U.S. Health Reform—Monitoring and Impact Redistribution of Federal Funding Across States Due to the Financing of Medicaid and CHIP October 2020 By John Holahan and Jennifer M. Haley Support for this research was provided by the Robert Wood Johnson Foundation. The views expressed here do not necessarily reflect the views of the Foundation. With support from the Robert Wood Johnson Foundation (RWJF), the Urban Institute is undertaking a comprehensive monitoring and tracking project to examine the implementation and effects of health reform. The project began in May 2011 and will take place over several years. The Urban Institute will document changes to the implementation of national health reform to help states, researchers and policymakers learn from the process as it unfolds. Reports that have been prepared as part of this ongoing project can be found at www.rwjf.org and www.healthpolicycenter.org. BACKGROUND Recent discussions about how to manage state budget crises their state received in overall federal spending, whereas the related to the coronavirus outbreak and accompanying other 42 states received more payments from the federal recession have revived long-standing debates about the government than they paid in federal taxes. distribution of federal dollars across states. Specifically, The financing structures of Medicaid and CHIP, specifically, discussion has centered on how much each state receives are intended to redistribute funds from higher-income to from the federal government compared with the amount lower-income states. Medicaid and CHIP financing is based each state contributes in federal tax revenues. In this paper, on total Medicaid/CHIP spending in the state, with costs we examine spending on Medicaid and the Children’s Health shared between federal and state governments. For most Insurance Program (CHIP), together the largest joint federal- types of Medicaid/CHIP spending, the share paid by the state programs in the US, comparing the distribution of federal government, called the federal medical assistance federal spending to the states with states’ contributions to percentage (FMAP), varies inversely with state per capita the federal treasury. We show that most high-income states income, meaning the federal government pays a higher receive fairly high levels of federal Medicaid/CHIP spending share of Medicaid/CHIP expenses in states with lower per but contribute even more to the federal treasury. Lower- capita incomes (hereafter called low-income states) and income states spend considerably less on Medicaid/CHIP and a lower share of such expenses in states with higher per thus receive less federal money for the programs, but at the capita incomes (hereafter called high-income states). In same time contribute even less to the federal treasury. 2020, state Medicaid FMAPs range from the mandated 50.0 Past research has compared all federal spending provided percent minimum (which applies to 12 states) to a high of to states with the federal tax payments from individuals and 77.0 percent in Mississippi, with the CHIP matching rate businesses across states and found considerable resulting proportionally higher in each state. income redistribution. A recent analysis found that New As noted, the federal Medicaid/CHIP payments made to a York accounts for the largest share of tax contributions from state depend on overall state spending on those programs. individuals and businesses of all states, but in 2018 it received They vary both by the generosity of the programs (i.e., $22.0 billion less in federal spending than it contributed. The expansiveness of eligibility and benefits provided, which difference is redistributed to other states, and in 2018, that states can extend beyond federal minimum requirements) meant New York’s taxpayers received $1,125 less per resident and the state’s federal matching rate. If a state has a low in federal funding than they paid in federal taxes.1 Other federal matching rate but broad eligibility and a rich benefit high-income states, such as Connecticut, Massachusetts, and package, it could receive considerable federal dollars. In New Jersey, also contributed more to the federal treasury contrast, a state with a high federal matching rate but limited than they received. Overall in 2018, residents and businesses eligibility and benefits could receive fewer federal dollars. in eight states made greater federal tax contributions than U.S. Health Reform—Monitoring and Impact 2 In a state where a high share of the population has low In this paper, we focus on how federal dollars for financing incomes, a higher share of the population may or may not be Medicaid and CHIP are distributed across states and how this enrolled in Medicaid/CHIP, depending on the state’s eligibility distribution differs from states’ contributions to the federal rules and the ease