This is the accessible text file for GAO report number GAO-13-434 
entitled 'Medicaid: Alternative Measures Could Be Used to Allocate 
Funding More Equitably' which was released on May 10, 2013. 

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

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

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

May 2013: 

Medicaid: 

Alternative Measures Could Be Used to Allocate Funding More Equitably: 

GAO-13-434: 

GAO Highlights: 

Highlights of GAO-13-434, a report to congressional requesters. 

Why GAO Did This Study: 

Medicaid is the largest federal program assisting states in financing 
medical and health-related services for certain low-income 
individuals. States and the federal government share in the financing 
of the Medicaid program, with the federal government matching most 
state expenditures for Medicaid services on the basis of a statutory 
formula known as the FMAP, which is based solely on state PCI in 
relation to national PCI. Prior GAO work has raised concerns about the 
FMAP, noting that PCI does not accurately represent states’ 
populations in need of Medicaid services or states’ ability to finance 
services, and does not account for geographic cost differences among 
states. GAO was asked to examine ways to improve the allocation of 
Medicaid funding. 

GAO considered whether available data sources could be used to develop 
measures to more equitably allocate Medicaid funding. To do so, GAO 
reviewed its prior reports and other studies, examined data sources 
produced by federal agencies, and illustrated how selected data could 
be used to develop measures to allocate Medicaid funding. GAO based 
its analysis on commonly used equity standards and focused its efforts 
on readily available data sources, which are not inclusive of all 
possibilities. 

The Department of Health and Human Services provided technical 
comments, which GAO incorporated as appropriate. 

What GAO Found: 

GAO identified multiple data sources that could be used to develop 
measures to allocate Medicaid funding to states more equitably than 
the current funding formula-—known as the Federal Medical Assistance 
Percentage (FMAP)-—which is based solely on per capita income (PCI). 
To be equitable from the perspective of beneficiaries and allow states 
to provide a comparable level of services to each person in need, a 
funding allocation mechanism should take into account the demand for 
services in each state and geographic cost differences among states. 
To be equitable from the perspective of taxpayers, an allocation 
mechanism should ensure that taxpayers in poorer states are not more 
heavily burdened than those in wealthier ones, by taking into account 
state resources. To illustrate, GAO identified at least one federal 
data source that could be used to develop measures of each of these 
aspects, in order to allocate Medicaid funding more equitably. 

* Demand for services. A measure of the demand for Medicaid services 
should account for both the size of the target population in need of 
services and the health services needs of that population. Nationally 
representative federal surveys, such as the U.S. Census Bureau’s 
American Community Survey (ACS) and Current Population Survey (CPS), 
are available data sources that can be used to directly estimate the 
number of persons residing in each state with incomes low enough to 
qualify them as potentially in need of Medicaid services. These 
estimates can then be adjusted to reflect variation in health services 
needs within the identified population, using additional information 
collected in the surveys or from data sources external to the surveys, 
such as Medicaid data on enrollment or spending. 

* Geographic cost differences. A measure of geographic cost 
differences should account for all components of health care costs, 
including the cost of the personnel who provide services, which 
represents the greatest share of costs. National data that can be used 
to estimate average wages for health care personnel by state include 
the Occupational Employment Statistics survey conducted by the Bureau 
of Labor Statistics. 

* State resources. A measure of state resources should account for all 
income-—regardless of whether the state taxes the income or not. While 
PCI includes the personal income of state residents, it excludes other 
taxable income, such as undistributed corporate profits. In contrast, 
the Total Taxable Resources (TTR) measure, as generated by the 
Department of the Treasury from multiple data sources, comprises not 
only the income included in PCI but also other significant sources of 
taxable income. As a result, nationwide, the TTR measure of income was 
42 percent larger on a per capita basis than PCI in 2010, and provided 
a more comprehensive measure of state resources. 

GAO’s analysis shows that measures of the demand for services, 
geographic cost differences, and state resources can be combined in 
various ways to provide a basis for allocating Medicaid funds more 
equitably among states. 

View [hyperlink, http://www.gao.gov/products/GAO-13-434]. For more 
information, contact Carolyn L. Yocom at (202) 512-7114 or 
yocomc@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Available Data Sources Could Be Used to Develop Measures to Allocate 
Medicaid Funding More Equitably: 

Agency Comments: 

Appendix I: Examples of Alternative Measures That Could Be Used to 
More Equitably Allocate Federal Medicaid Funding: 

Appendix II: GAO Contact and Staff Acknowledgments: 

Related GAO Products: 

Tables: 

Table 1: Estimates of Population in Poverty, by State: 

Table 2: Estimates of Medicaid-Service-Need Weights by Age and 
Disability Category: 

Table 3: Estimates of Demand for Services, Including Adjustment for 
Service Need Based on the Composition of the Medicaid Population, by 
State: 

Table 4: Estimates of Demand for Services, Including Adjustment for 
Service Need Based on the Composition of the State Poverty Population, 
by State: 

Table 5: Estimates of Cost Indexes for Health Care Services, by State: 

Table 6: Average Total Taxable Resources (TTR) and Per Capita TTR, by 
State, 2008 through 2010: 

Abbreviations: 

ACS: American Community Survey: 

CMS: Centers for Medicare & Medicaid Services: 

CPS: Current Population Survey: 

FMAP: Federal Medical Assistance Percentage: 

FPL: federal poverty level: 

HHS: Department of Health and Human Services: 

IPPS: Inpatient Prospective Payment System: 

MACPAC: Medicaid and CHIP Payment and Access Commission: 

MedPAC: Medicare Program Advisory Commission: 

MSIS: Medicaid Statistical Information System: 

OES: Occupational Employment Statistics: 

PCI: per capita income: 

PPACA: Patient Protection and Affordable Care Act: 

TTR: Total Taxable Resources: 

[End of section] 

United States Government Accountability Office: 
GAO: 
441 G St. N.W. 
Washington, DC 20548: 

May 10, 2013: 

The Honorable Orrin G. Hatch: 
Ranking Member: 
Committee on Finance: 
United States Senate: 

The Honorable Fred Upton: 
Chairman: 
Committee on Energy and Commerce: 
House of Representatives: 

Created in 1965, Medicaid is the largest federal program assisting 
states in financing medical and health-related services for certain 
low-income individuals, such as children and individuals who are 
disabled or elderly. Within broad federal requirements, states have 
some flexibility in deciding the range of medical services to provide 
and which individuals to cover, and state Medicaid programs 
accordingly vary in scope. In fiscal year 2011, Medicaid served an 
estimated 70 million enrollees and had expenditures totaling an 
estimated $432.4 billion, $275.1 billion of which was financed by the 
federal government, with the rest financed by the states. Medicaid 
enrollment and expenditures have increased significantly over the most 
recent 10-year period for which data are available--enrollment growth 
averaged 4.2 percent per year between 2001 and 2011, and expenditures 
grew, on average, by 6.5 percent per year. 

States and the federal government share in the financing of the 
Medicaid program, with the federal government matching most state 
expenditures for Medicaid services on the basis of a statutory formula 
known as the Federal Medical Assistance Percentage (FMAP). The FMAP 
calculates the federal matching rate for each state on the basis of 
the state's per capita income (PCI) in relation to the national PCI. 
[Footnote 1] Under the FMAP, the federal government pays a larger 
portion of Medicaid expenditures in states with low PCI relative to 
the national average, and a smaller portion for states with higher 
PCIs. PCI is used in the formula as a proxy for both state resources 
and the low-income population in need of Medicaid services in each 
state.[Footnote 2] 

Our prior work has noted multiple concerns regarding how the FMAP 
formula allocates funds among states. Specifically, we have noted that 
PCI, the sole measure included in the FMAP formula, is an incomplete 
measure of states' resources, is a poor proxy for the size or 
characteristics of the states' population in need of Medicaid 
services, and does not take into account geographic differences in the 
cost of providing health care services.[Footnote 3] In addition, we 
have expressed concerns that the FMAP does not take into account 
states' current economic circumstances due to lags in the calculation 
of PCI and implementation of the FMAP.[Footnote 4] 

Given these concerns regarding the Medicaid formula and the overall 
growth in Medicaid enrollment and expenditures, you asked us to 
examine ways to improve the allocation of Medicaid funding. In this 
report, we considered whether available data sources could be used to 
develop measures to more equitably allocate Medicaid funding. To do 
this, we applied two equity standards that we and others have commonly 
used to evaluate federal funding mechanisms: beneficiary equity, which 
stipulates that funds should be distributed so that each state can 
provide a comparable level of services to each person in need; and 
taxpayer equity, which stipulates that each state should receive 
enough funds to be able to provide a comparable level of services as 
others while contributing about the same proportion of its own 
resources. With these standards in mind, we reviewed prior GAO reports 
and other relevant studies, examined potential data sources produced 
by federal agencies, and developed illustrative examples of how 
selected data sources--largely federal statistical data, such as data 
from surveys conducted by the U.S. Census Bureau--could be used to 
develop measures to more equitably allocate Medicaid funding. We 
reviewed relevant documentation for each data source that we used and 
determined that the data were sufficiently reliable for our purposes. 
We focused our efforts on readily available data sources, most of 
which had been used in our prior reports. Accordingly, the data 
sources we discuss are not inclusive of all possibilities. Appendix I 
in this report provides additional details of our analysis. 

We conducted this performance audit from September 2012 to May 2013 in 
accordance with generally accepted government auditing standards. 
Those standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe 
that the evidence obtained provides a reasonable basis for our 
findings and conclusions based on our audit objectives. 

Background: 

In prior work spanning more than three decades, we have emphasized 
that, in federal-state programs such as Medicaid, funds should be 
allocated to states in a manner that is equitable from the 
perspectives of both beneficiaries and taxpayers.[Footnote 5] To be 
equitable from the perspective of beneficiaries, a funding allocation 
mechanism should take into account: 

* demand for services, which varies with the size and characteristics 
of the target population; and: 

* geographic cost differences, which affect the level of funding 
needed by each state to provide a comparable level of services to each 
person in need. 

In general, the demand for services depends not only on the size of 
the target population, but also on the amount or level of services 
required, which may vary with the characteristics of the population. 
An allocation mechanism that adjusts for cost differences would 
reflect the variation in the costs of key inputs used to provide 
services, such as the cost of the personnel involved.[Footnote 6] 

To be equitable from the perspective of taxpayers, a funding formula 
should ensure that taxpayers in poorer states are not more heavily 
burdened than those in wealthier ones. To do so, it must take into 
account each state's ability to finance its share of the costs of a 
given program from its own sources, which we refer to as state 
resources.[Footnote 7] A good measure of a state's resources would 
include all types of taxable resources and would not be affected by an 
individual state's taxing authorities or policies, in that it would 
measure all income flows and not just those that a state currently 
taxes. 

Our prior work has found that the FMAP does not adequately address the 
demand for services, geographic cost differences, and state resources. 
As we previously reported, PCI is a poor proxy for the size of a 
state's population in need of Medicaid services, as two states with 
similar PCI can have substantially different numbers of low-income 
residents.[Footnote 8] Moreover, PCI does not take into account 
differences among states in the health care services needs of this 
population. As we have reported, persons who are elderly typically use 
health care services at higher rates than younger adults and children. 
[Footnote 9] Similarly, disabled individuals typically have greater 
health care needs than nondisabled individuals. Accordingly, a state 
with the same number of Medicaid-eligible individuals, but a higher 
number of Medicaid-eligible elderly or disabled individuals than 
another state, may face a higher demand for Medicaid services. For 
example, in fiscal year 2010, Medicaid spent approximately $15,000 for 
each beneficiary aged 65 or older and $17,000 for each disabled 
beneficiary, compared with approximately $3,000 per child and $4,000 
per adult under age 65. 

