresholds for any state or other 
jurisdiction with lower median household incomes. The consequence would 
be that families in poorer states or other jurisdictions would have 
less access to need-based federal assistance than would families with 
the same income in more affluent areas. 

Second, it estimates poverty thresholds based on the responsiveness of 
poverty thresholds to changes in inflation-adjusted income over time. 
We note that under this method, areas with lower median household 
income and higher cost of living would have even lower poverty 
thresholds than the thresholds established with the first alternative 
approach that HHS describes. The consequence would be that families in 
poorer areas with higher cost of living would have a further reduction 
in access to federal assistance. 

HHS states that using these two approaches would produce poverty 
thresholds for different insular areas ranging between 14 and 86 
percent of the current U.S. poverty thresholds. HHS says that it does 
not advocate lower guidelines for the insular areas but finds it 
inequitable to have higher guidelines for those areas because of the 
lower median income in those areas than that of the 48 states. 

[End of section] 

Enclosure IV: Comments from the Census Bureau: 

Note: GAO comments supplementing those in the report text appear at the 
end the this enclosure. 

United States Department Of Commerce: 
The Under Secretary for Economic Affairs:
Washington, D.C. 20230: 

October 15, 2009: 

Mr. David Gootnick:  
Director, International Affairs and Trade: 
Government Accountability Office: 
441 G Street, NW: 
Washington, DC 20548: 

Dear Mr. Gootnick:   

Thank you for the opportunity to provide comments on the draft report 
entitled "Poverty Determination in U.S. Insular Areas" (GAO-09-1041R). 
We have no comments on the background section of the draft report, as 
it appears to accurately describe the development and use of U.S. 
poverty thresholds. 

We would, however, urge the Government Accountability Office (GAO) to 
study the issue of poverty guidelines for the insular areas further 
before making recommendations that would increase poverty guidelines by 
a significant amount in most of these areas. The 1995 National Academy 
of Sciences report, "Measuring Poverty: A New Approach," suggested that 
differences in housing costs may represent a viable basis for adjusting 
poverty thresholds when more comprehensive inter-area price adjustments 
are not available. The reasons for this suggestion include: 1) housing 
costs represent a significant portion of poverty budgets; 2) housing 
costs tend to differ more across geographic areas than other consumer 
items; and 3) housing cost differentials, across geographic areas, are 
often available to a greater extent than geographic cost differentials 
of other consumer items. 

See comment 1. 

The 2008 American Community Survey contains information about 2008 
housing costs in Puerto Rico, and Census 2000 results include housing 
data for Puerto Rico and all of the other insular areas. In addition, 
Housing and Urban Development Fair Market Rents are available for most 
of these areas. We suggest that GAO should examine these data to 
ascertain whether the suggested Office of Personnel Management COLA-
based adjustments seem reasonable in light of housing-cost 
differentials based on U.S. Census Bureau data and other sources. 

See comment 2. 

Again, thank you for the opportunity to comment on this draft report. 

Sincerely,

Signed by: 

Rebecca M. Blank: 

GAO Comments: 

The following are GAO's comments on the Census letter, dated October 
15, 2009. 

1. Census urges us to further study the poverty guidelines in the 
insular areas before making any recommendations that would increase 
those guidelines by a significant amount. Note that we have not made 
any recommendations in this report. 

2. Census suggests that we compare OPM's foreign area COLAs with 
insular area housing-cost differentials based on data such as the 2008 
American Community Survey (ACS), Census 2000, and the Housing and Urban 
Development Fair Market Rents. We note the following: 

* We did not consider alternative approaches for geographic cost-of- 
living adjustment of the poverty guidelines, such as the use of housing 
data, because we focused only on the approach used in 1970 to construct 
the Alaska and Hawaii poverty guidelines and because these are the only 
poverty guidelines that have been established specific to any 
geographic areas. 

* OPM analysis includes a housing cost component that is adjusted for 
differences in the quality of housing, using statistical analysis. OPM 
then aggregates housing and other types of expenditures in constructing 
its cost-of-living index. 

