STATE OF FLORIDA et al v. UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES et al

Filing 83

NOTICE Errata re Exhibits in support of 82 Defendants' Motion for Summary Judgment by TIMOTHY F GEITHNER, KATHLEEN SEBELIUS, HILDA L SOLIS, UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES, UNITED STATES DEPARTMENT OF LABOR, UNITED STATES DEPARTMENT OF THE TREASURY (Attachments: # 1 Table of Exhibits, # 2 Exhibit 1, # 3 Exhibit 2, # 4 Exhibit 3, # 5 Exhibit 4, # 6 Exhibit 5, # 7 Exhibit 6, # 8 Exhibit 7, # 9 Exhibit 8, # 10 Exhibit 9, # 11 Exhibit 10, # 12 Exhibit 11, # 13 Exhibit 12, # 14 Exhibit 13, # 15 Exhibit 14, # 16 Exhibit 15, # 17 Exhibit 16, # 18 Exhibit 17, # 19 Exhibit 18, # 20 Exhibit 19, # 21 Exhibit 20, # 22 Exhibit 21, # 23 Exhibit 22, # 24 Exhibit 23, # 25 Exhibit 24, # 26 Exhibit 25, # 27 Exhibit 26, # 28 Exhibit 27, # 29 Exhibit 28, # 30 Exhibit 29, # 31 Exhibit 30, # 32 Exhibit 31, # 33 Exhibit 32, # 34 Exhibit 33, # 35 Exhibit 34, # 36 Exhibit 35, # 37 Exhibit 36, # 38 Exhibit 37, # 39 Exhibit 38, # 40 Exhibit 39, # 41 Exhibit 40, # 42 Exhibit 41, # 43 Exhibit 42, # 44 Exhibit 43) (BECKENHAUER, ERIC)

STATE OF FLORIDA et al v. UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES et al

Doc. 83 Att. 35

Exhibit 34

Dockets.Justia.com

k acios e ri s s i o n mm
medicaid
and the

on

uninsured

Medicaid Coverage and Spending in Health Reform:

National and StatebyState Results for Adults at or Below 133% FPL
Prepared by: John Holahan and Irene Headen Urban Institute May 2010

k acios e ri s s i o n mm
and the

medicaid

uninsured
The Kaiser Commission on Medicaid and the Uninsured provides information and analysis on health care coverage and access for the low-income population, with a special focus on Medicaid's role and coverage of the uninsured. Begun in 1991 and based in the K a i s e r F a m i l y F o u n d a t i o n ' s Wa s h i n g t o n , D C office, the Commission is the largest operating program of the Foundation. The Commission's work is conducted by Foundation staff under the guidance of a bipartisan group of national leaders and e x p e r t s i n h e a l t h c a r e a n d p u b l i c p o l i c y.
James R. Ta l l o n

Chairman Diane Rowland, Sc.D.

Executive

Director

Executive Summary
The Patient Protection and Affordable Care Act (PPACA) expands Medicaid to nearly all individuals under age 65 with incomes up to 133 percent of the federal poverty line (FPL) which will extend coverage to large numbers of the nation's uninsured population, especially adults. However, the ultimate reach of the program will depend heavily on both federal and state actions to implement the new law. The Congressional Budget Office (CBO) has provided national estimates of the impacts of health reform, but does not provide statebystate estimates. We know that the impact of health reform will vary across states based on coverage levels in states today. This analysis provides national and statebystate estimates of the increases in coverage and the associated costs compared to a baseline scenario without the Medicaid expansions in health reform. Nationally and across states, this analysis shows that: Medicaid expansions will significantly increase coverage and reduce the number of uninsured The federal government will pay a very high share of new Medicaid costs in all states Increases in state spending are small compared to increases in coverage and federal revenues and relative to what states would have spent if reform had not been enacted

Today there is a great deal of variation across states in terms of Medicaid coverage, the uninsured, state fiscal capacity, leadership and priorities. These variations make it impossible to know how each state individually will respond to the new health reform law. There are a range of implementation scenarios that will impact the number of people who participate or sign up for coverage and these participation rates are directly related to the estimates of coverage and cost for health reform. Since it is impossible to predict the behavior of each state, this analysis examines two participation rate scenarios that are applied uniformly across states; however, we recognize that some states may implement reform to achieve coverage levels above expectations and others may be slower to implement reform or face implementation barriers that result in lower coverage levels. The two modeled scenarios are: 1. Standard Participation Scenario. This scenario attempts to approximate participation rates used by the CBO to estimate the national impact of the Medicaid expansion and then examines the results by state. These results assume moderate levels of participation similar to current experience among those made newly eligible for coverage and little additional participation among those currently eligible. This scenario assumes 57 percent participation among the newly eligible uninsured and lower participation across other coverage groups. 2. Enhanced Outreach Scenario. This scenario examines the impact and reach of Medicaid assuming a more aggressive outreach and enrollment campaign by federal and state governments as well as key stakeholders including community based organizations and providers that would promote more robust participation among those newly eligible (75 percent participation among the newly eligible that are currently uninsured and lower participation across other coverage groups) and higher participation among those currently eligible for coverage than in the standard scenario.

1

1

Even in a scenario with higher participation, we did not assume that there will be full or 100 percent participation. We did not model a participation rate lower than the standard, but this scenario might result in coverage levels that are not a substantial improvement over what would have occurred in the absence of reform (or baseline levels). This analysis estimates the impact of the coverage provisions for adults in health reform between 2014 and 2019 but does not account for other Medicaid changes in the law. For a more detailed description of the methods used in the analysis for this brief and a description of how the changes in the Medicaid match rates are applied to different populations, see the full text of the report and boxes 1 and 2 at the end of the executive summary.

Standard Participation Scenario
This scenario assumes that states will implement health reform and achieve levels of participation similar to current enrollment in Medicaid among those made newly eligible for coverage; however, this scenario assumes little additional participation among those currently eligible. These results attempt to approximate participation rates used by the CBO. National Results Medicaid expansions will significantly increase coverage and reduce the number of uninsured. Medicaid enrollment is projected to increase by 15.9 million by 2019. This new coverage would result in a reduction of uninsured adults under 133 percent of poverty of 11.2 million, a 45 percent reduction in 2019 (Figure 1). States with more limited coverage and higher uninsured rates prereform (like Texas) will see larger decreases in the uninsured compared to states with broader coverage and fewer uninsured prereform (like Massachusetts). The federal government will pay a very high share of new Medicaid costs in all states. In this scenario, federal spending would increase by $443.5 billion and state spending would increase by $21.1 billion between 20142019 (Figure 2). Thus about 95 percent of all new spending would be by the federal government. Spending in 2014 is expected to be relatively small, particularly for states because enrollment is being phasedin and the federal matching rate for new eligibles is 100 percent. Overall and state spending increases by 2019 as coverage is phased in to full implementation levels and federal matching rates for new eligibles fall to 93 percent from 100 percent.
Figur e 1

Figure 2

Stan dar d Scenario: Changes in Coverage from Medicaid Expansion in PPACA in 2019 (in millions)
15. 9

Standard Scenario: Changes in Costs from Medicaid Expansion in PPACA 2014-2019
(in billions)
$21.1 State