Additionally, the FMAP does not include any measure of geographic 
differences in the costs of providing health care services, which can 
vary widely. For example, in our work on Medicare payment policy, we 
have noted that when such differences were taken into account, 
physicians practicing in the District of Columbia received a higher 
overall payment for the same service--20 percent higher for a mid-
level office visit in 2007--compared with physicians practicing in 
South Carolina.[Footnote 10] PCI does not account for these cost 
differences, as they are largely driven by the cost of the personnel 
who provide the services (wages, for example). 

Finally, PCI does not adequately account for state resources. Although 
PCI measures income received by state residents, such as wages, rents, 
and interest income, it does not include other components of state 
resources, such as corporate income produced within the state but not 
received by state residents. We previously reported that PCI 
especially understated the taxable resources in energy-exporting 
states, such as Alaska and Wyoming, and in states that house numerous 
corporate headquarters, such as Delaware.[Footnote 11] 

Available Data Sources Could Be Used to Develop Measures to Allocate 
Medicaid Funding More Equitably: 

We identified multiple data sources that could be used to develop 
measures of the demand for Medicaid services, geographic cost 
differences, and state resources. These measures could be combined in 
various ways to provide a basis for allocating Medicaid funds more 
equitably among states. 

Demand for services. A measure of demand for Medicaid services should 
account for both the size of the target population in need of services 
and the health services needs of that population. Nationally 
representative federal surveys, such as the U.S. Census Bureau's 
American Community Survey (ACS) and Current Population Survey (CPS), 
are available data sources that can be used to directly estimate the 
number of persons residing in each state with incomes low enough to 
qualify them as potentially in need of Medicaid services.[Footnote 12] 
The specific income level chosen for this measure, such as the federal 
poverty level (FPL), could vary depending on Medicaid eligibility 
criteria for the target population.[Footnote 13] These estimates could 
then be adjusted to reflect variation in health services needs within 
the identified population. These adjustments could be based on 
additional information collected in the surveys or on sources external 
to the surveys, such as Medicaid data on enrollment or spending. In 
appendix I, we illustrate how ACS data could be used, alone or in 
combination with Medicaid data, to measure states' relative demand for 
Medicaid services. 

Geographic cost differences. A measure of geographic differences in 
the cost of providing health care services should account for the cost 
of the personnel who provide the services, the cost of medical 
equipment and supplies, and the rental cost of facilities in which the 
services are provided. Of these three components, personnel costs 
represent the greatest share of total costs. For example, in the 2012 
Medicare physician fee schedule, the Centers for Medicare & Medicaid 
Services (CMS) attributed an average of 72 percent of the total cost 
of physician services to personnel costs,[Footnote 14] which vary 
substantially across geographic areas. Data that can be used to 
estimate average wages for health care personnel by state are 
available through several national data-collection efforts, including 
the Occupational Employment Statistics (OES) survey conducted by the 
Bureau of Labor Statistics and the cost reports that hospitals must 
submit to CMS in order to be paid by Medicare under its acute care 
hospital Inpatient Prospective Payment System (IPPS).[Footnote 15] In 
appendix I, we illustrate how OES wage data could be used to measure 
geographic cost differences. 

State Resources. A measure of state resources should account for all 
income, regardless of whether the state taxes the income or not, as 
states' taxing authorities or policies may vary. While PCI--the 
measure of state resources used in the current Medicaid funding 
formula--includes the personal income of state residents, it excludes 
other taxable income, such as undistributed corporate profits. In 
contrast, Total Taxable Resources (TTR) comprises not only the income 
included in PCI but also other significant sources of taxable income. 
For example, TTR includes personal income received by state residents 
as well as income produced within a state but received by individuals 
who reside out-of-state, which is considered a portion of the Gross 
State Product. As a result, nationwide, the TTR measure of income was 
42 percent larger on a per capita basis than PCI in 2010. In appendix 
I, we illustrate how estimates of state TTR, generated by the 
Department of the Treasury from data provided primarily by the Bureau 
of Economic Analysis, could be used to measure state resources. 

Combining Measures. Measures of the demand for services, geographic 
cost differences, and state resources could be combined in various 
ways to provide a basis for allocating Medicaid funds equitably among 
states.[Footnote 16] For example, when determining states' ability to 
fund Medicaid services, rather than simply considering total state 
resources or state resources per capita, a funding formula could 
reflect state resources in relation to the population in need of 
Medicaid services--that is, in relation to demand for services. This 
would result in a more equitable allocation of funding, because two 
states with similar resources and populations may have very different 
numbers of residents in need of Medicaid services. 

For example, one way to combine measures would be to calculate each 
state's TTR per person in poverty, using a measure of the poverty 
population that is adjusted to account for both the service needs of 
the population and geographic differences in the cost of services. 
[Footnote 17] We have previously found that such a combination of 
measures could be the basis of an alternative to the existing FMAP 
formula, whereby this calculation of TTR per person in poverty would 
replace the use of PCI in the formula.[Footnote 18] These measures 
could also be combined in other ways and could be used to take into 
account specific circumstances, such as economic downturns.[Footnote 
19] 

Agency Comments: 

We provided the Secretary of Health and Human Services with a draft of 
this report. The Department of Health and Human Services provided 
technical comments, which we incorporated as appropriate. 

We are sending copies of this report to the Administrator of CMS and 
other interested parties. In addition, the report will be available at 
no charge on the GAO website at [hyperlink, http://www.gao.gov]. 

If you or your staff have any questions about this report, please 
contact me at (202) 512-7114 or yocomc@gao.gov. Contact points for our 
Offices of Congressional Relations and Public Affairs may be found on 
the last page of this report. GAO staff who made key contributions to 
this report are listed in appendix II. 

Signed by: 

Carolyn L. Yocom: 
Director, Health Care: 

[End of section] 

Appendix I: Examples of Alternative Measures That Could Be Used to 
More Equitably Allocate Federal Medicaid Funding: 

This appendix illustrates how selected data sources could be used to 
measure (1) the demand for services, (2) geographic differences in the 
cost of health services, and (3) state resources. We considered and 
evaluated multiple data sources, focusing our efforts on federally 
available data sources, most of which had been used in our prior 
reports.[Footnote 20] 

Demand for Services: 

To illustrate possible measures of the demand for Medicaid services, 
we used data from the American Community Survey (ACS) to estimate the 
size of the target population in need of services and then used ACS 
data and Medicaid program data from the Centers for Medicare & 
Medicaid Services (CMS) to adjust for differences in the health 
services needs of the target population in each state. 

Measuring the Size of the Target Population: 

To illustrate one possible measure of the size of a potential target 
population by state, we used data and population weights from the ACS 
3-year file for 2009-2011 to estimate the number of people living in 
each state with family incomes at or below 100 percent of the federal 
poverty level (FPL),[Footnote 21] hereafter referred to as our 
unadjusted estimates of the population in poverty. We chose the ACS as 
our data source because of its large sample size and because its 
sample includes people living in institutions, such as nursing homes. 
Residents of nursing homes are an important component of the potential 
Medicaid-eligible population but are not included in other key 
surveys, such as the Current Population Survey. While the ACS does not 
provide family income data for respondents living in institutions, we 
included respondents living in institutions in the population in 
poverty if they reported receipt of Medicaid. 

In addition to using the ACS data to generate unadjusted estimates of 
the population in poverty, we also used the data to generate an 
adjusted estimate of the population in poverty that takes into account 
cost-of-living differences among states, such as differences in the 
costs of renting an apartment. To do so, we used state-specific 
regional price parity estimates published by the Bureau of Economic 
Analysis to adjust family income relative to the FPL prior to 
estimating the population in poverty.[Footnote 22] For states with 
higher than average cost of living, this adjustment has the effect of 
increasing the amount of income that a family can have while still 
being considered to be in poverty, while the reverse is true for 
states with lower than average cost of living. For example, table 1 
shows that in Alaska the number of people counted as being in poverty 
is 8.1 percent higher after the cost-of-living adjustment, because 
Alaska has a higher than average cost of living. Conversely, the 
number of people counted as being in poverty in Wyoming is 8.7 percent 
lower after the cost of living adjustment, because Wyoming has a lower 
than average cost of living. Additional possible options for measuring 
the size of the target population include using higher or lower family 
income thresholds (for example, 133 percent of FPL) for specific 
population subgroups or for the entire population, or calculating 
family income in a manner that more closely corresponds to Medicaid 
eligibility rules than does the ACS family income variable.[Footnote 
23] 

Table 1: Estimates of Population in Poverty, by State: 

State: Alabama; 
Population in poverty: 893,216; 
Cost-of-living-adjusted population in poverty: 789,200; 
Percentage difference: -11.6%. 

State: Alaska; 
Population in poverty: 72,387; 
Cost-of-living-adjusted population in poverty: 78,215; 
Percentage difference: 8.1%. 

State: Arizona; 
Population in poverty: 1,136,376; 
Cost-of-living-adjusted population in poverty: 1,123,273; 
Percentage difference: -1.2%. 

State: Arkansas; 
Population in poverty: 548,677; 
Cost-of-living-adjusted population in poverty: 473,915; 
Percentage difference: -13.6%. 

State: California; 
Population in poverty: 5,929,053; 
Cost-of-living-adjusted population in poverty: 6,717,844; 
Percentage difference: 13.3%. 

State: Colorado; 
Population in poverty: 687,964; 
Cost-of-living-adjusted population in poverty: 680,010; 
Percentage difference: -1.2%. 

State: Connecticut; 
Population in poverty: 385,454; 
Cost-of-living-adjusted population in poverty: 426,196; 
Percentage difference: 10.6%. 

State: Delaware; 
Population in poverty: 108,723; 
Cost-of-living-adjusted population in poverty: 113,331; 
Percentage difference: 4.2. 

State: District of Columbia; 
Population in poverty: 113,653; 
Cost-of-living-adjusted population in poverty: 125,540; 
Percentage difference: 10.5%. 

State: Florida; 
Population in poverty: 3,079,970; 
Cost-of-living-adjusted population in poverty: 3,079,970; 
Percentage difference: 0.0%. 

State: Georgia; 
Population in poverty: 1,727,122; 
Cost-of-living-adjusted population in poverty: 1,607,924; 
Percentage difference: -6.9%. 

State: Hawaii; 
Population in poverty: 153,295; 
Cost-of-living-adjusted population in poverty: 179,899; 
Percentage difference: 17.4%. 

State: Idaho; 
Population in poverty: 240,044; 
Cost-of-living-adjusted population in poverty: 220,514; 
Percentage difference: -8.1%. 

State: Illinois; 
Population in poverty: 1,846,804; 
Cost-of-living-adjusted population in poverty: 1,846,804; 
Percentage difference: 0.0%. 

State: Indiana; 
Population in poverty: 995,798; 
Cost-of-living-adjusted population in poverty: 916,037; 
Percentage difference: -8.0%. 

State: Iowa; 
Population in poverty: 392,032; 
Cost-of-living-adjusted population in poverty: 341,986; 
Percentage difference: -12.8%. 

State: Kansas; 
Population in poverty: 389,181; 
Cost-of-living-adjusted population in poverty: 340,643; 
Percentage difference: -12.5%. 

State: Kentucky; 
Population in poverty: 829,055; 
Cost-of-living-adjusted population in poverty: 730,839; 
Percentage difference: -11.8%. 

State: Louisiana; 
Population in poverty: 855,925; 
Cost-of-living-adjusted population in poverty: 796,600; 
Percentage difference: -6.9%. 

State: Maine; 
Population in poverty: 183,849; 
Cost-of-living-adjusted population in poverty: 177,418; 
Percentage difference: -3.5%. 

State: Maryland; 
Population in poverty: 582,585; 
Cost-of-living-adjusted population in poverty: 646,042; 
Percentage difference: 10.9%. 

State: Massachusetts; 
Population in poverty: 770,818; 
Cost-of-living-adjusted population in poverty: 838,820; 
Percentage difference: 8.8%. 

State: Michigan; 
Population in poverty: 1,648,959; 
Cost-of-living-adjusted population in poverty: 1,561,684; 
Percentage difference: -5.3%. 