with which residents can enroll in the treasury. We calculate total federal dollars paid to all states for programs (i.e., related to administrative hurdles, enrollment Medicaid and CHIP, splitting the funds between traditional assisters, and outreach efforts). Medicaid/CHIP (including coverage for children and nonelderly adults, long-term care, and disproportionate share hospital But whether a state’s contribution to the federal government payments) and Medicaid expansion to people with incomes up is balanced with the funding they receive from the federal to 138 percent of the federal poverty level under the Affordable government also depends on each state’s relative contribution Care Act (ACA). We then examine how each state’s share of to the federal treasury. Individuals and businesses in states pay total federal Medicaid/CHIP spending differs from the share various federal taxes that finance federal programs and costs, of federal individual and business taxes it pays. For example, if including Medicaid and CHIP; Medicaid accounts for about 9 California contributes 13.3 percent of all federal taxes paid by percent of all federal spending, and CHIP accounts for a lower individuals and businesses, we estimate how federal Medicaid/ share.2 Because federal taxes tend to be progressive, states with CHIP payments to California under current law differ from lower incomes tend to pay a smaller share of federal revenues what they would be if California received 13.3 percent of total per capita than states with higher incomes. Thus, lower-income federal spending on Medicaid/CHIP. We also assess how relative states benefit from both receiving the higher federal matching payments and contributions vary depending on whether a rate and paying less to the federal treasury. Higher-income state has expanded Medicaid eligibility under the ACA, and we states have a lower federal matching rate and contribute more assess the federal Medicaid/CHIP funding distribution if every to the federal treasury. state had adopted the expansion. DATA AND METHODS We analyze prepandemic federal Medicaid/CHIP spending the ACA Medicaid expansion population. HIPSM merges for 2020 under current law and a scenario where each state two years of data from the American Community Survey, receives a share of total Medicaid/CHIP spending equal to so sample sizes are large enough to generate state-specific their contribution to federal household and business taxes. estimates. The model simulates each state’s eligibility rules, First, to assess spending under current law, we compile including whether the state has expanded Medicaid under detailed 2020 spending by state for six categories of program the ACA, and estimates expenditures separately for the spending encompassing most Medicaid/CHIP expenditures: traditional Medicaid/CHIP and Medicaid expansion populations for states that expanded Medicaid. The model calibrates state- 1. Traditional spending on Medicaid for the nonelderly, specific enrollment and spending estimates to the most recent including payments to managed-care organizations publicly available data sources from late 2019. 2. CHIP spending HIPSM is limited to health insurance coverage and acute 3. Disproportionate share hospital spending care spending for the nonelderly. Thus, we use other data sources for spending on acute care for the elderly, Medicare 4. Medicaid spending on Medicare beneficiaries ages premium payments, long-term services and supports, 65 and older, including premiums, cost-sharing and disproportionate share hospital spending. The most requirements, and payments for acute care recent available data on acute care for the elderly are from 5. Long-term services and supports the 2013–14 Medicaid Statistical Information System, and 2018 spending data on Medicare premiums and long-term 6. Adults eligible for Medicaid under the ACA’s Medicaid services and supports are from the Medicaid and CHIP expansion in the District of Columbia and the 35 states Payment and Access Commission.4 For these data elements, that expanded Medicaid eligibility as of mid-2020 we inflate estimates to 2020 levels based on actuarial data We use the Urban Institute’s Health Insurance Policy on program spending growth from the Centers for Medicare Simulation Model (HIPSM)3 to estimate 2020 spending for & Medicaid Services. Finally, 2020 disproportionate share nonelderly people enrolled in Medicaid/CHIP, which includes hospital allotments are published by the Medicaid and the traditional Medicaid population, CHIP enrollees, and CHIP Payment and Access Commission.5,6 Together, these U.S. Health Reform—Monitoring and Impact 3 sources provide comprehensive, state-specific data on most each state by the Internal Revenue Service in fiscal year Medicaid/CHIP spending components.7 2019.11 Taxes include business income taxes, individual income