$ 443.5 Federal

-11.2 Medic aid Unins ured
Total $464.7

2

2

00

Increases in state spending are small compared to increases in coverage and federal revenues and relative to what states would have spent if reform had not been enacted (baseline). Nationally, enrollment is expected to increase by 27.4 percent compared to baseline. This increase in enrollment far exceeds increases in state spending relative to baseline of 1.4 percent. Due to the large increase in federal matching rates, the federal increases in Medicaid spending compared to Figur e 3 baseline are expected to be 22.1 percent with overall spending increases of 13.2 percent. Stan dar d Scenario: Enrollment and Spending Increases Over Baseline 2014-2019 (Figure 3) The federal matching rates prereform and preARRA average 57.1 percent. The federal matching rate after reform is the combination of 27 .4% current matching rates on current eligibles, 22.1% expansion state match rate for certain childless adults, and the higher federal matching rates on 13. 2% new eligibles. The aggregate match rates for Medicaid or the share of total Medicaid spending 1.4% financed by the federal government is expected Enrollm ent in Stat e Spending F ederal Spending Tot al Spending to increase from 57.1 percent (under current law) 2019 to 61.6 percent; however, states that have had large increases in the number of new eligibles will see the greatest increases in matching rates. StatebyState Results The impact of the Medicaid expansions under health reform will vary across states based on current levels of coverage and current match rates for states. The next section reviews the variation in the impact of costs and coverage across states. For statebystate results of the standard scenario see Table 1. For purposes of this discussion we group the results into the experience in three types of states. For each group we will use the results from two states as illustrative of the experience for other states in that group: States with low Medicaid eligibility for adults today (Alabama and Texas) States that have broader coverage today for parents but have no Medicaid coverage for childless adults (California and New Jersey), and Expansion states that cover both parents and childless adults in Medicaid today (Massachusetts and New York). 1

1

F y s u h o r th i s a n a l s i w e a s s m e t a t th e re are seven "expansion states" which include: Arizona, Delaware, Hawaii, Maine, Massachusetts, New York, and Vermont.

3

3

The Medicaid expansion will result in large reductions in the uninsured across states, but especially in states that have higher levels of uninsured today. Overall, the Medicaid expansion is expected to result in a decrease in the number of uninsured of 11.2 million people, or 45 percent of the uninsured adults below 133 percent of poverty. States with low coverage levels and Figur e 4 higher uninsured rates today will see larger Sta n da rd Scenario: Percent Reduction in Uninsured reductions (Alabama 53.2 percent and Texas Adults < 133% FPL Due to Medicaid Expansion in 2019 49.4). States with broader coverage levels for 5 3 .2 % 49.4% parents today, no coverage for childless adults 45.3% 4 4 .5 % 41.5% and high uninsured rates will also see large reductions in the uninsured (California 41.5 percent and New Jersey 45.3 percent). States 14.8% with lower uninsured rates today will see smaller 10.2% reductions (Massachusetts 10.2 percent reduction and New York 14.8 percent). (Figure 4) Tot al AL TX CA NJ MA NY Broa de r Overall, Texas and California could each see a Low Coverage E xpans ion Coverage for Level s Today States Parents Today reduction in the uninsured of about 1.4 million compared to baseline in 2019. The actual federal share of the costs of the Medicaid expansion varies based on state coverage levels today, but it is always very high. States with low coverage levels today will see the vast majority of the costs of new enrollment financed by the federal government over the 2014 to 2019 period because most of their increased enrollment is from individuals made eligible by health reform who qualify for the high newly eligible match rate (for Alabama, 96 percent and Texas, 95 percent). States with broader coverage of parents today have the majority of costs financed by the federal government, but at slightly lower levels because they experience a higher participation of those currently eligible whose coverage is reimbursed at the states' regular match rates (California, 94 percent and New Jersey 94 percent). For expansion states, the level of federal financing varies with the proportion of current eligibles to newly Figure 5 eligible or those eligible for the expansion match Standard Scenario: Federal Share of Costs of the rate. Massachusetts, a state with no new Medicaid Expansion 2014-2019 eligibles, will actually achieve some savings because the benefit of the expansion match rate 100.0% 99.4% 95.6% 95.3% 95.4% 93.7% 94.4% for current and new coverage of childless adults outweighs any new state costs related to increases in participation for parents at the regular Medicaid match rate. States with state funded coverage programs for adults benefit because these adults will be considered newly Total AL TX CA NJ MA NY eligible for Medicaid and qualify for the newly Broader Low Coverage Expansion eligible match rate. Generally, states will benefit Coverage for Levels Today States Parents Today from a large influx of federal dollars and new coverage is likely to reduce the need for state payments for uncompensated care. (Figure 5)

4

4

00

Compared to projected enrollment without health reform, increases in new enrollment and coverage will far exceed new state costs, but these increases vary based on current levels of coverage across states. States with more modest coverage today are expected to see large increases in enrollment compared to projections without health reform. Increases in enrollment will be lower in states that have already covered a large share of these populations. Increases in enrollment far exceed increases in state spending relative to baseline estimates and this differential is biggest in states with low Figur e 6 coverage today. For example, Texas could see an S tan dard Scenario: Enrollment and State increase in enrollment of 46 percent but an Spending Increases Over Baseline 2014-2019 increase in state spending of about 3 percent. Federal spending in Texas is expected to increase Enrollment in 20 19 Stat e Spending 4 5 .5 % by 39 percent compared to baseline. States with 38.1% 36.9% low coverage today are expected to see large 27.4% increases in federal spending relative to baseline 2 0 .1 % both because of the very favorable matching rate - 2.1% on new eligibles and because these states also 6.0% 3.6% 3 .0 % 2.0% 1.4% 1 .5 % 1 .2 % 0.0% have a high regular Medicaid match rate for Tot al AL TX CA NJ MA NY current eligibles. Increases in coverage and B roa de r Low Coverage E xpans ion spending will be lower in states that have already Coverage for Level s Today States Parents Today covered a large share of these populations. (Figure 6)

Enhanced Outreach Scenario
This scenario examines the impact on Medicaid and the uninsured assuming a more aggressive outreach and enrollment campaign at both the federal and state levels that would promote more robust participation in Medicaid and further reduce the number of uninsured in this lowincome population compared to the standard scenario. The enhanced scenario also assumes that individuals respond favorably to the new mandate for coverage. Even though the large majority of Figur e 7 those eligible for Medicaid will be exempt from Chan ges in Coverage from Medicaid the penalties for failure to comply with the Expansion in PPACA in 2019 mandate, a new culture of coverage along with (in millions) outreach efforts are likely to yield more M edicaid Unins ured 22. 8 participation. These factors would increase 15.9 participation of both those made newly eligible for coverage under health reform and eligible for coverage prior to changes in reform. Under the enhanced outreach scenario applied uniformly across states, Medicaid enrollment could increase by 22.8 million by 2019 resulting in a 17.5 million reduction in uninsured adults under 133 percent of poverty (a 70 percent reduction). (Figure 7)
-11.2 -17.5