* GAO has identified challenges in the approach used to estimate the 
Housing and Urban Development Fair Market Rents and has recommended 
ways to improve the accuracy of Fair Market Rents estimates.[Footnote 
32] 

Enclosure V: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

David Gootnick, (202) 512-3149 or gootnickd@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Emil Friberg (Assistant 
Director), Gergana Danailova-Trainor, Kathleen Scholl, Thomas McCool, 
Marissa Jones, Reid Lowe, and Kathryn Larin made key contributions to 
this report.

[End of section] 

Footnotes: 

[1] According to the U.S. Census Bureau (Census), in 1999, the 
percentage of individuals in poverty ranged from 23 in Guam--nearly 
twice as high as the continental U.S. poverty rate of 12 percent--to 61 
in American Samoa (see encl. II). For more information about reliance 
on federal programs in American Samoa, the CNMI, Guam, and the USVI, 
see GAO, U.S. Insular Areas: Economic, Fiscal, and Financial 
Accountability Challenges, [hyperlink, http://www.gao.gov/cgi-
bin/getrpt?GAO-07-119] (Washington, D.C.: Dec. 12, 2006). 

[2] This reasoning does not take into account any differences in 
consumption patterns between federal employees and insular area poor 
populations. In addition, the OPM nonforeign area cost-of-living 
allowances for federal employees are limited by statute to 25 percent 
of basic pay. 

[3] The U.S. government's nonforeign area COLAs are adjustments to 
basic pay for federal white-collar workers based on differences in 
living costs between Alaska, Hawaii, and the insular areas (Guam and 
the Northern Mariana Islands, Puerto Rico, and the U.S. Virgin Islands) 
and Washington, D.C; OPM does not classify American Samoa as a 
nonforeign COLA area. To set the nonforeign area COLA rates, OPM 
surveys the prices of more than 300 items, including goods and 
services, housing, transportation, and miscellaneous expenses, in each 
of the areas and in the Washington, D.C., metropolitan area. Under Pub. 
Law No. 111-82, National Defense Authorization Act for Fiscal Year 
2010, Oct. 29, 2009, the nonforeign COLA will be phased out and 
locality pay will be phased in over a 3-year period starting Jan. 1, 
2010. The legislation also freezes the COLA rates that are in effect on 
the day of enactment of the act. The nonforeign area COLAs are a 
different adjustment than annual inflation adjustments (also commonly 
known as COLAs). The nonforeign area COLAs also differ from federal 
locality pay--that is, comparability payments in addition to basic pay, 
with the amount of locality pay based on differences between federal 
and private sector pay rates for particular sets of jobs within 
particular pay areas. In contrast to locality pay, nonforeign COLAs are 
not federally taxed and are not considered basic pay in determining an 
employee's retirement benefits, life insurance, or premium pay. 

[4] The original thresholds were developed in 1963-1964 by Mollie 
Orshansky, an economist working for the Social Security Administration 
(SSA). Orshansky used somewhat different procedures to calculate 
thresholds for one-and two-person units, to allow for small families' 
relatively larger fixed costs. 

[5] In measuring poverty, Census considers all before-tax cash income, 
except capital gains, and excludes all noncash benefits, such as food 
stamps. 

[6] Office of Management and Budget, Statistical Policy Directive No. 
14: Definition of Poverty for Statistical Purposes (Washington, D.C., 
1978). 

[7] For a one-person family unit, Census publishes two separate 
thresholds for unrelated individuals younger than 65 years and 65 years 
or older. For a two-person family unit, Census publishes four 
thresholds based on age: (1) both individuals are younger than 65 
years; (2) both individuals are 65 or older; (3) one individual is 
younger than 65 years and the other is a related child younger than 18 
years; and (4) one individual is 65 years or older and the other is a 
related child younger than 18 years. For a family unit of three or more 
people, Census publishes 42 thresholds that vary by family size and the 
number of related children who are younger than 18 years. 

[8] See [hyperlink, 
http://www.census.gov/hhes/www/poverty/povdef.html]. 

[9] A federal interagency committee met in 1980-1981 to revise the 
poverty definition. These modifications affected the number of poor and 
poverty rate only slightly and were documented in P60-133, 
Characteristics of the Population Below the Poverty Level: 1980. See 
[hyperlink, http://www.census.gov/hhes/www/povmeas/ombdir14.html]. 