taxes withheld and FICA tax, individual income We compute federal spending under a 2020 baseline tax payments and self-employment tax, unemployment scenario8 that excludes changes to the FMAP enacted in the insurance tax, railroad retirement tax, estate and trust Families First Coronavirus Response Act.9 Further, estimates income tax, estate tax, gift tax, and excise taxes. Some of are based on prepandemic spending in 2020 and do not these categories are largely dedicated to Medicare, Social account for changes to enrollment or spending per enrollee Security, and unemployment compensation, but the data likely to occur during the pandemic and associated recession. available do not allow for disaggregating these specific This means we may understate Medicaid/CHIP spending types by state.12 We calculate the share of taxes paid to the by the federal government. Further, though the duration federal treasury by each state and then estimate how the of the public health emergency and financial crisis are distribution of federal Medicaid/CHIP spending would differ unknown, this analysis assumes prepandemic spending levels if it were calculated according to the state’s share of federal throughout the 2020 calendar year. We present spending taxes paid. We then compare these hypothetical federal estimates separately for the traditional Medicaid/CHIP Medicaid/CHIP dollars with the distribution of federal population (categories 1 through 5 above) and for the ACA Medicaid/CHIP spending under current law. Medicaid expansion population (category 6), both in dollars and as each state’s share of total federal spending. In addition, to rank states by the current-law generosity of their Medicaid/CHIP benefits, we sum estimated federal and For our analysis of Medicaid expansion spending, we also state Medicaid/CHIP spending in 2020 from HIPSM (for all of assess an alternative scenario in which the remaining the Medicaid/CHIP spending types listed above, including nonexpansion states have adopted the ACA’s Medicaid for the ACA expansion population), and we estimate the expansion, using estimates of additional federal Medicaid/ low-income population in each state (the number of state CHIP spending that would have been provided to residents with incomes below 200 percent of the federal nonexpansion states had they expanded eligibility in poverty level in 2020, according to HIPSM) to compute 2020.10 Modeling adoption of the expansion in additional Medicaid/CHIP spending per low-income resident. We then states not only adds expansion spending in those states rank each state by their relative spending levels. We also but increases overall spending, in turn affecting the indicate each state’s FMAP for federal fiscal year 2020 and distribution of dollars across states. each state’s per capita income in 2019.13 Next, we estimate each state’s federal tax contribution. We use data on the gross amount of taxes collected from RESULTS Table 1 shows states ranked by their total federal and South Carolina each had total Medicaid/CHIP spending state Medicaid/CHIP spending per resident with low below $5,000 per low-income resident, and nearly all had income (income at or below 200 percent of the federal Medicaid FMAPs above 60 percent. Illinois’ spending also fell poverty level). It also shows each state’s FMAP and per below this level, but it does not have low per capita income, capita income. The states with the highest Medicaid/CHIP as shown in Table 1. These results suggest that despite the spending (higher than $9,500 per low-income resident) higher Medicaid/CHIP matching rates in lower-income states tend to have high per capita incomes and lower FMAPs. that would seem to encourage spending, lower-income These include Connecticut, DC, Maryland, Massachusetts, states spend considerably less, on average, per resident with Minnesota, New York, Pennsylvania, and Vermont. Maine low income than do higher-income states. and Montana are also among the 10 highest-spending Table 2 shows federal Medicaid/CHIP spending for each states, but they have below-average per capita incomes. state for traditional Medicaid (all types of Medicaid/CHIP Nearly all of these states’ Medicaid FMAPs are at or close to spending listed above, except ACA expansion spending) 50 percent (except in Maine and Montana), reflecting their under the current funding structure and how this spending high per capita incomes.14 States with the lowest spending would be allocated across states if it were consistent with tend to have lower incomes and be in the South or West; the state’s share of federal tax contributions. The difference, Alabama, Florida, Georgia, Kansas, Nevada, Oklahoma, and shown in the far-right column, shows whether the state’s U.S. Health Reform—Monitoring and Impact 4 Table 1: Medicaid/CHIP Spending per Low-Income Resident, FMAP, Low-Income Population, and