St andard Scenario

Enhanced Scenario

5

5

Compared to the standard scenario, states will see larger reductions in the uninsured. Similar to the standard scenario, states with low coverage levels today will see larger reductions (Alabama 73 percent and Texas 74 Figur e 8 percent). States with broader coverage levels for parents but no coverage for childless adults and En hanc ed Scenario: Percent Reduction in Uninsured Adults < 133% FPL Due to Medicaid Expansion in 2019 high uninsured rates will also see large reductions in the uninsured (California 68 percent and New 72.9% 73.6% 70.6% 6 9 .5 % 67.6% Jersey 71 percent). States with lower uninsured rates today will see smaller reductions 46.7% 42.9% (Massachusetts 43 percent reduction and New York 47 percent). (Figure 8) In this scenario, California could see a reduction in the uninsured of 2.3 million and Texas could see a 2.1 million reduction compared to baseline projections in Tot al AL TX CA NJ MA NY Broa de r 2019. See Table 2 for the statebystate results of Low Coverage E xpans ion Coverage for Level s Today States Parents Today the enhanced participation scenario. Under these higher participation assumptions, new spending for Medicaid would continue to be mostly federal (92.5 percent) over the 2014 to 2019 period. State spending could increase by $43 billion while federal spending could increase by $532 billion. The share of spending borne by the federal government will be somewhat lower under the higher participation assumptions, primarily due to higher takeup among those who are eligible under prePPACA rules. Since the states will receive lower federal matching rates for those previously eligible, states will be responsible for a higher share of their Figur e 9 costs. Relative to baseline spending, Medicaid E nh an ced Scenario: Enrollment and State enrollment could increase by 39 percent, Spending Increases Over Baseline 2014-2019 significantly higher than state spending increases Enrollment in 20 19 Stat e Spending of 2.9 percent. Federal spending nationally in 63.5% this scenario could be about 27 percent higher 55.4% 4 7 .9 % than baseline projections. (Figure 9) . In this 3 9 .3 % scenario, the aggregate match rates for Medicaid 2 9 .9 % or the share of total Medicaid spending financed 1 6 .0 % - 1.0% by the federal government is expected to 5 .3 % 5.2% 5.1% 3.4% 2 .9 % 2 .5 % 1 .2 % increase from 57.1 percent (under current law) to Tot al AL TX CA NJ MA NY 62.1 percent; however, states with large B roa de r Low Coverage E xpans ion Coverage for Level s Today States increases in the number of new eligibles will see Parents Today the greatest increases in matching rates.

Limited Outreach Scenario
Right now, states are still in the midst of a major economic downturn facing historic declines in revenues and increased demand for public programs. The impact of the downturn varies across states and the economic recovery will vary across states as well. Heading into health reform, some states will move quickly to promote coverage with efforts that may begin in 2010, while others may move more slowly. Some are challenging and opposing health reform through amendments to their state statutes and constitutions, ballot initiatives and court challenges. Continuing an approach to Medicaid that dates back to its enactment in 1965, health reform revises the standards with which states that choose to participate in the program must comply. Because
6

6

00

Medicaid is voluntary, states may choose to not to participate and thereby forego the federal Medicaid funding to which participating states are entitled. States that elect not to implement these new requirements in effect would be making the choice not to participate. The outcome of state actions will affect the extent to which implementation of health reform reaches its fullest potential. If states fall short of implementation expectations, fewer individuals will be covered and more individuals will remain uninsured. Under this scenario, states would also forgo large sums of federal funding tied to the coverage of those made newly eligible under reform. Even though states would have higher numbers of uninsured in this scenario, they will also face a reduction in the federal dollars to support uncompensated care since the new law calls for reductions in disproportionate share hospital payments (DSH) of $14 billion over the 2014 to 2019 period.

Conclusion
The changes to the Medicaid program under the Patient Protection and Affordability Care Act (PPACA) significantly expand Medicaid coverage for adults. There will be large increases in coverage and federal funding in exchange for a small increase in state spending. States with low coverage levels and high uninsured rates will see the largest increases in coverage and federal funding. Higher levels of coverage will allow states to reduce payments they make to support uncompensated care costs. The impact of health reform will vary across states based on coverage levels in states today, state decisions about implementation and ultimately the number of individuals who sign up for coverage. It is impossible to know how individual states will respond, so this analysis looked at a range of participation assumptions that are applied uniformly across states, but in reality this will vary. Some states may not aggressively implement health reform and therefore not see significant reductions in the uninsured while other states will have higher levels of participation because of effective outreach and enrollment strategies and see greater reductions in the number of uninsured.

7

7

Box 1: Methods Summary
The Model Database. We use the 2007 and 2008 Current Population Survey (CPS) as our baseline data set (which provides data for 2006 and 2007). It is generally accepted that the CPS has an undercount of the Medicaid population. We adjust for the undercount with a partial adjustment to state administrative data. We then generate a 2009 dataset by growing the population to 2009. We account for the impact of unemployment on coverage which has the effect of reducing employer coverage, increasing Medicaid enrollment, and increasing the number of uninsured. We also benchmark to 2009 CPS total population estimates by state and estimate population growth to 2019 using growth rates based on Census population projections. Eligibility Simulation. To estimate the impact of health reform on states, we use a model developed at the Urban Institute's Health Policy Center (Health Insurance Policy Simulation Model or HIPSM). The model takes into account statelevel eligibility requirements for Medicaid and CHIP eligibility pathways and applies them to person and familylevel data from the Annual Social and Economic Supplement to the CPS to simulate the eligibility determination process. The model identifies eligibility for Section 1115 waiver programs which is critical for determining match rates for coverage in seven states: Arizona, Delaware, Hawaii, Maine, Massachusetts, New York, and Vermont. Participation Rates. Once we have identified individuals who are newly eligible for Medicaid, we then assess the likelihood that they will participate in Medicaid under reform. The uninsured are likely to participate at relatively higher rates postreform because they currently lack coverage but not all new participation will come from the ranks of the Baseline Standard Enhanced uninsured. Participation rates are also likely to increase for those who are currently Coverage Scenario Scenario eligible but not participating in Medicaid. Under the standard scenario, we use a set Current Eligibles of participation rates that attempt to approximate those used by CBO (57% ESI 3% 5% participation from the uninsured and lower rates for other coverage groups). The Nongroup 7% 10% actual participation rates assumed in the CBO estimates are not publicly available. Uninsured 10% 40% We also look at the impact of a scenario with aggressive broader outreach and New Eligibles enrollment efforts and stronger response to the individual mandate (even though ESI 25% 25% the Medicaid population is largely exempt from these requirements). In this Nongroup 54% 60% scenario we assume 75% participation of the uninsured and lower rates for other Uninsured 57% 75% coverage groups. Cost per Person. We make estimates on the costs per enrollee using data from HIPSM. These estimates are based on the Medical Expenditure Panel Survey (MEPS) but calibrated to reflect differences in health status of Medicaid eligibles who are currently uninsured, have nongroup coverage, or employersponsored insurance. Estimates from MEPS are adjusted to be consistent with targets from the Medicaid Statistical Information System (MSIS). Cost per enrollee is then grown to 2019 using growth rates taken from the CBO March 2009 baseline. The Baseline. We use estimates of state and federal Medicaid spending in the baseline, i.e. what would have happened without reform if current law continued, to assess the impact of reform. Baseline enrollment and national spending totals for the years 20092019 were calculated using published CBO estimates from March 2009 to grow data from the 2007 Medicaid Statistical Information Statistics (MSIS) and CMS Form64 Medicaid Financial Report (CMS64). Using published Federal Medical Assistance Percentages (FMAP) from the Department of Health and Human Services, we calculated the federal and state share of spending for each state. These 2007 federal spending counts were grown to match 2009 spending from the CBO by enrollment group at the national level. Then these same growth rates were applied to each state. Published 2009 FMAP rates were then used to calculate the state and total spending amounts in 2009. This process was repeated for each year, 2010 through 2019, using CBO estimates and the most recent FMAP rates for each year, without the adjustments made by the American Recovery and Reinvestment Act (ARRA). Other Assumptions. These estimates do not account for: increased participation for states with current Medicaid coverage levels above 133% FPL because after 2014 states are unlikely to continue to cover these individuals on Medicaid; costs associated with the increase in physician payment rates for primary care; the effects of reform for children; or the fiscal implications of the reductions of disproportionate share hospital payments. Finally, the analysis also does not account for any changes in Medicaid between 2010 and 2014. States are permitted to extend coverage to childless adults and receive their regular federal medical assistance percentages (FMAP) until 2014.

8

8

00

Box 2: Medicaid Match Rates for Coverage in Health Reform Summary
The health reform law establishes a new, minimum standard for Medicaid coverage that is uniform across the country and fills the biggest gaps in coverage for lowincome people. Specifically, the PPACA requires states by January 1, 2014, to extend Medicaid eligibility to all groups of people under age 65 with income up to 133 percent of the FPL who are not otherwise eligible for Medicaid.2 For most states, this will mean providing Medicaid to adults without children for the first time, as well as increasing their income eligibility threshold for parents to 133 percent of the federal poverty line. The law specifies different match rates for individuals eligible for coverage as of December 1, 2009; those made newly eligible for coverage under health reform and for certain expansion states. Regular Medicaid Matching Rate: The regular Medicaid matching rate is determined by a formula that has been in place since the program was enacted in 1965. It ranges from 50 percent to 76 percent, and is designed to provide more federal support to states with lower per capita incomes. In 2014, it will continue to be used for "alreadyeligible" individuals (people who qualify for Medicaid under the rules in effect on December 1, 2009).

NewlyEligible Matching Rate: The newlyeligible matching rate assures that the federal government finances much of the cost of the Medicaid expansion to 133 percent of the FPL included in the health reform legislation. It is set at 100 percent in FY2014 through FY2016, 95 percent in 2017, 94 percent in 2018, 93 percent in 2019, and 90 percent in 2020 and beyond. Beginning in 2014, it is available for nonelderly adults with income up to 133 percent of the FPL who are not eligible for Medicaid under the rules that a state had in place on December 1, 2009.

"Expansion" States Matching Rate: The transitionmatching rate is designed to provide some additional federal help to "expansion" states (states that expanded coverage for adults to at least 100 percent of the FPL prior to enactment of health reform). These states can receive a phasedin increase in their federal matching rate for adults without children under age 65 beginning on January 1, 2014 so that by 2019 it will equal the enhanced matching rate available for newlyeligible adults. This analysis assumes that there are seven states that fall into this category: Arizona, Delaware, Hawaii, Massachusetts, Maine, New York, and Vermont.

Enhanced Matching Rates for Parents and Childless Adults, 2014 and Beyond NewlyEligible Parents & Childless Adults (up to 133% FPL) 100% 100% 100% 95% 94% 93% 90% MedicaidEligible Childless Adults in "Expansion" States Only Transition Percentage used to Calculate Enhanced Match Example: State with 50% Original FMAP Regular FMAP + [(Newly Eligible Enhanced Match Rate ­ Regular FMAP) x Transition Percentage] 75% 80% 85% 86% 89.6% 93% 90% Example: State with 60% Original FMAP Regular FMAP + [(Newly Eligible Enhanced Match Rate ­ Regular FMAP) x Transition Percentage] 80% 84% 88% 88% 90.6% 93% 90%

Year

2014 2015 2016 2017 2018 2019 2020 on

50% 60% 70% 80% 90% 100% 100%

To promote coordination, the gross income standard that will be used for the premium tax credits available in the Exchanges also will apply to most existing Medicaid eligibility groups. A standard five percent of income disregard will be built into the gross income test for Medicaid to compensate for the loss of other, existing Medicaid disregards. In addition, states will no longer be able to impose asset tests on most Medicaid populations.

2

9

9

Table 1: Standard Participation Scenario
Coverage in 2019 % Reduction in Total New Previously Uninsured Medicaid Uninsured Adults < 133% Enrollees* Newly Enrolled FPL
Northeast

Spending in 20142019 (in millions)

Change From Baseline 20142019

State Federal Total Spending Spending Spending $263 $118 $1,274 $63 $533 $50 $1,054 $70 $26 $1,202 $478 $147 $166 $686 $421 $431 $106 $32 $830 $32 $205 $470 $455 $3 $42 $1,233 $714 $515 $337 $533 $429 $1,029 $549 $470 $716 $2,619 $498 $164 $4,686 $1,857 $2,137 $1,204 $9,030 $8,049 $17,086 $1,559 $112 $19,259 $8,535 $2,800 $3,477 $14,252 $7,836 $8,395 $2,345 $595 $17,130 $717 $4,252 $10,305 $9,401 $387 $902 $20,050 $14,551 $11,878 $7,273 $9,112 $9,865 $20,712 $12,179 $10,919 $11,072 $52,537 $9,629 $3,781 $4,949 $1,738 $864 $1,267 $9,563 $8,099 $18,140 $1,629 $86 $20,461 $9,013 $2,947 $3,643 $14,938 $8,257 $8,826 $2,451 $627 $17,960 $748 $4,457 $10,776 $9,856 $390 $944 $21,283 $15,265 $12,393 $7,610 $9,645 $10,294 $21,741 $12,728 $11,389 $11,788 $55,156 $10,127 $3,945

% Federal Spending 94.7% 100%* 100%* 95.0% 94.4% 99.4% 94.2% 95.7% 100%* 94.1% 94.7% 95.0% 95.4% 95.4% 94.9% 95.1% 95.7% 94.9% 95.4% 95.8% 95.4% 95.6% 95.4% 99.2% 95.6% 94.2% 95.3% 95.8% 95.6% 94.5% 95.8% 95.3% 95.7% 95.9% 93.9% 95.3% 95.1% 95.9% 94.6% 97.4% 93.7% 95.4% 100%* 96.0% 95.6% 94.8% 95.9% 95.9% 96.0% 95.6% 95.6% 95.4%

Enrollment State Federal Total in 2019 Spending Spending Spending 20.1% 11.8% 2.0% 38.8% 38.1% 6.0% 21.7% 20.0% 2.8% 25.8% 29.4% 25.3% 42.0% 30.2% 32.9% 29.8% 36.2% 44.0% 31.9% 25.9% 20.8% 36.9% 27.9% 6.7% 16.1% 34.7% 40.4% 37.3% 32.4% 32.4% 41.2% 38.2% 51.2% 38.4% 20.9% 45.5% 41.8% 29.5% 38.5% 7.7% 20.1% 47.7% 38.0% 39.4% 54.5% 61.7% 28.3% 60.6% 56.1% 25.2% 40.0%
27.4%

Connecticut Maine Massachusetts** New Hampshire New Jersey New York Pennsylvania Rhode Island Vermont
Midwest

114,083 43,468 29,921 55,918 390,490 305,945 482,366 41,185 ,484 4 631,024 297,737 114,691 143,445 589,965 251,783 307,872 83,898 28,864 667,376 31,317 205,987 351,567 200,690 12,081 28,900 951,622 646,557 329,000 366,318 245,996 320,748 633,485 357,150 344,109 330,932 1,798,314 372,470 121,635 42,794 105,428 2,008,796 245,730 84,130 85,883 57,356 136,563 145,024 294,600 138,918 295,662 29,899 15,904,173

5,864 7 7,877 2 0,401 1 4,625 3 292,489 223,175 282,014 9,147 2 3,214 429,258 215,803 4,498 7 9,265 8 430,744 132,511 207,678 0,364 5 7,198 1 462,024 8,594 1 127,862 244,804 154,836 7,916 5,308 1 683,477 479,138 250,704 277,746 174,484 256,920 429,272 261,157 247,478 245,691 1,379,713 245,840 5,675 9 3,106 3 1,095 8 1,406,101 166,471 2,381 4 9,078 5 7,978 3 100,813 111,279 211,542 8,284 7 189,463 9,099 1 11,221,455

48.0% 47.4% 10.2% 48.7% 45.3% 14.8% 41.4% 50.6% 10.2% 42.5% 44.2% 44.1% 50.9% 50.6% 44.2% 45.5% 53.9% 45.1% 50.0% 51.9% 50.6% 53.2% 47.6% 15.9% 49.1% 44.4% 49.4% 57.1% 50.7% 46.2% 54.9% 46.6% 53.1% 56.4% 43.3% 49.4% 50.6% 56.7% 48.4% 13.6% 41.5% 50.0% 50.0% 53.9% 49.6% 47.0% 52.6% 56.7% 52.5% 52.2% 53.0% 44.5%

1.2% 1.5% 2.1% 1.1% 1.2% 0.0% 1.4% 0.7% 0.6% 1.6% 2.5% 1.4% 1.7% 2.0% 1.2% 1.7% 1.5% 1.4% 1.6% 1.1% 0.9% 3.6% 4.7% 0.1% 0.9% 1.9% 2.7% 3.5% 1.7% 1.7% 4.8% 2.6% 4.0% 3.6% 2.5% 3.0% 1.8% 2.4% 2.1% 0.2% 1.5% 1.8% 0.5% 2.5% 3.7% 2.9% 2.1% 3.6% 3.7% 1.2% 1.2% 1.4%

21.0% 12.9% 3.5% 21.3% 20.9% 3.3% 17.7% 14.6% 1.9% 25.9% 22.9% 15.7% 24.0% 21.5% 22.0% 19.5% 23.5% 16.9% 19.2% 16.4% 12.7% 35.9% 38.9% 6.2% 8.3% 24.3% 28.9% 32.2% 21.6% 29.6% 37.0% 29.0% 48.2% 36.0% 20.4% 38.9% 35.1% 20.4% 36.9% 4.2% 23.0% 37.1% 46.8% 27.1% 40.0% 49.8% 21.3% 50.6% 35.3% 26.0% 26.8% 22.1%

11.1% 7.7% 0.7% 11.2% 11.1% 1.7% 10.5% 8.1% 0.9% 13.8% 16.1% 10.3% 14.8% 14.8% 11.6% 13.0% 14.4% 10.8% 12.8% 10.5% 8.0% 25.7% 29.1% 3.3% 6.1% 14.3% 19.8% 24.0% 14.4% 15.6% 28.9% 19.7% 32.7% 26.3% 14.3% 24.7% 18.4% 15.6% 19.5% 2.9% 12.3% 19.4% 24.0% 19.4% 27.9% 27.1% 15.5% 33.1% 26.2% 13.6% 14.0% 13.2%

Illinois Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska North Dakota Ohio South Dakota Wisconsin
South

Alabama Arkansas Delaware District of Columbia Florida Georgia Kentucky Louisiana Maryland Mississippi North Carolina Oklahoma South Carolina Tennessee Texas Virginia West Virginia
West

Alaska Arizona California Colorado Hawaii Idaho Montana Nevada New Mexico Oregon Utah Washington Wyoming
Total

$117 $2,046 $2,163 $56 $2,091 $2,147 $2,982 $44,694 $47,676 $286 $5,917 $6,203 $28 $2,999 $2,971 $101 $2,402 $2,502 $100 $2,178 $2,278 $188 $3,445 $3,633 $194 $4,510 $4,704 $438 $10,302 $10,739 $174 $4,129 $4,304 $380 $8,271 $8,651 $32 $683 $715 $21,148 $443,530 $464,678

*Includes newly enrolled 1115 waiver eligible population. **Massachusetts has a low share of uninsured within the newly enrolled due to low levels of uninsurance in the baseline. Note: These estimates relate solely to the Medicaid expansion and do not account for other changes in health reform such as access to subsidized coverage in the exchanges or state or federal savings from reduced uncompensated care or the transition of individuals from statefunded programs to Medicaid in 2014.

10

Table 2: Enhanced Outreach Scenario
Coverage in 2019 % Reduction in Uninsured Previously Total New Adults < 133% Uninsured Medicaid FPL Enrollees* Newly Enrolled
Northeast

Spending in 20142019 (in millions)

Change From Baseline 20142019

State Federal Spending Spending $440 $65 $628 $117 $1,078 $2,859 $2,041 $100 $8 $2,468 $899 $257 $260 $1,096 $745 $773 $155 $57 $1,335 $46 $314 $693 $761 $90 $62 $2,537 $1,233 $695 $536 $1,060 $581 $1,791 $789 $615 $1,523 $4,514 $863 $217 $219 $739 $6,544 $470 $30 $133 $155 $338 $278 $555 $227 $567 $49 $43,218 $5,048 $2,105 $2,783 $1,470 $11,129 $17,170 $19,489 $1,768 $283 $22,109 $10,112 $3,298 $4,033 $16,944 $9,116 $10,228 $2,732 $709 $19,578 $844 $4,912 $11,404 $11,523 $686 $1,068 $24,260 $17,916 $13,220 $8,937 $10,881 $10,959 $24,720 $13,436 $12,109 $13,128 $62,056 $11,129 $4,182 $2,379 $4,861 $54,936 $6,925 $3,414 $2,896 $2,558 $4,100 $5,608 $11,723 $4,695 $9,573 $818 $531,958

Total Spending $5,488 $2,040 $2,155 $1,586 $12,207 $20,030 $21,530 $1,868 $291 $24,577 $11,010 $3,555 $4,293 $18,040 $9,861 $11,001 $2,886 $766 $20,913 $890 $5,226 $12,097 $12,284 $776 $1,129 $26,797 $19,149 $13,915 $9,472 $11,941 $11,539 $26,511 $14,225 $12,724 $14,651 $66,570 $11,992 $4,399 $2,598 $5,600 $61,481 $7,395 $3,444 $3,028 $2,713 $4,438 $5,885 $12,279 $4,921 $10,139 $867 $575,176

% Federal Spending 92.0% 100%* 100%* 92.6% 91.2% 85.7% 90.5% 94.6% 97.4% 90.0% 91.8% 92.8% 93.9% 93.9% 92.4% 93.0% 94.6% 92.5% 93.6% 94.9% 94.0% 94.3% 93.8% 88.4% 94.5% 90.5% 93.6% 95.0% 94.3% 91.1% 95.0% 93.2% 94.5% 95.2% 89.6% 93.2% 92.8% 95.1% 91.6% 86.8% 89.4% 93.6% 99.1% 95.6% 94.3% 92.4% 95.3% 95.5% 95.4% 94.4% 94.3% 92.5%

Enrollment State Federal Total in 2019 Spending Spending Spending 27.3% 16.2% 5.2% 53.3% 55.4% 16.0% 30.8% 26.2% 9.7% 37.2% 42.2% 36.1% 56.2% 41.6% 45.6% 42.4% 47.8% 61.0% 43.1% 34.6% 28.0% 47.9% 39.9% 15.9% 21.5% 50.2% 56.7% 48.1% 44.9% 45.9% 53.9% 53.5% 67.4% 49.4% 29.9% 63.5% 56.7% 37.9% 53.9% 22.4% 29.9% 65.6% 49.7% 53.1% 75.0% 88.6% 39.4% 79.6% 72.8% 33.6% 53.6% 39.3% 2.0% 0.8% 1.0% 2.1% 2.5% 1.2% 2.7% 1.1% 0.2% 3.3% 4.8% 2.4% 2.6% 3.2% 2.1% 3.1% 2.2% 2.5% 2.6% 1.6% 1.4% 5.3% 7.9% 1.6% 1.3% 3.8% 4.6% 4.7% 2.8% 3.4% 6.4% 4.6% 5.8% 4.7% 5.4% 5.1% 3.1% 3.2% 3.9% 2.9% 3.4% 2.9% 0.5% 3.3% 5.7% 5.2% 3.0% 4.6% 4.8% 1.8% 1.9% 2.9% 22.6% 14.7% 4.5% 26.0% 25.7% 7.1% 20.2% 16.5% 4.9% 29.7% 27.1% 18.4% 27.8% 25.6% 25.6% 23.8% 27.4% 20.2% 22.0% 19.3% 14.7% 39.7% 47.7% 11.0% 9.9% 29.4% 35.6% 35.8% 26.5% 35.3% 41.1% 34.6% 53.2% 39.9% 24.2% 45.9% 40.5% 22.6% 42.9% 9.9% 28.3% 43.4% 53.3% 32.7% 47.0% 59.3% 26.5% 57.6% 40.2% 30.1% 32.0% 26.5% 12.3% 9.1% 1.8% 14.0% 14.1% 4.1% 12.4% 9.2% 2.9% 16.6% 19.6% 12.4% 17.5% 17.9% 13.9% 16.2% 16.9% 13.2% 14.9% 12.5% 9.4% 28.9% 36.3% 6.6% 7.3% 18.0% 24.9% 26.9% 17.9% 19.4% 32.4% 24.0% 36.6% 29.4% 17.8% 29.8% 21.8% 17.4% 23.4% 7.5% 15.8% 23.2% 27.8% 23.5% 33.3% 33.1% 19.4% 37.9% 30.0% 15.9% 17.0% 16.4%

Connecticut Maine Massachusetts** New Hampshire New Jersey New York Pennsylvania Rhode Island Vermont Illinois Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska North Dakota Ohio South Dakota Wisconsin

154,664 59,502 75,569 76,744 567,852 820,623 682,880 53,841 15,509 911,830 427,311 163,264 192,006 812,818 348,684 437,735 110,820 40,017 901,023 41,847 277,116 455,952 286,347 28,839 38,763 1,376,753 907,203 423,757 507,952 348,140 419,571 887,560 470,358 443,020 474,240 2,513,355 504,466 156,582 59,914 305,634 2,986,362 337,706 110,203 115,730 78,840 196,168 201,855 386,845 180,478 395,577 40,041 22,809,862

113,876 41,858 43,508 52,146 455,627 706,575 458,200 40,850 13,443 694,012 337,987 117,621 131,528 635,231 211,781 324,276 71,053 26,457 670,992 27,160 188,043 335,547 234,695 23,317 22,891 1,073,391 721,558 337,987 409,869 267,555 350,091 661,292 367,541 334,296 372,894 2,055,888 365,514 129,185 49,061 273,008 2,291,221 249,208 64,167 85,523 56,889 156,025 163,105 292,651 113,872 276,096 27,488 17,524,046

72.1% 71.1% 42.9% 73.4% 70.6% 46.7% 67.2% 70.9% 42.9% 68.8% 69.1% 69.6% 75.1% 74.6% 70.7% 71.0% 76.0% 69.4% 72.6% 75.8% 74.3% 72.9% 72.1% 46.9% 73.4% 69.7% 74.4% 77.0% 74.8% 70.8% 74.8% 71.8% 74.8% 76.2% 65.7% 73.6% 75.2% 76.5% 71.7% 45.6% 67.6% 74.8% 75.7% 78.1% 74.3% 72.7% 77.1% 78.4% 76.3% 76.1% 76.2% 69.5%

Midwest

South

Alabama Arkansas Delaware District of Columbia Florida Georgia Kentucky Louisiana Maryland Mississippi North Carolina Oklahoma South Carolina Tennessee Texas Virginia West Virginia Alaska Arizona California Colorado Hawaii Idaho Montana Nevada New Mexico Oregon Utah Washington Wyoming

West

Total

*Includes newly enrolled 1115 waiver eligible population. **Massachusetts has a low share of uninsured within the newly enrolled due to low levels of uninsurance in the baseline. Note: These estimates relate solely to the Medicaid expansion and do not account for other changes in health reform such as access to subsidized coverage in the exchanges or state or federal savings from reduced uncompensated care or the transition of individuals from statefunded programs to Medicaid in 2014.

11

Introduction
This paper examines the impacts of the Patient Protection and Affordable Care Act (PPACA) on state and federal Medicaid coverage and associated costs. The PPACA will expand Medicaid coverage to large numbers of the nation's uninsured population. Currently, Medicaid provides fairly broad coverage of children, but there is less extensive coverage of parents and coverage of nondisabled childless adults is generally prohibited unless a state has a waiver. The law would expand Medicaid to nearly all individuals under age 65 with incomes up to 133 percent of the federal poverty line (FPL). As has been true in the past, undocumented immigrants are not eligible for Medicaid. There is a great deal of variation today across states in terms of Medicaid coverage, uninsured rate and fiscal capacity so it is no surprise that the Medicaid expansion in health reform will affect states differently. While the new coverage requirements do not take effect until 2014, several states have raised concerns about the fiscal implications of expanded eligibility for Medicaid particularly because states are currently dealing with the severe economic downturn and the resulting sharp decline in their revenues. This analysis shows that while there will be significant increases in coverage and new federal revenues, there will be only small increases in how much more money states will be expected to spend on Medicaid from their own funds.

13

Federal Matching Rates Under PPACA
Under the PPACA, the federal government will finance the vast majority of spending for those made newly eligible for Medicaid. The PPACA will provide states, for all new eligibles, with 100 percent federal funding in 20142016, 95 percent federal financing in 2017, 94 percent federal financing in 2018, 93 percent federal financing in 2019 and 90 percent federal financing for 2020 and subsequent years. However, some states, prior to passage of the PPACA had already made childless adults eligible for Medicaid up to 100 percent FPL at lower federal matching rates than those described above under the new law. Policymakers did not want those states that had gone further than others to be financially worse off under the PPACA. Consequently, the new law phases in an increase in the federal match rates so that by 2019, federal matching rates for childless adults who have been eligible for Medicaid through Section 1115 waivers will equal the rate for newly eligible populations at 93 percent As Medicaid eligibility expands under the PPACA, new efforts are made at the state and federal level for program outreach, enrollment procedures are simplified, and the requirement to obtain coverage shifts perceptions of individual responsibility, we also anticipate significant increases in the enrollment of uninsured individuals currently eligible for Medicaid. With the exception of the childless adults in the waiver states described above, the federal government will pay current matching rates for any new enrollees who are eligible under prePPACA Medicaid rules. Under PPACA, states are now required to maintain eligibility standards in place on March 23, 2010. The different matching rates are shown in Table 1. In general, one could think of states as falling into one of three categories: States who will have very large numbers of new eligibles starting in 2014 like Alabama and Texas. These tend to be states in the south and some in the west that have low levels of current eligibility and coverage. Most of their new enrollees will be newly eligible under PPACA and they will receive the high federal matching rates for them.

14

00

States that have already covered large numbers of adults, mostly parents, through their Medicaid programs, using poverty related provisions of Medicaid law (these states do not cover childless adults through waiver programs). These include many states such as California and New Jersey. Because of higher participation rates among current eligibles, a smaller share of their new enrollees under PPACA will be from those made newly eligible.

States that currently cover parents and childless adults in Medicaid today like Massachusetts and New York, or "expansion states". o Massachusetts and Vermont that already cover childless adults with incomes above 133 percent of the FPL through Section 1115 waiver programs. These states will have no new eligibles; they will, however, receive the higher "waiver" matching rates on those currently eligible childless adults, including prior and new enrollees. o States that have extended coverage through Section 1115 waiver programs to childless adults but did not do so for those all the way up to 133 percent of the FPL. These states, Arizona, Hawaii, Delaware, Maine, and New York will receive the waiver matching rate for the childless adults that are currently eligible under these rules. Because PPACA expands eligibility for those up to 133 percent of the FPL, these states will receive the law's higher matching rates for their new eligibiles.

A major determinant of the financial impact on states of the Medicaid reforms in the PPACA will be the numbers of eligible people in each of the eligibility categories (current eligibles, new eligibles, parents, childless adults) who actually enroll in the program, i.e. the group specific participation rates. The Congressional Budget Office (CBO) seems to have assumed relatively modest participation rates in Medicaid, primarily because the law imposes no financial penalties for the lowest income people who do not obtain health insurance coverage, and this would include much if not all of those eligible for Medicaid (CBO, 2009). However, there are likely to be strong outreach efforts on the part of state and

15

federal governments, community based organizations as well as on the part of health care providers. Moreover, there will be some new generalized societal pressure to obtain insurance coverage since most people will be required to do so. For example, the insurance coverage rate for lowincome people in Massachusetts is very high, even though this population is not penalized for being uninsured. While we do not expect all states to achieve the coverage rates observed in Massachusetts, participation could be higher than assumed by CBO. Thus we present results that approximate CBO participation rate assumptions as well as a set of assumptions with somewhat higher participation rates. The key results below can be summarized as follows. Medicaid enrollment will clearly increase under health reform, by about 16.0 million and possibly more. The federal government will pay a very high share of new Medicaid spending under reform in all states. States with very low coverage rates today are perhaps the greatest beneficiaries because most of their new enrollment is from new eligibles for whom there is the extremely high federal matching rate. States with broader current coverage today, particularly of parents, have somewhat lower share of new spending borne by the federal government but the federal share still approximates 90 percent. The seven states with Section 1115 waiver programs that have provided extensive coverage to nonparents benefit from the phasein of the higher expansion rate as well as the higher match on any new eligibles. States with statefunded programs that cover adults benefit from the fact that these adults are all considered new Medicaid eligibles. Other states that do not have state funded programs but make substantial contributions to uncompensated care can thus reduce the spending and will benefit from a large influx of federal dollars. While most states will experience some increase in spending, this is quite small relative to the federal matching payments and low relative to the costs of uncompensated care that they would bear if they were no health reform.

16

00

We do not address a number of topics that would affect state revenues. We did not assume changes in state Medicaid eligibility levels above 133% FPL (after 2014), although the availability of federal subsidies for the purchase of coverage through the new health insurance exchanges mean that states are likely to stop covering these somewhat higher income people through Medicaid. This change will affect a sizable share of the medically needy population and will provide significant savings to states that have, in the past, extended coverage in this way. These individuals will then obtain subsidized coverage through the new insurance exchanges. If states do continue to cover those with incomes above 133 percent FPL, there could be higher participation because of reform ­ we do not account for this either. Second, we did not include estimates of increased costs resulting from higher physician payment rates under Medicaid. The effects of these rate increases will be fully borne by the federal government in 2013 and 2014, but not thereafter. Third, we did not examine the eligibility provisions that affect children, but these provisions generally seem to benefit states financially. Under the new law, states are required to maintain coverage levels for children in Medicaid and CHIP through 2019 and funding for CHIP is extended from 2013 through 2015. If CHIP is reauthorized by 2015, the new law provides states with a 23 percentage point increase in the CHIP match rate (up to 100 percent) and if if CHIP is not reauthorized, we assume that these children (i.e., those above 133% FPL) would likely be enrolled in exchanges with all subsidies financed by the federal government. Fourth, we did not examine the fiscal implications of the reductions of disproportionate share hospital payments. Most states will be affected by these provisions, but the payment reductions will be small in comparison to estimates of the spending changes presented in this paper. Finally, we did not examine any changes in Medicaid between 2010 and 2014. States are permitted to ex tend coverage to childless adults and receive their regular federal medical assistance percentages (FMAP) until 2014. States with state funded programs for childless adults may well take advantage of the opportunity to enroll these populations in Medicaid and will achieve significant savings by doing so.

17

Methods
The Model Database. We use the 2007 and 2008 Current Population Survey (CPS) as our baseline data set (which provides data from 2006 and 2007). Two years of data are used to increase sample size, but estimates can still be imprecise, particularly in smaller states.1 As described below, we attached eligibility indicators to identify those eligible for Medicaid under PPACA rules. The CPS has excellent income information and allows us to identify, with a fair degree of accuracy, those who would become eligible under the law's Medicaid eligibility expansion to individuals with incomes up to 133 percent of the FPL. The CPS also provides information on health insurance coverage (or lack of coverage) during the past year. However, it is generally accepted that the CPS undercounts the number of people enrolled in Medicaid, as evidenced by a substantial discrepancy between state Medicaid administrative data and CPS estimates (cites). We make a partial adjustment of the CPS data to state administrative data totals (see Dubay, Holahan, and Cook, 2007, for a complete description of this adjustment). We then reweight the total population to hit 2009 population estimates. We account for the impact of unemployment on coverage using estimates made by Holahan and Garrett (2009). This analysis estimated the impact of unemployment on changes in employer and public coverage and the uninsured. Assuming a nine percent unemployment rate, we used these results to reduce employer coverage, increase Medicaid enrollment, and increase the number of uninsured. We also benchmark to 2009 CPS total population estimates by state in addition to taking into account coverage and income distributions in the reweighting process. We estimate population growth to 2019 using growth rates based on Census population projections. Eligibility Simulation. To estimate the impact of health reform on states, we need to simulate current eligibility. Once we identify whether individuals are currently eligible, we can then estimate the impact of expanding coverage to 133 percent of the FPL.2 Individuals eligible for Medicaid, the Children's Health

18

00

Insurance Program (CHIP), and stateonly financed programs are identified using a detailed Medicaid and CHIP eligibility model developed at the Urban Institute's Health Policy Center (Dubay and Cook, 2009). The model takes into account statelevel eligibility requirements for Medicaid and CHIP eligibility and applies them to person and familylevel data from the Annual Social and Economic Supplement to the CPS, simulating the eligibility determination process. The model also accounts for the pathways by which individuals can gain eligibility. Most important for our purposes, it identifies eligibility for Section 1115 waiver programs. Because Section 1115 waiver eligibles are treated differently under reform in seven states, it is important to identify those who are eligible for and currently enrolled through 1115 waiver programs. The states that we identified as meeting section 1115 program benchmark standards include Arizona, Delaware, Hawaii, Maine, Massachusetts, New York, and Vermont. Familylevel characteristics used in determining eligibility, such as income, are based on the health insurance unit (HIU). The model takes into account childcare expenses, work expenses, and earnings disregards in determining eligibility in the baseline. However, because the CPS does not collect information on monthly income, it is not possible to determine how eligibility status changes as a result of income fluctuations throughout the year. For noncitizens, the eligibility simulation also takes into account length of residency in the United States in states where this is a factor in eligibility. To account for the possibility that some foreign born individuals are unauthorized immigrants and therefore not eligible for public health insurance coverage, the model imputes legal immigrant status. Legal immigrant status is imputed based on a model that identified immigration status on the March 2004 CPS and then was used to predict immigration status on the March 2007 and 2008 CPS file used here. Estimates derived from the model are consistent with those produced using the March 2008 CPS (Passel and Cohen, 2009).

19

Participation Rates. Once we have identified individuals who are newly eligible for Medicaid under PPACA rules, we then assess the likelihood that they will participate in Medicaid under reform. The uninsured are likely to participate at relatively high rates postreform, but not all new participation will come from the ranks of the uninsured. Some who now have employersponsored or nongroup coverage will see Medicaid as a preferred alternative, due to low or no premiums, better benefits, and lower or no cost sharing. Some are likely to drop private coverage and take up Medicaid once eligible. Participation rates are also likely to increase for those who are currently eligible but not participating in Medicaid, regardless of whether they currently have employersponsored coverage, nongroup coverage, or are uninsured, due to expanded outreach and simplified enrollment processes expected under the PPACA. Thus we make assumptions about increased takeup rates among those populations as well. In our first scenario, we make Medicaid participation assumptions that approximate those used by CBO. We adjust takeup rates so that our expenditure estimates are in line with CBO's estimates. This includes lower takeup rates for the early years of the reform plan when CBO assumes lower expenditures due to a phasing up of new Medicaid enrollment. In the early years, CBO assumes little increase in enrollment of those eligible for Medicaid under prePPACA rules. In the second scenario, we assume that the takeup rates will be higher than under the CBO consistent assumptions. The justification for higher participation rate assumptions are that individuals will respond to the presence of the new legal requirement to have coverage even though this population is largely exempt from any financial penalties for noncompliance, expectations of strong outreach efforts on the part of advocacy organizations, and the incentives providers will face to enroll beneficiaries, particularly in the light of reductions of disproportionate share hospital payments. These factors will primarily affect takeup by the uninsured and the assumptions made are consistent with the participation rates embedded in the Urban Institute's Health Insurance Policy Simulation Model

20

00

(HIPSM). The assumptions are calibrated to reflect evidence on takeup rates in public programs as well as the literature on the crowding out of private coverage under public program expansions. The two alternative sets of takeup rates are presented in Table 2 (without the phasein adjustment). Cost per Person. We use estimates of the costs per enrollee from HIPSM. These estimates are based on the Medical Expenditure Panel Survey (MEPS) but calibrated to reflect differences in health status of Medicaid eligibles who are currently uninsured, have nongroup coverage, or employersponsored insurance. HIPSM estimates are adjusted to be consistent with targets from the Medicaid Statistical Information System (MSIS). Costs per enrollee are then inflated to 2019 using growth rates from the CBO March 2009 baseline. The Baseline. In order to assess the impacts of reform, we must first construct estimates of state and federal spending in the absence of PPACA, i.e., baseline spending. Baseline enrollment and national spending totals for the years 20092019 were calculated by applying CBO's predicted Medicaid enrollment and spending growth rates from the March 2009 baseline to data on enrollment and spending from the 2007 Medicaid Statistical Information Statistics (MSIS). We adjust MSIS spending data to spending on Medicaid benefits reported by the CMS64, since the CMS64 data is considered to be a more accurate data source due to its use in the calculation of federal matching payments for the states. The "adjusted" MSIS then provides 2007 estimates of enrollment and spending for children, adults, disabled and aged for each state. These 2007 federal spending counts were grown to match 2009 spending from the CBO by enrollment group at the national level. Then these same growth rates were applied to each state. Published 2009 FMAP rates were then used to calculate the state and total

21

spending amounts in 2009 (Federal Register, 2007). This process was repeated for each year, 2010 through 2019, using CBO estimates and the most recent FMAP rates for each year (Federal Register, 2008 & 2009), without the adjustments made by the American Recovery and Reinvestment Act (ARRA).

National Results
Impact on Coverage. Table 3 presents the 2019 national coverage impacts of the Medicaid provisions in PPACA under the two alternative participation rate assumptions. Under the lower participation rate assumptions (keyed to CBO assumptions), 15.9 million lowincome individuals will be added to Medicaid under PPACA. Of these, 15.0 million are those who will be newly eligible under PPACA rules; 94.1 percent of new enrollees would be those who become eligible after the PPACA Medicaid expansion. In addition, 200,000 will be individuals already eligible for Medicaid through Section 1115 waiver programs who would newly enroll because of reform. Another 0.8 million are those adults (primarily parents) who are currently eligible for Medicaid and who would take up coverage under reform. The table also shows that there are 400,000 childless adults already enrolled in Section 1115 waiver programs who would receive enhanced matching payments. Under the higher participation rate assumption, significantly more of today's uninsured population who are currently eligible but not enrolled in Medicaid would enter the program, including "waiver" populations (i.e., childless adults currently eligible through 1115 waivers). In total, 2.8 million people who are currently eligible but not covered by Medicaid would enroll under the higher assumptions, in addition to 19.4 million made newly eligible for the program. Total new enrollment under this scenario would be 22.8 million, 85.0 percent of which would be newly eligible people. Table 3 also shows that Medicaid enrollment would increase by 27.4 percent relative to the baseline under the lower participation rate assumption and by 39.3 percent under the higher participation rate assumption. Further, Table 3 also shows that the Medicaid coverage under the lower participation rate

22

00

assumption would reduce the number of uninsured by 11.2 million; 4.7 million new enrollees would have had other coverage in the absence of PPACA. Under the higher participation rate assumption, the Medicaid expansion would reduce the number of uninsured by about 17.5 million. Overall Impact on Cost. In aggregate, across the years 20142019, state spending will increase by $21.1 billion under the lower participation rate assumption and federal spending will increase by $443.5 billion. Thus about 95 percent of all new spending will be paid for by the federal government. Under the higher participation rate assumption, state spending will increase by $43.2 billion while federal spending will increase by $532.0 billion. The share of spending borne by the federal government will be somewhat lower under the higher participation assumptions, primarily due to higher takeup among those who are eligible under prePPACA rules. Since the states will receive lower federal matching rates for those previously eligible, states will be responsible for a higher share of their costs (Table 1). The second panel of Table 4 shows that new spending in 2014 wi