[10] According to HHS, a number of federal programs use the poverty 
guidelines, or percentage multiples of the guidelines, as an 
eligibility criterion, while others do not. Agencies that use the 
guidelines as a criterion compare them with varying types of income 
(e.g., before tax or after tax, gross or net) to determine eligibility. 
For more detailed information about federal programs that use the 
guidelines for this purpose, see Congressional Research Service, Cash 
and Noncash Benefits for Persons with Limited Income: Eligibility 
Rules, Recipient and Expenditure Data, FY2002-FY2004, Order Code 
RL33340 (Washington, D.C., Library of Congress, 2006). 

[11] HHS updates the poverty guidelines at least annually as required 
by 42 U.S.C. 9902(2) and publishes the guidelines in the Federal 
Register. 

[12] In 1970, the Office of Economic Opportunity instituted separate 
poverty guidelines for Alaska and Hawaii that are higher than the 
continental U.S. (48-state) guidelines--respectively, 25 percent and 15 
percent higher than the national guidelines--in view of substantially 
higher costs of living in those states (see Federal Register, vol. 35, 
no. 70, April 10, 1970, p. 5948). The percentage adjustment of the 
poverty guidelines has remained constant since 1970. 

[13] The Bureau of Labor Statistics does not collect data to construct 
CPIs specific to the insular areas. In addition, the bureau does not 
collect data from the insular areas to include in the national CPI 
calculation. 

[14] See "Annual Update of the HHS Poverty Guidelines," Federal 
Register, vol. 73, no. 15, January 23, 2008, p. 3972. 

[15] CPIs have been computed periodically for the insular areas by 
other sources; however, these CPIs cannot be used to construct 
inflation-adjusted thresholds for the insular areas because no 1964 
insular area thresholds are available for use as baselines. 

[16] The adjustment factors for Alaska and Hawaii guidelines were set 
by the Office of Economic Opportunity in 1970; while current COLAs 
could be used to update and adjust the level of those guidelines on an 
annual basis, the guidelines have not been updated with more recent 
cost-of-living data. We considered the approach used in constructing 
poverty guidelines for Alaska and Hawaii because these are the only 
poverty guidelines that have been established specific to any 
geographic areas; we did not examine the universe of possible 
alternative approaches to defining poverty guidelines for the insular 
areas. 

[17] Likewise, the method initially used to establish the guidelines in 
Alaska and Hawaii does not take into account any differences in 
consumption patterns between Alaska and Hawaii federal employees and 
the poor populations in these states. 

[18] Some programs that use formulas to distribute funds to the insular 
areas base these allocations on total population size rather than on 
poverty statistics. 

[19] For details on how the annual limits are established, see GAO, 
Medicaid and CHIP: Opportunities Exist to Improve U.S. Insular Area 
Demographic Data That Could Be Used to Help Determine Federal Funding, 
[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-09-558R] (Washington, 
D.C.: June 30, 2009). 

[20] U.S. Department of Health, Education, and Welfare, The Measure of 
Poverty: A Report to Congress as Mandated by the Education Amendments 
of 1974 (Washington, D.C: 1976), available at [hyperlink, 
http://www.census.gov/hhes/www/povmeas/pdf/measureofpoverty.pdf]; 
Constance F. Citro and Robert T. Michael (ed.), Measuring Poverty: A 
New Approach (Washington, D.C.: National Academy Press, 1995), 
available at [hyperlink, 
http://www.census.gov/hhes/www/povmeas/toc.html]; and GAO, Poverty 
Measurement: Adjusting for Geographic Cost-of-Living Difference, 
[hyperlink, GAO/GGD-95-64] (Washington, D.C.: March 9, 1995). 

[21] HHS bases the argument on the empirical relationship discovered by 
Engel that poorer families spend a larger fraction of their income on 
food than richer families. We note that although the 1955 food share in 
the insular areas might have been higher than in the contiguous states, 
it is likely that if there were 1964 food plans for the areas, the cost 
of these food plans would also be higher because food is imported in 
those areas. Therefore, the higher cost of the food plans could 
diminish the effect of the higher food share in the insular areas, 
rendering uncertain the overall level of the insular areas’ specific 
poverty thresholds. 