per Capita Income, by State Estimated Federal and State Medicaid and CHIP Spending Estimated Low- Per Capita Income, 2019 FY 2020 Baseline State per Low-Income Resident, 2020 Income Population Medicaid FMAP (Thousands) Dollars Rank Dollars Rank DC 15,585 1 70% 215 84,538 1 Connecticut 14,120 2 50% 882 79,087 2 Minnesota 12,857 3 50% 1,346 59,683 14 Vermont 12,582 4 54% 183 56,691 19 Massachusetts 11,618 5 50% 1,908 74,967 3 Maine 10,823 6 64% 384 50,950 30 Pennsylvania 10,252 7 52% 3,695 58,775 16 New York 9,847 8 50% 7,501 71,440 4 Maryland 9,647 9 50% 1,530 65,683 7 Montana 9,500 10 65% 322 49,074 35 New Hampshire 9,308 11 50% 294 63,880 9 Delaware 9,129 12 58% 285 54,264 22 Oregon 8,723 13 61% 1,300 52,937 26 Kentucky 8,375 14 72% 1,624 44,017 47 Alaska 8,112 15 50% 295 62,102 11 West Virginia 8,090 16 75% 668 42,336 50 North Dakota 8,073 17 50% 183 57,501 17 Washington 8,065 18 50% 2,048 64,898 8 Ohio 8,022 19 63% 3,589 50,546 32 Louisiana 7,986 20 67% 1,803 48,008 40 Rhode Island 7,959 21 53% 298 56,542 20 New Mexico 7,685 22 73% 935 43,984 48 Missouri 7,656 23 66% 1,998 49,589 34 California 7,648 24 50% 13,587 66,661 6 Indiana 7,551 25 66% 2,179 48,657 37 New Jersey 7,548 26 50% 2,235 70,979 5 Arkansas 7,502 27 71% 1,177 44,845 46 Michigan 7,439 28 64% 3,125 50,320 33 Colorado 7,296 29 50% 1,550 61,348 12 Wisconsin 6,921 30 59% 1,555 53,583 24 Iowa 6,637 31 61% 827 52,636 27 Virginia 6,483 32 50% 2,356 60,116 13 Arizona 6,289 33 70% 2,547 46,233 43 Mississippi 6,093 34 77% 1,256 39,368 51 Wyoming 5,927 35 50% 162 63,316 10 South Dakota 5,919 36 58% 252 53,925 23 Tennessee 5,915 37 65% 2,402 48,761 36 North Carolina 5,908 38 67% 3,670 47,803 42 Idaho 5,867 39 70% 627 45,642 44 Nebraska 5,425 40 55% 537 54,871 21 Hawaii 5,127 41 53% 469 57,450 18 Utah 5,101 42 68% 1,002 48,395 38 Texas 5,042 43 61% 10,619 52,504 28 Illinois 4,991 44 50% 3,865 58,935 15 Oklahoma 4,773 45 66% 1,516 47,951 41 Alabama 4,691 46 72% 1,837 43,880 49 South Carolina 4,558 46 71% 1,889 45,314 45 Nevada 4,405 48 64% 1,117 50,883 31 Georgia 4,288 49 67% 3,892 48,199 39 Florida 4,091 50 61% 7,348 51,989 29 Kansas 3,395 51 59% 886 53,453 25 Average/Total 7,012 107,771 Source: Authors’ analysis of estimated Medicaid/Children’s Health Insurance Program prepandemic spending data for 2020 and 2019 per capita income from the Federal Reserve Bank of St. Louis (see note 13). Notes: FMAP = federal medical assistance percentage. FPL = federal poverty level. CHIP = Children’s Health Insurance Program. FY = fiscal year. Low-income populations have incomes below 200 percent of the federal poverty level. FMAPs shown apply to most Medicaid spending types (e.g., Affordable Care Act expansion and CHIP matching rates are calculated differently) and reflect prepandemic levels. U.S. Health Reform—Monitoring and Impact 5 Table 2: Federal Spending on Traditional (Nonexpansion) Medicaid/CHIP under Current Law and a Scenario Where Such Spending Aligns with State Federal Tax Contributions, 2020 Federal Spending When Aligned with Federal Spending under Current Law Difference State Federal Tax Contributions  State (Millions of % of Total Federal % of Total Federal Dollars) Millions of Dollars Millions of Dollars Spending Spending Alabama 6,334 1.6% 2,978 0.7% 3,355 Alaska 1,315 0.3% 606 0.2% 709 Arizona 8,766 2.2% 5,364 1.3% 3,402 Arkansas 5,370 1.3% 3,435 0.9% 1,935 California 45,211 11.4% 53,032 13.3% -7,821 Colorado 4,941 1.2% 6,737 1.7% -1,795 Connecticut 5,579 1.4% 6,414 1.6% -836 Delaware 1,332 0.3% 2,255 0.6% -924 DC 2,165 0.5% 3,093 0.8% -928 Florida 18,860 4.7% 23,596 5.9% -4,736 Georgia 11,508 2.9% 10,595 2.7% 913 Hawaii 1,102 0.3% 1,039 0.3% 63 Idaho 2,075 0.5% 1,274 0.3% 801 Illinois 8,919 2.2% 18,231 4.6% -9,312 Indiana 9,070 2.3% 6,811 1.7% 2,258 Iowa 3,001 0.8% 2,799 0.7% 202 Kansas 1,853 0.5% 2,959 0.7% -1,106 Kentucky 7,333 1.8% 3,999 1.0% 3,334 Louisiana 7,508 1.9% 4,430 1.1% 3,078 Maine 2,516 0.6% 913 0.2% 1,603 Maryland 6,601 1.7% 8,816 2.2% -2,216 Massachusetts 10,667 2.7% 13,486 3.4% -2,819 Michigan 12,101 3.0% 9,166 2.3% 2,935 Minnesota 7,866 2.0% 11,532 2.9% -3,666 Mississippi 5,958 1.5% 1,267 0.3% 4,691 Missouri 10,204 2.6% 7,207 1.8% 2,997 Montana 1,531 0.4% 714 0.2% 816 Nebraska 1,624 0.4% 2,871 0.7% -1,247 Nevada 2,651 0.7% 2,670 0.7% -20 New Hampshire 1,254 0.3% 1,372 0.3% -118 New Jersey 7,560 1.9% 15,758 4.0% -8,198 New Mexico 3,588 0.9% 1,042 0.3% 2,546 New York 32,123 8.1% 34,266 8.6% -2,142 North Carolina 14,953 3.8% 9,862 2.5% 5,091 North Dakota 673 0.2% 746 0.2% -73 Ohio 15,910 4.0% 16,257 4.1% -347 Oklahoma 4,882 1.2% 3,273 0.8% 1,609 Oregon 5,862 1.5% 3,937 1.0% 1,926 Pennsylvania 18,597 4.7% 15,951 4.0% 2,646 Rhode Island 1,111 0.3% 1,609 0.4% -499 South Carolina 6,178 1.6% 3,133 0.8% 3,046 South Dakota 898 0.2% 902 0.2% -4 Tennessee 9,479 2.4% 7,838 2.0% 1,640 Texas 33,699 8.5% 32,843 8.3% 856 Utah 2,978 0.7% 2,734 0.7% 244 Vermont 1,110 0.3% 506 0.1% 604 Virginia 6,000 1.5% 9,389 2.4% -3,390 Washington 6,717 1.7% 11,303 2.8% -4,586 West Virginia 3,475 0.9% 791 0.2% 2,684 Wisconsin 6,512 1.6% 5,940 1.5% 572 Wyoming 492 0.1% 265 0.1% 226 Total 398,007 100.0% 398,007 100.0% Source: Authors’ analysis of estimated 2020 Medicaid/Children’s Health Insurance Program spending data and federal revenue data for fiscal year 2019 from the Internal Revenue Service (see note 11). Notes: CHIP = Children’s Health Insurance Program. Estimates are based on pre-COVID-19 baseline spending and do not account for changes to enrollment or spending per enrollee that may occur under the pandemic and associated recession. Traditional Medicaid includes Medicaid for the nonelderly and Children’s Health Insurance Program coverage (both fee-for-service and managed care), disproportionate share hospital spending, spending on long-term services and supports, and Medicaid spending on Medicare beneficiaries ages 65 and older, including premiums, cost-sharing requirements, and payments for acute care. Federal tax contributions include business income taxes, individual income tax payments and self-employment taxes, railroad retirement taxes, estate and trust income taxes, estate taxes, gift taxes, and excise taxes. U.S. Health Reform—Monitoring and Impact 6 traditional Medicaid/CHIP allocation would have been In Table 3, we examine the amount of federal funding for higher or lower if allocated according to its federal tax states’ ACA Medicaid expansion populations, along with contributions, and by what amount. As shown in Table 2, state contributions to the federal Treasury. For each state, Alabama receives 1.6 percent of federal Medicaid/CHIP we compare spending on the expansion population with dollars under the pre-COVID-19 structure, whereas it what that spending would be if it were allocated to align contributed 0.7 percent of tax contributions to the federal with tax contributions by state residents and businesses. treasury. Therefore, Alabama receives more federal dollars The resulting redistribution is considerably different than for traditional Medicaid/CHIP than it contributes toward that for traditional Medicaid/CHIP. All of the nonexpansion the program. Conversely, California receives 11.4 percent states are, by definition, net contributors to expansion of federal Medicaid/CHIP spending but contributes 13.3 population spending: they pay their share of federal taxes percent of federal taxes. Therefore, California pays more to finance care for the expansion population, but they toward the program than it receives. Table 2 shows that 22 receive no federal dollars from the program because they states, including DC, contribute a higher share of federal have declined to participate. A handful of states that have revenue than the share of federal Medicaid/CHIP spending adopted the expansion are, nonetheless, net contributors, they receive; the opposite is true for the remaining 29 states. with the largest being DC, Illinois, Massachusetts, Minnesota, and New Jersey. The largest net recipients of The largest net contributors, defined as states with a net federal Medicaid expansion funds are California ($5.4 billion) contribution (tax payments less federal matching payments) and New York ($2.7 billion). Arizona, Kentucky, Louisiana, of $3.0 billion or more, are California, Florida, Illinois, and Michigan are also significant net recipients. Minnesota, New Jersey, Virginia, and Washington. On the other hand, seven states receive $3.0 billion or more in federal If we assume all states had chosen to adopt the ACA’s Medicaid/CHIP payments than would be the case if they Medicaid expansion by 2020, using HIPSM estimates received their state’s share of total tax revenue. These states of the amount of federal spending each of the current are Alabama, Arizona, Kentucky, Louisiana, Mississippi, North nonexpansion states would receive had they expanded, Carolina, and South Carolina. the redistribution changes yet again. As shown in Table 4, almost all of the current nonexpansion states would become Among the states with the highest Medicaid/CHIP spending net federal funding recipients if they expanded Medicaid (Table 1), several are large net contributors despite also eligibility for their residents. A few, such as Florida, Kansas, making significant federal tax contributions. That is, Tennessee, and Wisconsin, would remain net funding Connecticut, DC, Minnesota, Maryland, Massachusetts, contributors, but their net contributions would be much and New York are among the states with Medicaid/CHIP smaller than if they had not expanded (as shown in table programs spending the most per low-income resident, but 3). In addition, states like Alabama, Georgia, Mississippi, all contribute a higher share to the federal treasury than the Missouri, North Carolina, Oklahoma, South Carolina, and share of federal Medicaid/CHIP spending they receive. Many West Virginia would all become significant net recipients of states spending far less per low-income resident are net federal dollars. These net recipient and contributor patterns recipients, including Alabama, Georgia, and South Carolina. are similar to those that emerged for traditional programs Some of the less generous programs, however, are in states only, as shown in Table 2. that contribute a higher share to the treasury than they receive in federal matching payments, such as Florida, Illinois, Kansas, and Nevada. U.S. Health Reform—Monitoring and Impact 7 Table 3: Federal Spending on ACA Medicaid Expansion under Current Law and a Scenario Where Such Spending Aligns with State Federal Tax Contributions, 2020 Federal Spending When Aligned with   Federal Spending under Current Law Difference State Federal Tax Contributions (Millions of State % of Total Federal % of Total Federal Dollars) Millions of Dollars Millions of Dollars Spending Spending Alabama 0 0.0% 599 0.7% -599 Alaska 232 0.3% 122 0.2% 110 Arizona 3,377 4.2% 1,078 1.3% 2,299 Arkansas 1,364 1.7% 690 0.9% 673 California 16,044 20.1% 10,658 13.3% 5,387 Colorado 1,884 2.4% 1,354 1.7% 530 Connecticut 1,342 1.7% 1,289 1.6% 53 Delaware 316 0.4% 453 0.6% -138 DC 218 0.3% 622 0.8% -403 Florida 0 0.0% 4,742 5.9% -4,742 Georgia 0 0.0% 2,129 2.7% -2,129 Hawaii 366 0.5% 209 0.3% 158 Idaho 704 0.9% 256 0.3% 448 Illinois 1,875 2.3% 3,664 4.6% -1,789 Indiana 2,619 3.3% 1,369 1.7% 1,250 Iowa 763 1.0% 563 0.7% 201 Kansas 0 0.0% 595 0.7% -595 Kentucky 3,272 4.1% 804 1.0% 2,469 Louisiana 3,024 3.8% 890 1.1% 2,134 Maine 231 0.3% 184 0.2% 48 Maryland 1,996 2.5% 1,772 2.2% 224 Massachusetts 1,520 1.9% 2,710 3.4% -1,s190 Michigan 4,320 5.4% 1,842 2.3% 2,478 Minnesota 1,446 1.8% 2,317 2.9% -871 Mississippi 0 0.0% 255 0.3% -255 Missouri 0 0.0% 1,448 1.8% -1,448 Montana 763 1.0% 144 0.2% 620 Nebraska 0 0.0% 577 0.7% -577 Nevada 786 1.0% 537 0.7% 249 New Hampshire 245 0.3% 276 0.3% -30 New Jersey 2,091 2.6% 3,167 4.0% -1,076 New Mexico 2,187 2.7% 209 0.3% 1,978 New York 9,617 12.0% 6,886 8.6% 2,731 North Carolina 0 0.0% 1,982 2.5% -1,982 North Dakota 143 0.2% 150 0.2% -7 Ohio 3,498 4.4% 3,267 4.1% 230 Oklahoma 0 0.0% 658 0.8% -658 Oregon 1,902 2.4% 791 1.0% 1,111 Pennsylvania 3,369 4.2% 3,206 4.0% 164 Rhode Island 330 0.4% 323 0.4% 7 South Carolina 0 0.0% 630 0.8% -630 South Dakota 0 0.0% 181 0.2% -181 Tennessee 0 0.0% 1,575 2.0% -1,575 Texas 0 0.0% 6,600 8.3% -6,600 Utah 786 1.0% 549 0.7% 236 Vermont 223 0.3% 102 0.1% 121 Virginia 3,239 4.0% 1,887 2.4% 1,352 Washington 3,128 3.9% 2,272 2.8% 857 West Virginia 764 1.0% 159 0.2% 605 Wisconsin 0 0.0% 1,194 1.5% -1,194 Wyoming 0 0.0% 53 0.1% -53 Total 79,986 100.0% 79,986 100.0% Source: Authors’ analysis of estimated 2020 Medicaid/Children’s Health Insurance Program spending data and federal revenue data for fiscal year 2019 from the Internal Revenue Service (see note 12). Notes: ACA = Affordable Care Act. Estimates are based on pre-COVID-19 baseline spending and do not account for changes to enrollment or spending per enrollee that may occur under the pandemic and associated recession. Federal tax contributions include business income taxes, individual income tax payments and self-employment taxes, railroad retirement taxes, estate and trust income taxes, estate taxes, gift taxes, and excise taxes. U.S. Health Reform—Monitoring and Impact 8 Table 4: Federal Spending on ACA Medicaid Expansion under Current Law and a Scenario Where Such Spending Aligns with State Federal Tax Contributions, Assuming All States Adopted Medicaid Expansion by Mid-2020, 2020 Federal Spending When Aligned with   Federal Spending under Current Law Difference State Federal Tax Contributions (Millions of State % of Total Federal % of Total Federal Dollars) Millions of Dollars Millions of Dollars Spending Spending Alabama 2,203 1.8% 923 0.7% 1,280 Alaska 232 0.2% 188 0.2% 44 Arizona 3,377 2.7% 1,663 1.3% 1,714 Arkansas 1,364 1.1% 1,065 0.9% 299 California 16,044 13.0% 16,443 13.3% -398 Colorado 1,884 1.5% 2,089 1.7% -205 Connecticut 1,342 1.1% 1,989 1.6% -647 Delaware 316 0.3% 699 0.6% -384 DC 218 0.2% 959 0.8% -741 Florida 7,092 5.7% 7,316 5.9% -225 Georgia 3,937 3.2% 3,285 2.7% 652 Hawaii 366 0.3% 322 0.3% 44 Idaho 704 0.6% 395 0.3% 309 Illinois 1,875 1.5% 5,653 4.6% -3,778 Indiana 2,619 2.1% 2,112 1.7% 507 Iowa 763 0.6% 868 0.7% -105 Kansas 840 0.7% 917 0.7% -78 Kentucky 3,272 2.7% 1,240 1.0% 2,033 Louisiana 3,024 2.5% 1,374 1.1% 1,651 Maine 231 0.2% 283 0.2% -52 Maryland 1,996 1.6% 2,734 2.2% -738 Massachusetts 1,520 1.2% 4,181 3.4% -2,661 Michigan 4,320 3.5% 2,842 2.3% 1,478 Minnesota 1,446 1.2% 3,575 2.9% -2,129 Mississippi 1,543 1.3% 393 0.3% 1,150 Missouri 2,744 2.2% 2,235 1.8% 509 Montana 763 0.6% 221 0.2% 542 Nebraska 554 0.4% 890 0.7% -336 Nevada 786 0.6% 828 0.7% -42 New Hampshire 245 0.2% 425 0.3% -180 New Jersey 2,091 1.7% 4,886 4.0% -2,795 New Mexico 2,187 1.8% 323 0.3% 1,864 New York 9,617 7.8% 10,624 8.6% -1,007 North Carolina 5,472 4.4% 3,058 2.5% 2,415 North Dakota 143 0.1% 231 0.2% -89 Ohio 3,498 2.8% 5,041 4.1% -1,543 Oklahoma 1,765 1.4% 1,015 0.8% 751 Oregon 1,902 1.5% 1,221 1.0% 682 Pennsylvania 3,369 2.7% 4,946 4.0% -1,576 Rhode Island 330 0.3% 499 0.4% -169 South Carolina 2,227 1.8% 971 0.8% 1,256 South Dakota 337 0.3% 280 0.2% 57 Tennessee 2,047 1.7% 2,430 2.0% -384 Texas 11,756 9.5% 10,183 8.3% 1,573 Utah 786 0.6% 848 0.7% -62 Vermont 223 0.2% 157 0.1% 66 Virginia 3,239 2.6% 2,911 2.4% 327 Washington 3,128 2.5% 3,505 2.8% -376 West Virginia 764 0.6% 245 0.2% 519 Wisconsin 729 0.6% 1,842 1.5% -1,113 Wyoming 174 0.1% 82 0.1% 92 Total 123,404 100.0% 123,404 100.0% Source: Authors’ analysis of estimated 2020 Medicaid/Children’s Health Insurance Program spending data and federal revenue data for fiscal year 2019 from the Internal Revenue Service (see note 11). Notes: ACA = Affordable Care Act. Estimates are based on pre-COVID-19 baseline spending and do not account for changes to enrollment or spending per enrollee that may occur under the pandemic and associated recession. Federal tax contributions include business income taxes, individual income tax payments and self-employment taxes, railroad retirement taxes, estate and trust income taxes, estate taxes, gift taxes, and excise taxes. U.S. Health Reform—Monitoring and Impact 9 CONCLUSION This analysis produced three central findings. First, Medicaid/ income states receive high shares of total federal Medicaid/ CHIP spending significantly contributes to the distribution CHIP spending (and have more generous programs than of federal funding across states. Second, funding is generally average), but their residents and businesses contribute even redistributed from high-income states to low-income states more to federal revenues. At the same time, many lower- through the traditional Medicaid and CHIP programs. Third, income states receive low shares of national Medicaid/CHIP there is considerable redistribution from some low-income spending (and have less generous programs per low-income states to other states in the financing of the ACA’s Medicaid resident), but their residents and business make even smaller expansion, because of the former having refused to expand contributions to the treasury. We also show that states that Medicaid eligibility. have not expanded Medicaid eligibility under the ACA make net contributions to the other states that have expanded, The redistribution from the traditional public insurance and many of these nonexpansion states have low per capita programs largely transfers funding from higher-income states incomes. If the remaining states were to expand Medicaid to lower-income states, with state income measured by per eligibility, however, they would receive more in federal capita income. This redistribution mostly occurs because of dollars than they contribute in taxes, as is the case with the the higher Medicaid/CHIP federal matching rates low-income traditional programs. states receive, but it also owes to the smaller tax contributions such states make to the federal treasury. Many higher- U.S. Health Reform—Monitoring and Impact 10 ENDNOTES 1 Schultz L, Cummings M. Giving or Getting? New York’s Balance of Payments with the 7 This analysis includes most types of federal and state-level Medicaid spending, Federal Government. Albany, NY: Rockefeller Institute of Government; 2020. but we exclude the following: administrative expenses, family planning services, https://rockinst.org/wp-content/uploads/2020/01/1-22-20-Balance-of-Payments. expenditures fully financed by the federal government for beneficiaries in the pdf. Accessed September 15, 2020. Qualifying Individual program, Community First Choice expenditures, and 2 Bogusz C, Feinstein A, Ready D, Salazar J. The federal budget in 2019. spending on home health services and the National Breast and Cervical Cancer Congressional Budget Office website. https://www.cbo.gov/system/files/2020- Early Detection Program. We also do not account for effects on Medicare claw- 04/56324-CBO-2019-budget-infographic.pdf. back payments related to Medicare Part D. However, we include services provided to American Indians and Alaska Natives eligible for the Indian Health Service at 3 Buettgens M, Banthin J. The Health Insurance Policy Simulation Model (HIPSM) Indian Health Service facilities or tribal or urban Indian centers. for 2020: Current Law Baseline and Methodology. Washington: Urban Institute; forthcoming. 8 Holahan J, Haley J, Buettgens M, Elmendorf C, Wang R. Increasing federal Medicaid matching rates to provide fiscal relief to states during the COVID-19 pandemic. 4 Medicaid and CHIP Payment and Access Commission. MACStats: Medicaid Urban Institute. 2020. https://www.urban.org/research/publication/increasing- and CHIP Data Book. Washington: Medicaid and CHIP Payment and Access federal-medicaid-matching-rates-provide-fiscal-relief-states-during-covid-19- Commission; 2019. pandemic. Published April 23, 2020. Accessed September 15, 2020. https://www.macpac.gov/publication/macstats-medicaid-and-chip-data-book-2/. Accessed September 15, 2020. 9 Families First Coronavirus Response Act, Pub. L. No. 116-127, 134 Stat. 177 (2020). 5 Baseline disproportionate share hospital (DSH) spending was derived from 10 Simpson M. The implications of Medicaid expansion in the remaining states: 2020 Medicaid and CHIP Payment and Access Commission data on state-level fiscal update. Urban Institute. 2020. https://www.urban.org/research/publication/ year (FY) 2020 DSH allotments; see table IA-2 in Medicaid and CHIP Payment implications-medicaid-expansion-remaining-states-2020-update. Published June and Access Commission. Report to Congress on Medicaid and CHIP. Washington: 9, 2020. Accessed September 15, 2020. We use estimates of federal spending Medicaid and CHIP Payment and Access Commission; 2020. https://www. provided by the author that exclude spending on marketplace subsidies. macpac.gov/wp-content/uploads/2020/03/March-2020-Report-to-Congress-on- 11 SOI tax stats – Gross collections, by type of tax and state – IRS data book table Medicaid-and-CHIP.pdf. Accessed September 15, 2020. Congress had previously 5. Internal Revenue Service website. https://www.irs.gov/statistics/soi-tax-stats- planned a $4 billion reduction in federal DSH allotments in FY 2020. However, the gross-collections-by-type-of-tax-and-state-irs-data-book-table-5. Accessed Coronavirus Aid, Relief, and Economic Security, or CARES, Act eliminated these September 15, 2020. cuts in FY 2020 and reduced and delayed cuts planned for FY 2021. To account for 12 As indicated in the Internal Revenue Service source above: “Collections of these changes, we use the Medicaid and CHIP Payment and Access Commission’s withheld individual income tax are not reported by taxpayers separately from total and federal “unreduced allotments” (which exclude the previously planned Old-Age, Survivors, Disability, and Hospital Insurance (OASDHI) taxes on salaries reductions) and then calculate state shares and implied federal matching rates. and wages (under the Federal Insurance Contributions Act or FICA) and individual 6 Medicaid and CHIP Payment and Access Commission. Report to Congress income tax payments along with taxes on self-employment income (under the on Medicaid and CHIP. Washington: Medicaid and CHIP Payment and Access Self-Employment Insurance Contributions Act or SECA). Thus, while Table 1 shows Commission; 2020. these amounts separately for the United States total, separate amounts are not https://www.macpac.gov/wp-content/uploads/2020/03/March-2020-Report-to- available by state.” Congress-on-Medicaid-and-CHIP.pdf. Accessed September 15, 2020. 13 Release tables: Per capita personal income by state, annual. Federal Reserve Bank of St. Louis website. https://fred.stlouisfed.org/release/ tables?rid=151&eid=257197. Accessed September 15, 2020. 14 DC’s FMAP is set at 70 percent. Copyright © October 2020. The Urban Institute. Permission is granted for reproduction of this file, with attribution to the Urban Institute. About the Authors and Acknowledgments John Holahan is an Institute Fellow and Jennifer M. Haley is a Research Associate in the Urban Institute’s Health Policy Center. The authors gratefully acknowledge helpful feedback from Jessica Banthin and Linda Blumberg. About the Robert Wood Johnson Foundation For more than 45 years the Robert Wood Johnson Foundation has worked to improve health and health care. We are working alongside others to build a national Culture of Health that provides everyone in America a fair and just opportunity for health and well-being. For more information, visit www.rwjf.org. Follow the Foundation on Twitter at www.rwjf.org/twitter or on Facebook at www.rwjf.org/facebook. About the Urban Institute The nonprofit Urban Institute is dedicated to elevating the debate on social and economic policy. For five decades, Urban scholars have conducted research and offered evidence-based solutions that improve lives and strengthen communities across a rapidly urbanizing world. Their objective research helps expand opportunities for all, reduce hardship among the most vulnerable, and strengthen the effectiveness of the public sector. For more information, visit www.urban.org. Follow the Urban Institute on Twitter or Facebook. More information specific to the Urban Institute’s Health Policy Center, its staff, and its recent research can be found at www.healthpolicycenter.org. U.S. Health Reform—Monitoring and Impact 11