State: Minnesota; 
Population in poverty: 628,483; 
Cost-of-living-adjusted population in poverty: 606,052; 
Percentage difference: -3.6%. 

State: Mississippi; 
Population in poverty: 664,170; 
Cost-of-living-adjusted population in poverty: 576,936; 
Percentage difference: -13.1%. 

State: Missouri; 
Population in poverty: 938,980; 
Cost-of-living-adjusted population in poverty: 812,178; 
Percentage difference: -13.5%. 

State: Montana; 
Population in poverty: 144,955; 
Cost-of-living-adjusted population in poverty: 134,372; 
Percentage difference: -7.3%. 

State: Nebraska; 
Population in poverty: 232,418; 
Cost-of-living-adjusted population in poverty: 203,135; 
Percentage difference: -12.6%. 

State: Nevada; 
Population in poverty: 401,723; 
Cost-of-living-adjusted population in poverty: 397,253; 
Percentage difference: -1.1%. 

State: New Hampshire; 
Population in poverty: 117,629; 
Cost-of-living-adjusted population in poverty: 123,594; 
Percentage difference: 5.1%. 

State: New Jersey; 
Population in poverty: 925,547; 
Cost-of-living-adjusted population in poverty: 1,047,850; 
Percentage difference: 13.2%. 

State: New Mexico; 
Population in poverty: 415,761; 
Cost-of-living-adjusted population in poverty: 386,481; 
Percentage difference: -7.0%. 

State: New York; 
Population in poverty: 3,001,976; 
Cost-of-living-adjusted population in poverty: 3,464,867; 
Percentage difference: 15.4%. 

State: North Carolina; 
Population in poverty: 1,656,411; 
Cost-of-living-adjusted population in poverty: 1,517,681; 
Percentage difference: -8.4%. 

State: North Dakota; 
Population in poverty: 78,796; 
Cost-of-living-adjusted population in poverty: 68,993; 
Percentage difference: -12.4%. 

State: Ohio; 
Population in poverty: 1,882,856; 
Cost-of-living-adjusted population in poverty: 1,698,819; 
Percentage difference: -9.8%. 

State: Oklahoma; 
Population in poverty: 649,564; 
Cost-of-living-adjusted population in poverty: 570,564; 
Percentage difference: -12.2%. 

State: Oregon; 
Population in poverty: 614,086; 
Cost-of-living-adjusted population in poverty: 593,279; 
Percentage difference: -3.4%. 

State: Pennsylvania; 
Population in poverty: 1,715,637; 
Cost-of-living-adjusted population in poverty: 1,680,290; 
Percentage difference: -2.1%. 

State: Rhode Island; 
Population in poverty: 144,196; 
Cost-of-living-adjusted population in poverty: 142,555; 
Percentage difference: -1.1%. 

State: South Carolina; 
Population in poverty: 847,614; 
Cost-of-living-adjusted population in poverty: 774,451; 
Percentage difference: -8.6%. 

State: South Dakota; 
Population in poverty: 118,358; 
Cost-of-living-adjusted population in poverty: 102,631; 
Percentage difference: -13.3%. 

State: Tennessee; 
Population in poverty: 1,133,239; 
Cost-of-living-adjusted population in poverty: 1,021,628; 
Percentage difference: -9.8%. 

State: Texas; 
Population in poverty: 4,553,162; 
Cost-of-living-adjusted population in poverty: 4,397,261; 
Percentage difference: -3.4%. 

State: Utah; 
Population in poverty: 363,480; 
Cost-of-living-adjusted population in poverty: 341,294; 
Percentage difference: -6.1%. 

State: Vermont; 
Population in poverty: 70,410; 
Cost-of-living-adjusted population in poverty: 70,040; 
Percentage difference: -0.5%. 

State: Virginia; 
Population in poverty: 937,069; 
Cost-of-living-adjusted population in poverty: 964,561; 
Percentage difference: 2.9%. 

State: Washington; 
Population in poverty: 918,816; 
Cost-of-living-adjusted population in poverty: 937,728; 
Percentage difference: 2.1%. 

State: West Virginia; 
Population in poverty: 338,273; 
Cost-of-living-adjusted population in poverty: 291,751; 
Percentage difference: -13.8%. 

State: Wisconsin; 
Population in poverty: 767,774; 
Cost-of-living-adjusted population in poverty: 699,022; 
Percentage difference: -9.0%. 

State: Wyoming; 
Population in poverty: 64,480; 
Cost-of-living-adjusted population in poverty: 58,865; 
Percentage difference: -8.7%. 

Total: United States; 
Population in poverty: 47,896,803; 
Cost-of-living-adjusted population in poverty: 47,496,835; 
Percentage difference: -0.8%. 

Source: GAO analysis of U.S. Census Bureau data. 

Note: Data are 2009-2011 data from the American Community Survey 
(ACS). To account for cost-of-living differences, we used state-
specific regional price parity estimates published by the Bureau of 
Economic Analysis to adjust family income relative to the federal 
poverty level (FPL). 

[End of table] 

Adjusting for Differences in Health Services Needs: 

An ideal estimate of the demand for Medicaid services would take into 
account the health needs of the target population--defined in this 
example as the population in poverty--based on individuals' specific 
health conditions and the costs of services to treat those conditions, 
an approach known as risk adjustment. A simplified approach is to 
identify categories--such as age or disability categories that 
describe the composition of the population--that are known to be 
correlated with different levels of health service needs and 
expenditures as a proxy for the health service needs of individuals 
within the categories. For example, the Medicaid and CHIP Payment and 
Access Commission (MACPAC) estimated that disabled individuals 
represented about 15 percent of Medicaid enrollees in fiscal year 
2009, yet accounted for about 43 percent of Medicaid benefit spending 
that year--in line with their greater need for health 
services.[Footnote 24] We used age and disability categories as proxy 
measures of the health services needs of this population. We 
distinguished the following categories, which we used to weight the 
estimated population in poverty obtained from the ACS: 

* children ages 0 through 18, 

* adults ages 19 through 64, 

* individuals with disabilities ages 0 through 64, 

* elderly individuals ages 65 through 84, and: 

* elderly individuals ages 85 and older. 

Other possible options for accounting for health service needs include 
using more detailed age categories; separating out certain Medicaid 
eligibility categories,[Footnote 25] such as foster children; or using 
age categories without consideration of disability status, as we did 
in a prior report.[Footnote 26] 

We used CMS Medicaid program data to estimate per-enrollee Medicaid 
spending in order to create Medicaid-service-need weights specific to 
each of the five categories. CMS has two key data sets that report 
state Medicaid expenditures--the Medicaid Statistical Information 
System (MSIS) and CMS-64. The MSIS data set was established as a 
national eligibility and claims data set and can provide a summary of 
expenditures linked to specific enrollees on the basis of their 
medical claims for care, while the CMS-64 data set aggregates states' 
expenditures and is used to reimburse the states for their Medicaid 
expenditures. We have previously reported that Medicaid expenditures 
as captured in the MSIS were less than the CMS-64 amounts and that 
MSIS data were considered less reliable when compared with the 
CMS-64.[Footnote 27] However, MSIS data are linked to specific 
enrollees, unlike the CMS-64 data, and therefore can be used on their 
own to measure per enrollee spending. For the purpose of providing an 
illustrative example for this report, we therefore relied on MSIS data 
to develop our estimates of Medicaid spending per enrollee and 
Medicaid-service-need weights. It is also possible to combine 
information from MSIS and the CMS-64 in order to develop more 
comprehensive estimates of per-enrollee Medicaid spending. Therefore, 
our per-enrollee estimates--based on MSIS data alone--differ from 
other published estimates that are based on a combination of CMS-64 
and MSIS data, such as certain estimates published by the CMS Office 
of the Actuary and MACPAC.[Footnote 28] We compared our calculated 
service-need weights with those that would be obtained if we had used 
per-enrollee spending estimates published by MACPAC and the CMS Office 
of the Actuary and found them to be similar for comparable categories. 
We used our own calculations because of our interest in dividing 
enrollees who were older than age 64 into two age groups. 

Specifically, we used fiscal year 2009 MSIS data, the latest year for 
which complete data were available for every state, to estimate 
Medicaid spending per enrollee for each category nationwide, measuring 
enrollees in terms of full-year equivalents rather than the total 
number of persons who were enrolled for any length of time during the 
year. We then compared Medicaid spending per enrollee for each 
category with average Medicaid spending per enrollee for all 
categories. For example, in table 2 the service-need-weight for the 
disabled category indicates that, on average, Medicaid spending per 
enrollee among disabled enrollees ages 0 through 64 is 2.6 times 
higher than the average for all enrollees. 

Table 2: Estimates of Medicaid-Service-Need Weights by Age and 
Disability Category: 

Category: Children ages 0 through 18; 
Average spending per enrollee: $2,526; 
Medicaid-service-need weight[A]: 0.4. 

Category: Adults ages 19 through 64; 
Average spending per enrollee: $3,692; 
Medicaid-service-need weight[A]: 0.6. 

Category: Individuals with disabilities ages 0 through 64; 
Average spending per enrollee: $16,027; 
Medicaid-service-need weight[A]: 2.6. 

Category: Elderly individuals ages 65 through 84; 
Average spending per enrollee: $10,472; 
Medicaid-service-need weight[A]: 1.7. 

Category: Elderly individuals ages 85 and older; 
Average spending per enrollee: $24,918; 
Medicaid-service-need weight[A]: 4.0. 

Category: All categories; 
Average spending per enrollee: $6,267; 
Medicaid-service-need weight[A]: 1.0. 

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) 
data. 

Notes: Categories are mutually exclusive. Data are fiscal year 2009 
data from the Medicaid Statistical Information System (MSIS). 

[A] Calculated as average spending per enrollee by category divided by 
average spending per enrollee for all categories. 

[End of table] 

We next examined two possible methods for calculating the demand for 
services measure, by adjusting the state populations in poverty on the 
basis of the Medicaid-service-need weights shown in table 2. For the 
first method, we adjusted the state population in poverty by applying 
the service-need weights according to the composition of the state 
Medicaid population--specifically, according to the extent to which 
each age and disability category is represented within a state's 
Medicaid population. To do so, we used fiscal year 2009 MSIS data to 
calculate the percentage of Medicaid enrollees in each category by 
state. We then multiplied the state poverty population by the 
percentage of Medicaid enrollees in each category in that state and by 
the national service-need weight for that category, and summed the 
results.[Footnote 29] 

This first method accounts for potential differences among categories 
in the participation rates among eligible individuals within those 
categories, as well as potential differences in the proportions of 
children, adults, disabled individuals, and elderly individuals within 
each state's population in poverty. It also reflects state-specific 
coverage decisions about eligibility criteria for each category, in 
that--all else being equal--states with stricter eligibility criteria 
for a given age or disability category relative to other states would 
have a smaller proportion of enrollees in that category, while the 
reverse would be true for states with more generous eligibility 
criteria relative to other states. 

Table 3 shows the results of calculating demand for services measures 
by this first method, which weights estimates of the state population 
in poverty for service needs on the basis of the composition of the 
Medicaid population. We applied the same service-need weights to (1) 
the unadjusted population in poverty and (2) the cost-of-living 
adjusted population in poverty, and then calculated the percentage 
difference compared with the unweighted population in poverty 
estimates shown in table 1. Table 3 illustrates, for example, that in 
West Virginia--a state with a higher than average proportion of 
disabled Medicaid enrollees--the demand for services measure increases 
by about 27 percent after using this method to account for the service 
needs of the population in poverty. The percentage differences shown 
in columns three and five of table 3, between weighted and unweighted 
measures, are the same for the population in poverty and the cost-of-
living-adjusted population in poverty because the same weights--
reflecting the distribution of each state's Medicaid population among 
age and disability categories and the relative cost of serving each 
category--are applied to each population measure. 

Table 3: Estimates of Demand for Services, Including Adjustment for 
Service Need Based on the Composition of the Medicaid Population, by 
State: 

State: Alabama; 
Population in poverty: 
Weighted by service need[A]: 1,047,183; 
Percentage difference compared with unweighted population in 
poverty[B]: 17.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 925,237; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 17.2%. 

State: Alaska; 
Population in poverty: 
Weighted by service need[A]: 65,533; 
Percentage difference compared with unweighted population in 
poverty[B]: -9.5%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 70,809; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -9.5%. 

State: Arizona; 
Population in poverty: 
Weighted by service need[A]: 914,545; 
Percentage difference compared with unweighted population in 
poverty[B]: -19.5%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 904,000; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -19.5%. 

State: Arkansas; 
Population in poverty: 
Weighted by service need[A]: 596,947; 
Percentage difference compared with unweighted population in 
poverty[B]: 8.8%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 515,607; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 8.8%. 

State: California; 
Population in poverty: 
Weighted by service need[A]: 5,271,251; 
Percentage difference compared with unweighted population in 
poverty[B]: -11.1%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 5,972,529; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -11.1%. 

State: Colorado; 
Population in poverty: 
Weighted by service need[A]: 650,065; 
Percentage difference compared with unweighted population in 
poverty[B]: -5.5%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 642,549; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -5.5%. 

State: Connecticut; 
Population in poverty: 
Weighted by service need[A]: 369,451; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 408,502; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -4.2%. 

State: Delaware; 
Population in poverty: 
Weighted by service need[A]: 97,983; 
Percentage difference compared with unweighted population in 
poverty[B]: -9.9%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 102,136; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -9.9%. 

State: District of Columbia; 
Population in poverty: 
Weighted by service need[A]: 122,937; 
Percentage difference compared with unweighted population in 
poverty[B]: 8.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 135,795; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 8.2%. 

State: Florida; 
Population in poverty: 
Weighted by service need[A]: 3,384,924; 
Percentage difference compared with unweighted population in 
poverty[B]: 9.9%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 3,384,924; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 9.9%. 

State: Georgia; 
Population in poverty: 
Weighted by service need[A]: 1,743,384; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.9%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,623,064; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 0.9%. 

State: Hawaii; 
Population in poverty: 
Weighted by service need[A]: 140,330; 
Percentage difference compared with unweighted population in 
poverty[B]: -8.5%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 164,684; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -8.5%. 

State: Idaho; 
Population in poverty: 
Weighted by service need[A]: 246,076; 
Percentage difference compared with unweighted population in 
poverty[B]: 2.5%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 226,055; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 2.5%. 

State: Illinois; 
Population in poverty: 
Weighted by service need[A]: 1,570,507; 
Percentage difference compared with unweighted population in 
poverty[B]: -15.0%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,570,507; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -15.0%. 

State: Indiana; 
Population in poverty: 
Weighted by service need[A]: 924,821; 
Percentage difference compared with unweighted population in 
poverty[B]: -7.1%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 850,745; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -7.1%. 

State: Iowa; 
Population in poverty: 
Weighted by service need[A]: 393,682; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.4%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 343,426; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 0.4%. 

State: Kansas; 
Population in poverty: 
Weighted by service need[A]: 436,818; 
Percentage difference compared with unweighted population in 
poverty[B]: 12.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 382,338; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 12.2%. 

State: Kentucky; 
Population in poverty: 
Weighted by service need[A]: 1,035,117; 
Percentage difference compared with unweighted population in 
poverty[B]: 24.9%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 912,489; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 24.9%. 

State: Louisiana; 
Population in poverty: 
Weighted by service need[A]: 889,593; 
Percentage difference compared with unweighted population in 
poverty[B]: 3.9%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 827,934; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 3.9%. 

State: Maine; 
Population in poverty: 
Weighted by service need[A]: 219,147; 
Percentage difference compared with unweighted population in 
poverty[B]: 19.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 211,481; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 19.2%. 

State: Maryland; 
Population in poverty: 
Weighted by service need[A]: 577,868; 
Percentage difference compared with unweighted population in 
poverty[B]: -0.8%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 640,811; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -0.8%. 

State: Massachusetts; 
Population in poverty: 
Weighted by service need[A]: 834,562; 
Percentage difference compared with unweighted population in 
poverty[B]: 8.3%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 908,187; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 8.3%. 

State: Michigan; 
Population in poverty: 
Weighted by service need[A]: 1,574,151; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.5%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,490,835; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -4.5%. 

State: Minnesota; 
Population in poverty: 
Weighted by service need[A]: 641,583; 
Percentage difference compared with unweighted population in 
poverty[B]: 2.1%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 618,685; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 2.1%. 

State: Mississippi; 
Population in poverty: 
Weighted by service need[A]: 777,059; 
Percentage difference compared with unweighted population in 
poverty[B]: 17.0%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 674,997; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 17.0%. 

State: Missouri; 
Population in poverty: 
Weighted by service need[A]: 982,026; 
Percentage difference compared with unweighted population in 
poverty[B]: 4.6%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 849,411; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 4.6%. 

State: Montana; 
Population in poverty: 
Weighted by service need[A]: 152,432; 
Percentage difference compared with unweighted population in 
poverty[B]: 5.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 141,303; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 5.2%. 

State: Nebraska; 
Population in poverty: 
Weighted by service need[A]: 234,093; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.7%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 204,599; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 0.7%. 

State: Nevada; 
Population in poverty: 
Weighted by service need[A]: 383,647; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.5%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 379,378; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -4.5%. 

State: New Hampshire; 
Population in poverty: 
Weighted by service need[A]: 120,245; 
Percentage difference compared with unweighted population in 
poverty[B]: 2.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 126,343; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 2.2%. 

State: New Jersey; 
Population in poverty: 
Weighted by service need[A]: 1,037,859; 
Percentage difference compared with unweighted population in 
poverty[B]: 12.1%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,175,003; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 12.1%. 

State: New Mexico; 
Population in poverty: 
Weighted by service need[A]: 344,122; 
Percentage difference compared with unweighted population in 
poverty[B]: -17.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 319,887; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -17.2%. 

State: New York; 
Population in poverty: 
Weighted by service need[A]: 2,994,488; 
Percentage difference compared with unweighted population in 
poverty[B]: -0.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 3,456,224; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -0.2%. 

State: North Carolina; 
Population in poverty: 
Weighted by service need[A]: 1,767,180; 
Percentage difference compared with unweighted population in 
poverty[B]: 6.7%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,619,173; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 6.7%. 

State: North Dakota; 
Population in poverty: 
Weighted by service need[A]: 85,362; 
Percentage difference compared with unweighted population in 
poverty[B]: 8.3%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 74,742; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 8.3%. 

State: Ohio; 
Population in poverty: 
Weighted by service need[A]: 1,903,295; 
Percentage difference compared with unweighted population in 
poverty[B]: 1.1%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,717,261; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 1.1%. 

State: Oklahoma; 
Population in poverty: 
Weighted by service need[A]: 625,371; 
Percentage difference compared with unweighted population in 
poverty[B]: -3.7%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 549,313; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -3.7%. 

State: Oregon; 
Population in poverty: 
Weighted by service need[A]: 645,999; 
Percentage difference compared with unweighted population in 
poverty[B]: 5.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 624,110; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 5.2%. 

State: Pennsylvania; 
Population in poverty: 
Weighted by service need[A]: 2,082,818; 
Percentage difference compared with unweighted population in 
poverty[B]: 21.4%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 2,039,906; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 21.4%. 

State: Rhode Island; 
Population in poverty: 
Weighted by service need[A]: 170,425; 
Percentage difference compared with unweighted population in 
poverty[B]: 18.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 168,485; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 18.2%. 

State: South Carolina; 
Population in poverty: 
Weighted by service need[A]: 869,468; 
Percentage difference compared with unweighted population in 
poverty[B]: 2.6%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 794,419; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 2.6%. 

State: South Dakota; 
Population in poverty: 
Weighted by service need[A]: 118,482; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.1%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 102,739; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 0.1%. 

State: Tennessee; 
Population in poverty: 
Weighted by service need[A]: 1,237,668; 
Percentage difference compared with unweighted population in 
poverty[B]: 9.2%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,115,772; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 9.2%. 

State: Texas; 
Population in poverty: 
Weighted by service need[A]: 4,412,370; 
Percentage difference compared with unweighted population in 
poverty[B]: -3.1%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 4,261,290; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -3.1%. 

State: Utah; 
Population in poverty: 
Weighted by service need[A]: 317,608; 
Percentage difference compared with unweighted population in 
poverty[B]: -12.6%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 298,222; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -12.6%. 

State: Vermont; 
Population in poverty: 
Weighted by service need[A]: 70,630; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.3%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 70,259; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 0.3%. 

State: Virginia; 
Population in poverty: 
Weighted by service need[A]: 1,009,078; 
Percentage difference compared with unweighted population in 
poverty[B]: 7.7%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 1,038,683; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 7.7%. 

State: Washington; 
Population in poverty: 
Weighted by service need[A]: 857,040; 
Percentage difference compared with unweighted population in 
poverty[B]: -6.7%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 874,680; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -6.7%. 

State: West Virginia; 
Population in poverty: 
Weighted by service need[A]: 428,467; 
Percentage difference compared with unweighted population in 
poverty[B]: 26.7%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 369,541; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 26.7%. 

State: Wisconsin; 
Population in poverty: 
Weighted by service need[A]: 824,210; 
Percentage difference compared with unweighted population in 
poverty[B]: 7.4%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 750,405; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 7.4%. 

State: Wyoming; 
Population in poverty: 
Weighted by service need[A]: 57,543; 
Percentage difference compared with unweighted population in 
poverty[B]: -10.8%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 52,532; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -10.8%. 

Total: United States; 
Population in poverty: 
Weighted by service need[A]: 47,896,803; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.0%; 
Cost-of-living-adjusted population in poverty: 
Weighted by service need[C]: 47,496,835; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 0.0%. 

Source: GAO analysis of U.S. Census Bureau and Centers for Medicare & 
Medicaid Services data. 

Notes: The data are 2009-2011 data from the American Community Survey 
(ACS) and 2009 data from the Medicaid Statistical Information System 
(MSIS). 

The percentage differences shown in columns three and five, between 
weighted and unweighted measures, are the same for the population in 
poverty and the cost-of-living-adjusted population in poverty because 
the same weights--reflecting the distribution of each state's Medicaid 
population among age and disability categories and the relative cost 
of serving each category--are applied to each population measure. 

[A] Calculated by multiplying the state population in poverty by the 
proportion each age and disability category represents among all state 
Medicaid enrollees and by the category service-need weight, then 
summing the results for each category. 

[B] Calculated in comparison to the unweighted, unadjusted population 
in poverty. 

[C] Calculated by multiplying the cost-of-living adjusted state 
population in poverty by the proportion each age and disability 
category represents among all state Medicaid enrollees and by the 
category service-need weight, then summing the results for each 
category. 

[D] Calculated in comparison to the unweighted, cost-of-living-
adjusted population in poverty. 

[End of table] 

For the second method, we adjusted the state population in poverty by 
applying the service-need weights on the basis of the composition of 
the state poverty population rather than on the basis of the 
composition of the Medicaid population. Specifically, we generated 
poverty population estimates from ACS data separately for each age and 
disability category, and then multiplied these estimates by the 
Medicaid-service-need weights, summing the results.[Footnote 30] 

This second method more directly measures differences in the 
representation of each category within each state's population in 
poverty and is not affected by individual state coverage decisions. 
However, it also does not account for differences in Medicaid 
participation rates among categories, nor does it track federal 
eligibility criteria for the categories. (For example, the adult 
poverty population includes adults who generally have not been 
eligible for Medicaid under federal law, and may or may not become 
eligible in future years depending on state decisions about whether to 
expand Medicaid as provided for under PPACA.)[Footnote 31] 

Table 4 shows the results of calculating a demand for services measure 
by this second method, which--in contrast to the previous method--
weights the state population in poverty for service need based on the 
composition of the population in poverty itself instead of the 
composition of the Medicaid enrollee population. Again, we applied the 
same service-need weights to (1) the unadjusted population in poverty 
and (2) the cost-of-living adjusted population in poverty, and then 
calculated the percentage difference compared with the unweighted 
population in poverty estimates shown in table 1. Due to the different 
weighting method employed, the results of the demand for services 
calculation shown in table 4 generally differ from those shown in 
table 3.[Footnote 32] For example, table 4 shows that the demand for 
services measure for West Virginia increases by about 13 percent when 
the service-need adjustment is made based on the composition of the 
poverty population, as opposed to by about 27 percent when the 
adjustment is made based on the composition of the Medicaid enrollee 
population as shown in table 3. 

Table 4: Estimates of Demand for Services, Including Adjustment for 
Service Need Based on the Composition of the State Poverty Population, 
by State: 

State: Alabama; 
Population in Poverty: 
Weighted by service need[A]: 916,994; 
Percentage difference compared with unweighted population in 
poverty[B]: 2.7%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 809,176; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 2.5%. 

State: Alaska; 
Population in Poverty: 
Weighted by service need[A]: 65,074; 
Percentage difference compared with unweighted population in 
poverty[B]: -10.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 69,654; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -10.9%. 

State: Arizona; 
Population in Poverty: 
Weighted by service need[A]: 941,599; 
Percentage difference compared with unweighted population in 
poverty[B]: -17.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 929,741; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -17.2%. 

State: Arkansas; 
Population in Poverty: 
Weighted by service need[A]: 559,659; 
Percentage difference compared with unweighted population in 
poverty[B]: 2.0%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 480,214; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 1.3%. 

State: California; 
Population in Poverty: 
Weighted by service need[A]: 5,013,297; 
Percentage difference compared with unweighted population in 
poverty[B]: -15.4%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 5,686,186; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -15.4%. 

State: Colorado; 
Population in Poverty: 
Weighted by service need[A]: 610,160; 
Percentage difference compared with unweighted population in 
poverty[B]: -11.3%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 604,119; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -11.2%. 

State: Connecticut; 
Population in Poverty: 
Weighted by service need[A]: 408,524; 
Percentage difference compared with unweighted population in 
poverty[B]: 6.0%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 450,223; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 5.6%. 

State: Delaware; 
Population in Poverty: 
Weighted by service need[A]: 103,156; 
Percentage difference compared with unweighted population in 
poverty[B]: -5.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 107,240; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -5.4%. 

State: District of Columbia; 
Population in Poverty: 
Weighted by service need[A]: 114,624; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.9%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 126,924; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 1.1%. 

State: Florida; 
Population in Poverty: 
Weighted by service need[A]: 2,909,093; 
Percentage difference compared with unweighted population in 
poverty[B]: -5.5%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 2,909,093; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -5.5%. 

State: Georgia; 
Population in Poverty: 
Weighted by service need[A]: 1,540,854; 
Percentage difference compared with unweighted population in 
poverty[B]: -10.8%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,435,464; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -10.7%. 

State: Hawaii; 
Population in Poverty: 
Weighted by service need[A]: 146,227; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.6%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 174,422; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -3.0%. 

State: Idaho; 
Population in Poverty: 
Weighted by service need[A]: 218,125; 
Percentage difference compared with unweighted population in 
poverty[B]: -9.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 199,745; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -9.4%. 

State: Illinois; 
Population in Poverty: 
Weighted by service need[A]: 1,669,156; 
Percentage difference compared with unweighted population in 
poverty[B]: -9.6%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,669,156; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -9.6%. 

State: Indiana; 
Population in Poverty: 
Weighted by service need[A]: 956,705; 
Percentage difference compared with unweighted population in 
poverty[B]: -3.9%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 877,350; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -4.2%. 

State: Iowa; 
Population in Poverty: 
Weighted by service need[A]: 397,264; 
Percentage difference compared with unweighted population in 
poverty[B]: 1.3%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 345,596; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 1.1%. 

State: Kansas; 
Population in Poverty: 
Weighted by service need[A]: 382,114; 
Percentage difference compared with unweighted population in 
poverty[B]: -1.8%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 333,048; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -2.2%. 

State: Kentucky; 
Population in Poverty: 
Weighted by service need[A]: 895,273; 
Percentage difference compared with unweighted population in 
poverty[B]: 8.0%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 779,190; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 6.6%. 

State: Louisiana; 
Population in Poverty: 
Weighted by service need[A]: 834,311; 
Percentage difference compared with unweighted population in 
poverty[B]: -2.5; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 770,853; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -3.2. 

State: Maine; 
Population in Poverty: 
Weighted by service need[A]: 212,808; 
Percentage difference compared with unweighted population in 
poverty[B]: 15.8%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 205,964; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 16.1%. 

State: Maryland; 
Population in Poverty: 
Weighted by service need[A]: 581,446; 
Percentage difference compared with unweighted population in 
poverty[B]: -0.2%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 642,543; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -0.5%. 

State: Massachusetts; 
Population in Poverty: 
Weighted by service need[A]: 856,630; 
Percentage difference compared with unweighted population in 
poverty[B]: 11.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 932,803; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 11.2%. 

State: Michigan; 
Population in Poverty: 
Weighted by service need[A]: 1,611,096; 
Percentage difference compared with unweighted population in 
poverty[B]: -2.3%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,522,285; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -2.5%. 

State: Minnesota; 
Population in Poverty: 
Weighted by service need[A]: 627,103; 
Percentage difference compared with unweighted population in 
poverty[B]: -0.2%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 603,857; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -0.4%. 

State: Mississippi; 
Population in Poverty: 
Weighted by service need[A]: 650,244; 
Percentage difference compared with unweighted population in 
poverty[B]: -2.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 562,051; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -2.6%. 

State: Missouri; 
Population in Poverty: 
Weighted by service need[A]: 964,236; 
Percentage difference compared with unweighted population in 
poverty[B]: 2.7%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 834,033; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 2.7%. 

State: Montana; 
Population in Poverty: 
Weighted by service need[A]: 139,985; 
Percentage difference compared with unweighted population in 
poverty[B]: -3.4%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 129,316; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -3.8%. 

State: Nebraska; 
Population in Poverty: 
Weighted by service need[A]: 225,390; 
Percentage difference compared with unweighted population in 
poverty[B]: -3.0%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 198,046; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -2.5%. 

State: Nevada; 
Population in Poverty: 
Weighted by service need[A]: 340,050; 
Percentage difference compared with unweighted population in 
poverty[B]: -15.4%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 336,691; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -15.2%. 

State: New Hampshire; 
Population in Poverty: 
Weighted by service need[A]: 126,953; 
Percentage difference compared with unweighted population in 
poverty[B]: 7.9%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 134,087; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 8.5%. 

State: New Jersey; 
Population in Poverty: 
Weighted by service need[A]: 901,993; 
Percentage difference compared with unweighted population in 
poverty[B]: -2.5%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,018,230; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -2.8%. 

State: New Mexico; 
Population in Poverty: 
Weighted by service need[A]: 365,621; 
Percentage difference compared with unweighted population in 
poverty[B]: -12.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 338,296; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -12.5%. 

State: New York; 
Population in Poverty: 
Weighted by service need[A]: 2,928,975; 
Percentage difference compared with unweighted population in 
poverty[B]: -2.4%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 3,384,800; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -2.3%. 

State: North Carolina; 
Population in Poverty: 
Weighted by service need[A]: 1,574,434; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.9%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,432,154; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -5.6%. 

State: North Dakota; 
Population in Poverty: 
Weighted by service need[A]: 82,997; 
Percentage difference compared with unweighted population in 
poverty[B]: 5.3%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 71,554; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 3.7%. 

State: Ohio; 
Population in Poverty: 
Weighted by service need[A]: 1,914,640; 
Percentage difference compared with unweighted population in 
poverty[B]: 1.7%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,720,652; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 1.3%. 

State: Oklahoma; 
Population in Poverty: 
Weighted by service need[A]: 652,381; 
Percentage difference compared with unweighted population in 
poverty[B]: 0.4%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 568,368; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -0.4%. 

State: Oregon; 
Population in Poverty: 
Weighted by service need[A]: 584,968; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.7%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 566,970; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -4.4%. 

State: Pennsylvania; 
Population in Poverty: 
Weighted by service need[A]: 1,841,165; 
Percentage difference compared with unweighted population in 
poverty[B]: 7.3%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,803,930; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 7.4%. 

State: Rhode Island; 
Population in Poverty: 
Weighted by service need[A]: 156,997; 
Percentage difference compared with unweighted population in 
poverty[B]: 8.9%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 154,398; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 8.3%. 

State: South Carolina; 
Population in Poverty: 
Weighted by service need[A]: 807,345; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.8%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 734,110; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -5.2%. 

State: South Dakota; 
Population in Poverty: 
Weighted by service need[A]: 118,266; 
Percentage difference compared with unweighted population in 
poverty[B]: -0.1%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 101,276; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -1.3%. 

State: Tennessee; 
Population in Poverty: 
Weighted by service need[A]: 1,148,919; 
Percentage difference compared with unweighted population in 
poverty[B]: 1.4%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 1,025,634; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 0.4%. 

State: Texas; 
Population in Poverty: 
Weighted by service need[A]: 3,915,084; 
Percentage difference compared with unweighted population in 
poverty[B]: -14.0%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 3,780,997; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -14.0%. 

State: Utah; 
Population in Poverty: 
Weighted by service need[A]: 289,619; 
Percentage difference compared with unweighted population in 
poverty[B]: -20.3%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 271,339; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -20.5%. 

State: Vermont; 
Population in Poverty: 
Weighted by service need[A]: 80,753; 
Percentage difference compared with unweighted population in 
poverty[B]: 14.7%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 80,165; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 14.5%. 

State: Virginia; 
Population in Poverty: 
Weighted by service need[A]: 918,232; 
Percentage difference compared with unweighted population in 
poverty[B]: -2.0%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 949,145; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -1.6%. 

State: Washington; 
Population in Poverty: 
Weighted by service need[A]: 881,651; 
Percentage difference compared with unweighted population in 
poverty[B]: -4.0%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 900,148; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -4.0%. 

State: West Virginia; 
Population in Poverty: 
Weighted by service need[A]: 383,995; 
Percentage difference compared with unweighted population in 
poverty[B]: 13.5%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 329,769; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: 13.0%. 

State: Wisconsin; 
Population in Poverty: 
Weighted by service need[A]: 756,778; 
Percentage difference compared with unweighted population in 
poverty[B]: -1.4%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 683,516; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -2.2%. 

State: Wyoming; 
Population in Poverty: 
Weighted by service need[A]: 62,976; 
Percentage difference compared with unweighted population in 
poverty[B]: -2.3%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 57,919; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -1.6%. 

Total: United States; 
Population in Poverty: 
Weighted by service need[A]: 45,385,937; 
Percentage difference compared with unweighted population in 
poverty[B]: -5.2%; 
Cost-of-living-adjusted population in Poverty: 
Weighted by service need[C]: 44,832,438; 
Percentage difference compared with unweighted, cost-of-living-
adjusted population in poverty[D]: -5.6%. 

Source: GAO analysis of U.S. Census Bureau and Centers for Medicare & 
Medicaid Services data. 

Notes: The data are 2009-2011 data from the American Community Survey 
(ACS) and 2009 data from the Medicaid Statistical Information System 
(MSIS). 

The percentage differences shown in columns three and five, between 
weighted and unweighted measures, differ for the population in poverty 
and the cost-of-living-adjusted population in poverty because the 
application of the cost-of-living adjustment affects not only the size 
of the population in poverty, but the distribution of that population 
across the age and disability categories used in weighting. 

[A] Calculated by multiplying the age and disability-category-specific 
state population in poverty by the service-need weight for the 
category, then summing the results for each category. 

[B] Calculated in comparison to the unweighted, unadjusted population 
in poverty. 

[C] Calculated by multiplying the age and disability-category-
specific, cost-of-living adjusted state population in poverty by the 
service-need weight for the category, then summing the results for 
each category. 

[D] Calculated in comparison to the unweighted, cost-of-living-
adjusted population in poverty. 

[End of table] 

Geographic Cost Differences: 

To illustrate possible measures of geographic differences in the cost 
of providing health care services, we used data from the Bureau of 
Labor Statistics' Occupational Employment Statistics (OES) survey to 
estimate personnel costs, which represent the greatest proportion of 
total health care costs. We then constructed two indexes of geographic 
cost differences, each of which gives a different weight to cost of 
personnel relative to the cost of other inputs, which include health 
care facilities, equipment, and supplies. 

To measure geographic differences in the cost of the personnel who 
provide health care services, we used data from the 2009-2011 OES to 
estimate average wages, by state, for health care workers. We used OES 
data instead of CMS's Inpatient Prospective Payment System (IPSS) 
hospital cost report data in part because of concerns about the 
latter. Specifically, the Medicare Program Advisory Commission 
(MedPAC) has noted that the IPPS hospital cost report data reflect at 
least to some extent hospitals' choices regarding wages and staff mix 
and proposed an alternative wage index based on OES data to adjust 
Medicare payments to hospitals.[Footnote 33] We also chose to use the 
OES data because it covers personnel in a wide variety of health care 
settings in addition to hospitals, including offices, clinics, nursing 
homes, and medical and dental laboratories, as well as health care 
providers who work for home health agencies. 

To construct a wage index, we used OES wage data for 22 broadly 
defined occupations within the health care and social assistance 
industry sector.[Footnote 34] For each state, we calculated a weighted 
average wage for the sector, on the basis of the number employed in 
each occupation nationwide, and then constructed an index by dividing 
each state's weighted average wage by the national weighted average 
wage for the sector. 

Data are not readily available on geographic differences in the cost 
of other inputs to health care services--medical equipment and 
supplies and the facilities through which health care services are 
delivered. However, because medical equipment and supplies are 
generally purchased in national markets, we assumed that the costs of 
these items did not vary across states.[Footnote 35] Similarly, we 
assumed that facility costs did not vary across states, as CMS does in 
calculating Medicare prospective payment rates for many services other 
than acute inpatient hospital or physician services. 

Accordingly, to calculate illustrative indexes of geographic 
differences in the cost of health care services by state, we weighted 
our OES wage index on the basis of the proportion of total cost of 
health care services attributed to personnel. Because estimates of the 
proportion of total health care costs that is attributable to 
personnel vary depending on the type of service, we calculated two 
separate indexes of geographic differences in health care costs--one 
in which we assumed personnel costs accounted for 50 percent of total 
costs and another in which we assumed personnel costs accounted for 77 
percent of the total. These percentages are the minimum and maximum 
proportions of health care costs attributed to personnel costs in 
Medicare's prospective payment systems for a variety of 
services.[Footnote 36] All other costs were assumed to account for 50 
and 23 percent of the total, respectively. So, for example, when we 
assumed that personnel costs accounted for 77 percent of total costs, 
if a state had personnel costs 10 percent higher than the national 
average, and thus an index of 1.10, the state's index for geographic 
cost of services was calculated as follows: (0.77) * (1.10) + (0.23) * 
(1.0) = 1.033. See table 5. 

Table 5: Estimates of Cost Indexes for Health Care Services, by State: 

State: Alabama; 
Cost index with 50 percent weight for personnel costs[A]: 0.95; 
Cost index with 77 percent weight for personnel costs[B]: 0.92. 

State: Alaska; 
Cost index with 50 percent weight for personnel costs[A]: 1.09; 
Cost index with 77 percent weight for personnel costs[B]: 1.14. 

State: Arizona; 
Cost index with 50 percent weight for personnel costs[A]: 1.01; 
Cost index with 77 percent weight for personnel costs[B]: 1.02. 

State: Arkansas; 
Cost index with 50 percent weight for personnel costs[A]: 0.93; 
Cost index with 77 percent weight for personnel costs[B]: 0.89. 

State: California; 
Cost index with 50 percent weight for personnel costs[A]: 1.10; 
Cost index with 77 percent weight for personnel costs[B]: 1.15. 

State: Colorado; 
Cost index with 50 percent weight for personnel costs[A]: 1.03; 
Cost index with 77 percent weight for personnel costs[B]: 1.05. 

State: Connecticut; 
Cost index with 50 percent weight for personnel costs[A]: 1.08; 
Cost index with 77 percent weight for personnel costs[B]: 1.12. 

State: Delaware; 
Cost index with 50 percent weight for personnel costs[A]: 1.05; 
Cost index with 77 percent weight for personnel costs[B]: 1.08. 

State: District of Columbia; 
Cost index with 50 percent weight for personnel costs[A]: 0.95: 1.09: 
1.01: 0.93: 1.10: 1.03: 1.08: 1.05: 1.10; 
Cost index with 77 percent weight for personnel costs[B]: 1.16. 

State: Florida; 
Cost index with 50 percent weight for personnel costs[A]: 0.99; 
Cost index with 77 percent weight for personnel costs[B]: 0.98. 

State: Georgia; 
Cost index with 50 percent weight for personnel costs[A]: 0.98; 
Cost index with 77 percent weight for personnel costs[B]: 0.97. 

State: Hawaii; 
Cost index with 50 percent weight for personnel costs[A]: 1.07; 
Cost index with 77 percent weight for personnel costs[B]: 1.10. 

State: Idaho; 
Cost index with 50 percent weight for personnel costs[A]: 0.96; 
Cost index with 77 percent weight for personnel costs[B]: 0.94. 

State: Illinois; 
Cost index with 50 percent weight for personnel costs[A]: 1.02; 
Cost index with 77 percent weight for personnel costs[B]: 1.02. 

State: Indiana; 
Cost index with 50 percent weight for personnel costs[A]: 0.97; 
Cost index with 77 percent weight for personnel costs[B]: 0.95. 

State: Iowa; 
Cost index with 50 percent weight for personnel costs[A]: 0.95; 
Cost index with 77 percent weight for personnel costs[B]: 0.92. 

State: Kansas; 
Cost index with 50 percent weight for personnel costs[A]: 0.96; 
Cost index with 77 percent weight for personnel costs[B]: 0.94. 

State: Kentucky; 
Cost index with 50 percent weight for personnel costs[A]: 0.95; 
Cost index with 77 percent weight for personnel costs[B]: 0.93. 

State: Louisiana; 
Cost index with 50 percent weight for personnel costs[A]: 0.94; 
Cost index with 77 percent weight for personnel costs[B]: 0.91. 

State: Maine; 
Cost index with 50 percent weight for personnel costs[A]: 0.99; 
Cost index with 77 percent weight for personnel costs[B]: 0.99. 

State: Maryland; 
Cost index with 50 percent weight for personnel costs[A]: 1.07; 
Cost index with 77 percent weight for personnel costs[B]: 1.11. 

State: Massachusetts; 
Cost index with 50 percent weight for personnel costs[A]: 1.09; 
Cost index with 77 percent weight for personnel costs[B]: 1.13. 

State: Michigan; 
Cost index with 50 percent weight for personnel costs[A]: 1.00; 
Cost index with 77 percent weight for personnel costs[B]: 1.00. 

State: Minnesota; 
Cost index with 50 percent weight for personnel costs[A]: 1.03; 
Cost index with 77 percent weight for personnel costs[B]: 1.05. 

State: Mississippi; 
Cost index with 50 percent weight for personnel costs[A]: 0.92; 
Cost index with 77 percent weight for personnel costs[B]: 0.87. 

State: Missouri; 
Cost index with 50 percent weight for personnel costs[A]: 0.95; 
Cost index with 77 percent weight for personnel costs[B]: 0.93. 

State: Montana; 
Cost index with 50 percent weight for personnel costs[A]: 0.94; 
Cost index with 77 percent weight for personnel costs[B]: 0.91. 

State: Nebraska; 
Cost index with 50 percent weight for personnel costs[A]: 0.96; 
Cost index with 77 percent weight for personnel costs[B]: 0.93. 

State: Nevada; 
Cost index with 50 percent weight for personnel costs[A]: 1.06; 
Cost index with 77 percent weight for personnel costs[B]: 1.09. 

State: New Hampshire; 
Cost index with 50 percent weight for personnel costs[A]: 1.02; 
Cost index with 77 percent weight for personnel costs[B]: 1.03. 

State: New Jersey; 
Cost index with 50 percent weight for personnel costs[A]: 1.08; 
Cost index with 77 percent weight for personnel costs[B]: 1.13. 

State: New Mexico; 
Cost index with 50 percent weight for personnel costs[A]: 0.98; 
Cost index with 77 percent weight for personnel costs[B]: 0.98. 

State: New York; 
Cost index with 50 percent weight for personnel costs[A]: 1.07; 
Cost index with 77 percent weight for personnel costs[B]: 1.11. 

State: North Carolina; 
Cost index with 50 percent weight for personnel costs[A]: 0.98; 
Cost index with 77 percent weight for personnel costs[B]: 0.97. 

State: North Dakota; 
Cost index with 50 percent weight for personnel costs[A]: 0.94; 
Cost index with 77 percent weight for personnel costs[B]: 0.91. 

State: Ohio; 
Cost index with 50 percent weight for personnel costs[A]: 0.99; 
Cost index with 77 percent weight for personnel costs[B]: 0.98. 

State: Oklahoma; 
Cost index with 50 percent weight for personnel costs[A]: 0.93; 
Cost index with 77 percent weight for personnel costs[B]: 0.89. 

State: Oregon; 
Cost index with 50 percent weight for personnel costs[A]: 1.05; 
Cost index with 77 percent weight for personnel costs[B]: 1.07. 

State: Pennsylvania; 
Cost index with 50 percent weight for personnel costs[A]: 1.00; 
Cost index with 77 percent weight for personnel costs[B]: 0.99. 

State: Rhode Island; 
Cost index with 50 percent weight for personnel costs[A]: 1.05; 
Cost index with 77 percent weight for personnel costs[B]: 1.07. 

State: South Carolina; 
Cost index with 50 percent weight for personnel costs[A]: 0.96; 
Cost index with 77 percent weight for personnel costs[B]: 0.94. 

State: South Dakota; 
Cost index with 50 percent weight for personnel costs[A]: 0.93; 
Cost index with 77 percent weight for personnel costs[B]: 0.89. 

State: Tennessee; 
Cost index with 50 percent weight for personnel costs[A]: 0.95; 
Cost index with 77 percent weight for personnel costs[B]: 0.92. 

State: Texas; 
Cost index with 50 percent weight for personnel costs[A]: 0.99; 
Cost index with 77 percent weight for personnel costs[B]: 0.98. 

State: Utah; 
Cost index with 50 percent weight for personnel costs[A]: 0.98; 
Cost index with 77 percent weight for personnel costs[B]: 0.96. 

State: Vermont; 
Cost index with 50 percent weight for personnel costs[A]: 1.01; 
Cost index with 77 percent weight for personnel costs[B]: 1.01. 

State: Virginia; 
Cost index with 50 percent weight for personnel costs[A]: 1.02; 
Cost index with 77 percent weight for personnel costs[B]: 1.02. 

State: Washington; 
Cost index with 50 percent weight for personnel costs[A]: 1.06; 
Cost index with 77 percent weight for personnel costs[B]: 1.09. 

State: West Virginia; 
Cost index with 50 percent weight for personnel costs[A]: 0.92; 
Cost index with 77 percent weight for personnel costs[B]: 0.87. 

State: Wisconsin; 
Cost index with 50 percent weight for personnel costs[A]: 1.00; 
Cost index with 77 percent weight for personnel costs[B]: 1.00. 

State: Wyoming; 
Cost index with 50 percent weight for personnel costs[A]: 0.99; 
Cost index with 77 percent weight for personnel costs[B]: 0.98. 

State: United States; 
Cost index with 50 percent weight for personnel costs[A]: 1.00; 
Cost index with 77 percent weight for personnel costs[B]: 1.00. 

Source: GAO analysis of U.S. Bureau of Labor Statistics data. 

Notes: Data are 2009-2011 data from the Occupational Employment 
Statistics (OES). 

[A] Calculated as 0.5 + 0.5 * (weighted average wage for health care 
industry workers in the state/weighted average wage for health care 
industry workers in the nation). 

[B] Calculated as 0.23 + 0.77 * (weighted average wage for health care 
industry workers in the state/weighted average wage for health care 
industry workers in the nation). 

[End of table] 

State Resources: 

To illustrate two possible measures of state resources, we used data 
on states' Total Taxable Resources (TTR), as reported by the 
Department of the Treasury. TTR is a measure of all income potentially 
subject to taxation that is either produced within a state or received 
by state residents from out-of-state sources. As such, TTR comprises 
the income included in per capita income (PCI), as well as income from 
other sources, such as corporate income and capital gains. For each 
state, TTR is reported by the Department of the Treasury both as a 
total dollar amount and in dollars per capita. The total dollar amount 
reflects state resources, and the per capita amount reflects the state 
resources relative to the size of the state population. We calculated 
average TTR and average per capita TTR for 2008 through 2010, the most 
recent years for which data are available. See table 6. 

Table 6: Average Total Taxable Resources (TTR) and Per Capita TTR, by 
State, 2008 through 2010: 

State: Alabama; 
Average TTR: $188 billion; 
Average per capita TTR: $39,627. 

State: Alaska; 
Average TTR: $49 billion; 
Average per capita TTR: $70,074. 

State: Arizona; 
Average TTR: $277 billion; 
Average per capita TTR: $43,689. 

State: Arkansas; 
Average TTR: $114 billion; 
Average per capita TTR: $39,322. 

State: California; 
Average TTR: $2.018 trillion; 
Average per capita TTR: $54,608. 

State: Colorado; 
Average TTR: $273 billion; 
Average per capita TTR: $55,007. 

State: Connecticut; 
Average TTR: $260 billion; 
Average per capita TTR: $72,903. 

State: Delaware; 
Average TTR: $67 billion; 
Average per capita TTR: $74,980. 

State: District of Columbia; 
Average TTR: $58 billion; 
Average per capita TTR: $97,608. 

State: Florida; 
Average TTR: $882 billion; 
Average per capita TTR: $47,246. 

State: Georgia; 
Average TTR: $431 billion; 
Average per capita TTR: $44,875. 

State: Hawaii; 
Average TTR: $70 billion; 
Average per capita TTR: $52,197. 

State: Idaho; 
Average TTR: $62 billion; 
Average per capita TTR: $39,938. 

State: Illinois; 
Average TTR: $701 billion; 
Average per capita TTR: $54,750. 

State: Indiana; 
Average TTR: $287 billion; 
Average per capita TTR: $44,494. 

State: Iowa; 
Average TTR: $151 billion; 
Average per capita TTR: $49,725. 

State: Kansas; 
Average TTR: $143 billion; 
Average per capita TTR: $50,377. 

State: Kentucky; 
Average TTR: $172 billion; 
Average per capita TTR: $39,765. 

State: Louisiana; 
Average TTR: $230 billion; 
Average per capita TTR: $51,262. 

State: Maine; 
Average TTR: $56 billion; 
Average per capita TTR: $42,355. 

State: Maryland; 
Average TTR: $351 billion; 
Average per capita TTR: $61,263. 

State: Massachusetts; 
Average TTR: $406 billion; 
Average per capita TTR: $62,377. 

State: Michigan; 
Average TTR: $397 billion; 
Average per capita TTR: $40,067. 

State: Minnesota; 
Average TTR: $286 billion; 
Average per capita TTR: $54,222. 

State: Mississippi; 
Average TTR: $107 billion; 
Average per capita TTR: $36,233. 

State: Missouri; 
Average TTR: $270 billion; 
Average per capita TTR: $45,270. 

State: Montana; 
Average TTR: $41 billion; 
Average per capita TTR: $41,571. 

State: Nebraska; 
Average TTR: $96 billion; 
Average per capita TTR: $52,886. 

State: Nevada; 
Average TTR: $143 billion; 
Average per capita TTR: $53,231. 

State: New Hampshire; 
Average TTR: $73 billion; 
Average per capita TTR: $55,339. 

State: New Jersey; 
Average TTR: $573 billion; 
Average per capita TTR: $65,453. 

State: New Mexico; 
Average TTR: $83 billion; 
Average per capita TTR: $40,807. 

State: New York; 
Average TTR: $1.215 trillion; 
Average per capita TTR: $62,936. 

State: North Carolina; 
Average TTR: $442 billion; 
Average per capita TTR: $46,841. 

State: North Dakota; 
Average TTR: $36 billion; 
Average per capita TTR: $54,299. 

State: Ohio; 
Average TTR: $505 billion; 
Average per capita TTR: $43,782. 

State: Oklahoma; 
Average TTR: $165 billion; 
Average per capita TTR: $44,300. 

State: Oregon; 
Average TTR: $195 billion; 
Average per capita TTR: $51,124. 

State: Pennsylvania; 
Average TTR: $617 billion; 
Average per capita TTR: $48,685. 

State: Rhode Island; 
Average TTR: $56 billion; 
Average per capita TTR: $53,162. 

State: South Carolina; 
Average TTR: $178 billion; 
Average per capita TTR: $38,806. 

State: South Dakota; 
Average TTR: $43 billion; 
Average per capita TTR: $53,033. 

State: Tennessee; 
Average TTR: $271 billion; 
Average per capita TTR: $43,070. 

State: Texas; 
Average TTR: $1.275 trillion; 
Average per capita TTR: $51,455. 

State: Utah; 
Average TTR: $123 billion; 
Average per capita TTR: $45,403. 

State: Vermont; 
Average TTR: $28 billion; 
Average per capita TTR: $45,564. 

State: Virginia; 
Average TTR: $465 billion; 
Average per capita TTR: $58,651. 

State: Washington; 
Average TTR: $370 billion; 
Average per capita TTR: $55,547. 

State: West Virginia; 
Average TTR: $69 billion; 
Average per capita TTR: $37,437. 

State: Wisconsin; 
Average TTR: $266 billion; 
Average per capita TTR: $46,857. 

State: Wyoming; 
Average TTR: $41 billion; 
Average per capita TTR: $72,917. 

State: United States; 
Average TTR: $15.675 trillion; 
Average per capita TTR: $51,108. 

Source: GAO analysis of Department of the Treasury data. 

[End of table] 

One possible alternative to TTR as a measure of state resources would 
be a measure based on the concept of the representative tax system, 
which refers to a hypothetical tax system that is representative of 
the taxes actually levied by all states in the aggregate. Such a 
measure would estimate the resources that would be available to each 
state if their tax policies were in line with those of most other 
states.[Footnote 37] 

[End of section] 

Appendix II: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Carolyn L. Yocom, (202) 512-7114 or yocomc@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Robert Copeland, Assistant 
Director; Emily Beller; Kye Briesath; Greg Dybalski; Nancy Fasciano; 
Kaycee M. Glavich; Drew S. Long; and Hemi Tewarson made key 
contributions to this report. 

[End of section] 

Related GAO Products: 

Older Americans Act: Options to Better Target Need and Improve Equity. 
[hyperlink, http://www.gao.gov/products/GAO-13-74]. Washington, D.C.: 
November 30, 2012. 

Medicaid: Data Sets Provide Inconsistent Picture of Expenditures. 
[hyperlink, http://www.gao.gov/products/GAO-13-47]. Washington D.C.: 
October 29, 2012. 

Medicaid: Prototype Formula Would Provide Automatic, Targeted 
Assistance to States during Economic Downturns. [hyperlink, 
http://www.gao.gov/products/GAO-12-38]. Washington, D.C.: November 10, 
2011. 

State and Local Governments: Knowledge of Past Recessions Can Inform 
Future Federal Fiscal Assistance. [hyperlink, 
http://www.gao.gov/products/GAO-11-401]. Washington, D.C.: March 31, 
2011. 

Medicaid: Improving Responsiveness of Federal Assistance to States 
during Economic Downturns. [hyperlink, 
http://www.gao.gov/products/GAO-11-395]. Washington, D.C.: March 31, 
2011. 

Formula Grants: Funding for the Largest Federal Assistance Programs Is 
Based on Census-Related Data and Other Factors. [hyperlink, 
http://www.gao.gov/products/GAO-10-263]. Washington, D.C.: December 
15, 2009. 

Medicaid: Strategies to Help States Address Increased Expenditures 
during Economic Downturns. [hyperlink, 
http://www.gao.gov/products/GAO-07-97]. Washington, D.C.: October 18, 
2006. 

Federal Assistance: Illustrative Simulations of Using Statistical 
Population Estimates for Reallocating Certain Federal Funding. 
[hyperlink, http://www.gao.gov/products/GAO-06-567]. Washington, D.C.: 
June 22, 2006. 

Medicaid Formula: Differences in Funding Ability among States Often 
Are Widened. [hyperlink, http://www.gao.gov/products/GAO-03-620]. 
Washington, D.C.: July 10, 2003. 

Medicaid Formula: Effects of Proposed Formula on Federal Shares of 
State Spending. [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-99-29R]. Washington, D.C.: 
February 19, 1999. 

Federal Grants: Design Improvements Could Help Federal Resources Go 
Further. [hyperlink, http://www.gao.gov/products/GAO/AIMD-97-7]. 
Washington, D.C.: December 18, 1996. 

Block Grants: Issues in Designing Accountability Provisions. 
[hyperlink, http://www.gao.gov/products/GAO/AIMD-95-226]. Washington, 
D.C.: September 1, 1995. 

Medicaid: Matching Formula's Performance and Potential Modifications. 
[hyperlink, http://www.gao.gov/products/GAO/T-HEHS-95-226]. 
Washington, D.C.: July 27, 1995. 

Block Grants: Characteristics, Experience, and Lessons Learned. 
[hyperlink, http://www.gao.gov/products/GAO/HEHS-95-74]. Washington, 
D.C.: February 9, 1995. 

Block Grants: Lessons Learned. [hyperlink, 
http://www.gao.gov/products/GAO/T-HEHS-95-80]. Washington, D.C.: 
February 9, 1995. 

Maternal and Child Health: Block Grant Funds Should Be Distributed 
More Equitably. [hyperlink, http://www.gao.gov/products/GAO/HRD-92-5]. 
Washington, D.C.: April 2, 1992. 

Changing Medicaid Formula Can Improve Distribution of Funds to States. 
[hyperlink, http://www.gao.gov/products/GAO/GGD-83-27]. Washington, 
D.C.: March 9, 1983. 

[End of section] 

Footnotes: 

[1] The FMAP is calculated using the following formula: FMAP = 1.00 - 
0.45 (State PCI / U.S. PCI)2. PCI is calculated by the U.S. Bureau of 
Economic Analysis. Federal law specifies that the FMAP will be no 
lower than 50 percent and no higher than 83 percent. The Department of 
Health and Human Services is required to publish FMAPs for states 
between October 1 and November 30of each fiscal year. See 42 U.S.C. §§ 
1396b(a)(1), 1396d(b). For fiscal year 2013, states' FMAPs range from 
50.00 percent to 73.43 percent. 

[2] Consistent with this intent, squaring PCI has the effect of making 
PCI appear in the formula twice as an attempt to reflect both state 
resources and the population in need of Medicaid services. 

[3] See, for example, GAO, Medicaid Formula: Differences in Funding 
Ability among States Often Are Widened, [hyperlink, 
http://www.gao.gov/products/GAO-03-620] (Washington, D.C.: July 10, 
2003). 

[4] This is of particular concern for states during economic 
downturns. See, for example, GAO, State and Local Governments: 
Knowledge of Past Recessions Can Inform Future Federal Fiscal 
Assistance, [hyperlink, http://www.gao.gov/products/GAO-11-401] 
(Washington, D.C.: Mar. 31, 2011); and Medicaid: Strategies to Help 
States Address Increased Expenditures during Economic Downturns, 
[hyperlink, http://www.gao.gov/products/GAO-07-97] (Washington, D.C.: 
Oct. 18, 2006). 

[5] See, for example, GAO, Older Americans Act: Options to Better 
Target Need and Improve Equity, [hyperlink, 
http://www.gao.gov/products/GAO-13-74] (Washington, D.C.: Nov. 30, 
2012); and the list of related GAO products at the end of this report. 

[6] See, for example, [hyperlink, 
http://www.gao.gov/products/GAO-03-620]. 

[7] In this report, we use the term "state resources" to refer to 
funding available to states for financing the nonfederal share of 
their Medicaid expenditures, which may include funding provided by 
local government sources. 

[8] For example, as we reported in 2003, New York and Maryland had 
similar PCIs, but the percentage of New York's population that was in 
poverty was almost twice Maryland's. See [hyperlink, 
http://www.gao.gov/products/GAO-03-620]. 

[9] See [hyperlink, http://www.gao.gov/products/GAO-03-620]. 

[10] GAO, Medicare: Geographic Areas Used to Adjust Physician Payments 
for Variation in Practice Costs Should Be Revised, [hyperlink, 
http://www.gao.gov/products/GAO-07-466] (Washington, D.C.: June 29, 
2007). 

[11] See [hyperlink, http://www.gao.gov/products/GAO-03-620]. 

[12] Unless funds were to be allocated on a per-enrollee grant basis, 
Medicaid program data on the number of actual Medicaid enrollees would 
generally not be an ideal data source for this measure because 
Medicaid eligibility criteria vary by state and because not all 
eligible individuals enroll in Medicaid. 

[13] For example, states are currently required to cover children up 
to age 6 in families with income at or below 133 percent of the 
federal poverty level (FPL) and children from ages 6 to 19 in families 
with income below 100 percent of the FPL. Under the Patient Protection 
and Affordable Care Act (PPACA), beginning in January 2014, states 
must cover all children ages 6 to 19 up to 133 percent of the FPL and 
states may extend coverage to nonpregnant, nonelderly adults up to 133 
percent of the FPL. 

[14] In the November 16, 2012, update to the Medicare physician fee 
schedule, CMS attributed 72 percent of the cost of physician services 
to personnel costs, 10 percent to office rent, and 17 percent to all 
other costs, including the cost of medical equipment and supplies. 
(Percentages do not sum to 100 due to rounding.) For other services, 
the percentages CMS attributes to personnel costs vary. 

[15] Under the IPPS, hospitals are paid a standardized amount for an 
entire inpatient episode. CMS calculates IPPS payments through a 
series of adjustments applied to national base payment rates covering 
operating and capital expenses. CMS adjusts hospital payments under 
IPPS using an area wage index to account for geographic differences in 
personnel costs across the country. The wage index reflects how 
average hospital wages in each geographic area compare to average 
hospital wages nationally, set at 1.0. 

[16] In technical comments provided on a draft of this report, the 
Department of Health and Human Services noted that an alternative 
allocation methodology could result in significant changes in federal 
funding for individual states and suggested that we consider detailing 
the projected funding effect of such an alternative. We do not provide 
such details because the measures we discuss could be combined in 
various ways in different allocation mechanisms; additionally, it was 
not within the scope of this engagement to develop or test new 
allocation mechanisms, only to consider possible sources of data. As 
previously noted, the data we used to illustrate these measures are 
not the only data sources that could be used. 

[17] The results of this calculation would differ somewhat depending 
on which measures of demand for services or geographic cost 
differences were selected. 

[18] For example, see GAO, Medicaid: Matching Formula's Performance 
and Potential Modifications, [hyperlink, 
http://www.gao.gov/products/GAO/T-HEHS-95-226] (Washington, D.C.: July 
27, 1995). 

[19] For example, because states' economic conditions affect both 
demand for services and state resources, we have suggested that both 
of these should be considered when targeting increased assistance to 
states during economic downturns. Such consideration is important 
because rising unemployment during economic downturns can lead to 
increases in the number of individuals who are eligible for Medicaid 
coverage at the same time that states' tax revenues decline. See, for 
example, GAO, Medicaid: Prototype Formula Would Provide Automatic, 
Targeted Assistance to States during Economic Downturns, [hyperlink, 
http://www.gao.gov/products/GAO-12-38] (Washington, D.C.: Nov. 10, 
2011). 

[20] The data sources we used have a significant time lag and may be 
appropriate when state economic conditions are stable or improving. 
However, funding adjustments to states during economic downturns may 
require more current data sources, such as unemployment rates, to 
reflect increasing demand for services. For example, see GAO, 
Medicaid: Prototype Formula Would Provide Automatic, Targeted 
Assistance to States during Economic Downturns, [hyperlink, 
http://www.gao.gov/products/GAO-12-38] (Washington, D.C.: Nov. 10, 
2011); Medicaid: Improving Responsiveness of Federal Assistance to 
States during Economic Downturns, [hyperlink, 
http://www.gao.gov/products/GAO-11-395] (Washington, D.C.: Mar. 31, 
2011); Medicaid: Strategies to Help States Address Increased 
Expenditures during Economic Downturns, [hyperlink, 
http://www.gao.gov/products/GAO-07-97] (Washington, D.C.: Oct. 18, 
2006). 

[21] The Census Bureau produces 1-year, 3-year, and 5-year ACS data 
files. The 2009-2011 3-year file is a multiyear combination of 1-year 
files, with appropriate adjustments to weights and other data 
elements. We chose the 3-year file over the 1-and 5-year files in 
order to include only the more recent years of available data while 
also increasing the sample size and thus the precision of our 
estimates. However, both the 1-year and 5-year files would also be 
appropriate options depending on relative preferences in terms of how 
current the data are versus the precision of estimates that can be 
generated from them. 

[22] Regional price parities are indexes that measure price-level 
differences across regions during a given period by comparing the 
average price level of an area to the national average price level for 
all areas. See Bettina H. Aten, Eric B. Figueroa, and Troy M. Martin, 
"Regional Price Parities for States and Metropolitan Areas, 2006-
2010," Survey of Current Business (Bureau of Economic Analysis, August 
2012). Alternative price indexes are available; for example, see Trudi 
Renwick, "Geographic Adjustments of Supplemental Poverty Measure 
Thresholds: Using the American Community Survey Five-Year Data on 
Housing Costs," U.S. Census Bureau, Social, Economic, and Housing 
Statistics Division, Working Paper 2011-21 (July 2011). 

[23] The ACS family income variable is based on all family members 
aged 15 years and older residing within a household. The Patient 
Protection and Affordable Care Act (PPACA) specified that, beginning 
in 2014, for purposes of determining Medicaid eligibility for most 
individuals, income will be defined based on the income of every 
individual in a household, including the income of individuals that a 
household's primary taxpayer expects to claim as tax dependents, 
unless the dependents are not required to file tax returns. 

[24] See MACPAC, Report to the Congress on Medicaid and CHIP 
(Washington, D.C.: June 2012). 

[25] Eligibility for Medicaid is based on a variety of categorical and 
financial requirements. For example, Medicaid eligibility focuses on 
certain categories of individuals, such as low-income children, 
individuals who are aged, and individuals with disabilities. 

[26] See GAO, Medicaid Formula: Differences in Funding Ability among 
States Often Are Widened, [hyperlink, 
http://www.gao.gov/products/GAO-03-620] (Washington, D.C.: July 10, 
2003), in which we used age categories of 0 through 20, 21 through 64, 
and 65 and older. 

[27] See GAO, Medicaid: Data Sets Provide Inconsistent Picture of 
Expenditures, [hyperlink, http://www.gao.gov/products/GAO-13-47] 
(Washington, D.C.: Oct. 29, 2012). 

[28] See Department of Health and Human Services, Centers for Medicare 
& Medicaid Services, Office of the Actuary, 2011 Actuarial Report on 
the Financial Outlook for Medicaid (Washington, D.C.: Mar. 16, 2012) 
and MACPAC, Report to the Congress on Medicaid and CHIP (Washington, 
D.C.: June 15, 2012). 

[29] Under this first method, the final demand for services measure is 
calculated as follows: State population in poverty * percentage of 
state Medicaid enrollees who were ages 0 through 18 * 0.4 + state 
population in poverty * percentage of state Medicaid enrollees who 
were ages 19 through 64 * 0.6 + state population in poverty * 
percentage of state Medicaid enrollees who were disabled (under age 
65) * 2.6 + state population in poverty * percentage of state Medicaid 
enrollees who were ages 65 through 84 * 1.7 + state population in 
poverty * percentage of state Medicaid enrollees who were ages 85 and 
older * 4.0. 

[30] Under this second method, the final demand for services measure 
is calculated as follows: State population in poverty * percentage of 
that population who were children ages 0 through 18 * 0.4 + state 
population in poverty * percentage of that population who were adults 
ages 19 through 64 * 0.6 + state population in poverty * percentage of 
that population who were disabled individuals under age 65 * 2.6 + 
state population in poverty * percentage of that population who were 
elderly individuals ages 65 through 84 * 1.7 + state population in 
poverty * percentage of that population who were elderly individuals 
ages 85 and older * 4.0. 

[31] These considerations could, however, be addressed by additional 
modifications to the measure. For example, the category-specific 
estimates of the population in poverty could be further weighted to 
account for each category's national Medicaid participation rate (that 
is, the national number of Medicaid enrollees in the category divided 
by the number of persons in the national poverty population who are in 
the category). For an example of this modification, see [hyperlink, 
http://www.gao.gov/products/GAO-03-620]. 

[32] The percentage differences shown in columns three and five of 
table 4, between weighted and unweighted measures, differ because the 
application of the cost-of-living adjustment affects not only the size 
of the population in poverty, but the distribution of that population 
across the age and disability categories used in weighting. In 
contrast, the cost-of-living adjustment does not affect the weighting 
used in table 3, which is based on distribution of states' Medicaid 
populations across categories. 

[33] See MedPAC, Report to the Congress: Promoting Greater Efficiency 
in Medicare (Washington, D.C.: June 15, 2007). 

[34] Because nursing home care represents a sizable percentage of 
total Medicaid expenditures--more than 11 percent in 2011, not 
including hospital-based nursing home care--we also constructed an 
alternative wage index using data for occupations within the nursing 
and residential care facilities industry sector. This index did not 
differ substantially from the one we constructed from wage data on 
occupations in the health care and social assistance industry sector, 
so we present only the latter index here. 

[35] Under Medicare's prospective payment systems, providers are paid 
a predetermined, fixed amount, which is derived on the basis of the 
classification system for the type of service provided (for example, 
diagnosis-related groups for inpatient hospital services). CMS uses 
separate prospective payment systems for different types of services, 
including hospital outpatient and home health services. For certain 
services, the payment calculation includes a geographic cost 
adjustment factor for differences in personnel costs, but no 
adjustment for differences in the other costs of providing health care 
services, including facility costs. However, for other services, 
adjustments are made for geographic differences in nonpersonnel costs. 
For example, for physician services, the Medicare payment system 
includes a Geographic Practice Cost Index based on apartment rents as 
captured in the ACS to account for geographic differences in health 
care facility costs. 

[36] These services include outpatient hospital services and home 
health care services, as well as services provided in ambulatory 
surgical centers, skilled nursing facilities, inpatient rehabilitative 
facilities, and psychiatric hospitals, among others. 

[37] For example, see Advisory Commission on Intergovernmental 
Relations, 1988 State Fiscal Capacity and Effort (Washington, D.C.: 
August 1990). 

[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 website [hyperlink, http://www.gao.gov]. Each 
weekday afternoon, GAO posts on its website newly released reports, 
testimony, and correspondence. To have GAO e-mail you a list of newly 
posted products, 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 
website, [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. 

Connect with GAO: 

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

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

Contact: 
Website: [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: 

Katherine Siggerud, Managing Director, siggerudk@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, DC 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, DC 20548. 

[End of document]