[22] See Israel Putnam, "Poverty Thresholds: Their History and Future 
Development" [November 1970], pp. 277, 280, and 281 in Mollie Orshansky 
[compiler], Documentation of Background Information and Rationale for 
Current Poverty Matrix [hyperlink, 
http://www.census.nov/hhes/www/povmeas/pdf/tp_i.pdf] (Technical Paper I 
of The Measure of Poverty), Washington, D.C., U.S. Department of 
Health, Education, and Welfare, 1977; and U.S. Office of Economic 
Opportunity, "Guidelines for Alaska and Hawaii," Federal Register, Vol. 
35, No. 70, April 10, 1970, p. 5948. 

[23] See U.S. Department of Health, Education, and Welfare, The Measure 
of Poverty: A Report to Congress as Mandated by The Education 
Amendments of 1974 [hyperlink, 
http://www.census.gov/hhes/www/povmeas/pdf/measureofpoverty.pdf] 
Washington, D.C., U.S. Government Printing Office, April 1976. 

[24] See Constance F. Citro and Robert T. Michael (editors), Measuring 
Poverty: A New Approach [hyperlink, 
http://www.census.aov/hhes/www/povmeas/toc.html], Washington, D.C., 
National Academy Press, 1995. 

[25] See U.S. General Accounting Office, Poverty Measurement: Adjusting 
for Geographic Cost-of-Living Difference ([hyperlink, 
http://www.gao.gov/cgi-bin/getrpt?GAO/GGD-95-64]) [hyperlink, 
http://www.gao.gov/archive/1995/ge95064.pdf], March 1995. 

[26] See Community Services Administration, "45 CFR Part 1060—General 
Characteristics of Community Action Programs; CSA Income Poverty 
Guidelines," Federal Register, Vol. 46, No. 43, March 5, 1981, pp. 
15270- 15271. 

[27] See Department of Health and Human Services, Office of the 
Secretary, "Annual Revision of Poverty Income Guidelines," Federal 
Register, Vol. 47, No. 69, April 9, 1982, pp. 15417-15418. 

[28] Ernst Engel, a 19th-century German statistician, formulated the 
empirical relationship which became known as Engel's Law: "The poorer 
is a family, the greater is the proportion of the total outgo which 
must be used for food....The proportion of the outgo used for food, 
other things being equal, is the best measure of the material standard 
of living of a population." (See Carle C. Zimmerman, "Ernst Engel's Law 
of Expenditures for Food," Quarterly Journal of Economics, Vol. 47, No. 
1, November 1932, P. 80.) 

[29] See U.S. Bureau of the Census, Statistical Abstract of the United 
States 1969 (90th Annual Edition), Washington, D.C., U.S. Department of 
Commerce, pp. 324 and 814. 

[30] See Citro and Michael (cited in footnote 3), pp. 140-141 and 143-
144; and G. M. Fisher, "Is There Such a Thing as an Absolute Poverty 
Line Over Time? Evidence from the United States, Britain, Canada, and 
Australia on the Income Elasticity of the Poverty Line" [hyperlink, 
http://www.census.gov/hhes/www/povmeas/papers/elastap4.html] (Poverty 
Measurement Working Paper, U.S. Census Bureau web site), August 1995. 

[31] HHS bases the argument on the empirical relationship discovered by 
Engel that poorer families spend a larger fraction of their income on 
food than richer families (see comments from HHS, footnote 7). 

[32] See GAO, Rental Housing: HUD Can Improve Its Process for 
Estimating Fair Market Rents, [hyperlink, http://www.gao.gov/cgi-
bin/getrpt?GAO-05-342] (Washington, D.C.: March 31, 2005). 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the performance 
and accountability of the federal government for the American people. 
GAO examines the use of public funds; evaluates federal programs and 
policies; and provides analyses, recommendations, and other assistance 
to help Congress make informed oversight, policy, and funding 
decisions. GAO's commitment to good government is reflected in its core 
values of accountability, integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 
[hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 
information. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

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

Congressional Relations: 

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

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: