Affordable, quality health care. For everyone.

How Would State-Based Individual Mandates Affect Health Insurance Coverage and Premium Costs?

Downloads

ABSTRACT

  • Issue: The Tax Cuts and Jobs Act of 2017 eliminated the financial penalty of the Affordable Care Act’s individual mandate. States could reinstate a similar penalty to encourage health insurance enrollment, ensuring broad sharing of health care costs across healthy and sick populations to stabilize the marketplaces.
  • Goal: To provide state-by-state estimates of the impact on insurance coverage, premiums, and mandate penalty revenues if the state were to adopt an individual mandate.
  • Methods: Urban Institute’s Health Insurance Policy Simulation Model (HIPSM) is used to estimate the coverage and cost impacts of state-specific individual mandates. We assume each state adopts an individual mandate similar to the ACA’s.
  • Findings and Conclusion: If all states implemented individual mandates, the number of uninsured would be lower by 3.9 million in 2019 and 7.5 million in 2022. On average, marketplace premiums would be 11.8 percent lower in 2019. State mandate penalty revenues would amount to $7.4 billion and demand for uncompensated care would be $11.4 billion lower. The impact on coverage and on premiums varies in significant ways across states. For example, in 2019, the number of people uninsured would be 19 percent lower in Colorado and 10 percent lower in California if they implemented their own mandates. With mandates in place, average premiums would be 4 percent lower in Alaska and 15 percent lower in Washington.

Background

One of the Affordable Care Act’s central aims was to reform insurance markets by sharing health care risks and costs more broadly across the healthy and sicker populations. Strategies to accomplish this goal include modified community rating, guaranteed issue, and benefit standards, with the greatest changes made to nongroup insurance markets. Spreading risks tends to decrease costs for people with medical needs and increase them for healthy people. As a consequence, financial incentives to become and remain insured regardless of health status are necessary to ensure the risk pool is large and stable. The ACA established the individual responsibility requirement — also referred to as the individual mandate — to require most people to enroll in minimum essential health care coverage or pay a tax penalty. The Tax Cut and Jobs Act of 2017 sets the ACA’s penalties for individuals who remain uninsured to $0, beginning in 2019.

The Congressional Budget Office (CBO) estimated that eliminating the individual mandate penalties would lead to an additional 3 million uninsured people in 2019.1 It also estimated that premiums in the nongroup insurance market will increase by 15 percent between 2018 and 2019. Because of the elimination of mandate penalties, fewer healthy people are estimated to enroll in nongroup insurance; thus, the average nongroup insurance enrollee will be more likely to have higher health care expenses. As a result, premiums will be higher. Other pending changes, such as expansion of short-term, limited-duration plans, are expected to worsen the nongroup risk pool and increase premiums as well. The changes, taken together, may lead to some insurers ending or limiting their participation in ACA-compliant nongroup insurance markets.2 Acting on these concerns, some states have considered or passed legislation to implement state-specific individual mandates.3 New Jersey enacted its individual mandate on May 30, 2018;4 Massachusetts did so in 2006, well before the passage of the ACA.

This analysis provides estimates of the effects of state-specific individual mandates on insurance coverage, nongroup insurance premiums, federal and state government spending (including penalty revenue to states), and demand for uncompensated care. Findings are provided nationally as if every state adopted its own individual mandate and for 48 states and the District of Columbia (but excluding Massachusetts and New Jersey because they have their own mandates under current law), assuming each state adopts a penalty structure similar to that of the ACA. We do not anticipate every state taking this approach, but present findings this way for ease of exposition and as a reference point for understanding the effects of the mandate. (A full description of our methods is available below.)

Key Findings

Our central estimates assume that state mandates are implemented in each state as soon as the federal penalties are eliminated in 2019. The effect of a mandate grows over time as health care costs grow relative to incomes; we show some of our results in 2022 to illustrate this. State mandates would have two central effects. First, more people would retain insurance coverage to avoid the penalty. Second, premiums in the nongroup market would be lower because the insurance pool will not lose healthy people that would otherwise drop their coverage without a mandate. As a result, even more people will enroll because of the lower premiums.

National Distribution of Health Insurance Coverage, 2019

If all states adopted a mandate, the number of uninsured would fall by 3.9 million people, a decrease of 11.4 percent (Exhibit 1). The uninsured rate would decline from 12.4 percent of the nonelderly (i.e., under age 65) to 11.0 percent. About 452,000 additional people would enroll in employer-sponsored insurance (through their own employer or a family member’s) with the mandates in place. Another 1.2 million people would enroll in nongroup coverage with subsidies. Another 1.7 million people would enroll in marketplace or nonmarketplace nongroup coverage without federal subsidies. Finally, 623,000 additional people would enroll in Medicaid or the Children’s Health Insurance Program (CHIP). In most cases, these will be children; when parents apply for marketplace coverage, they find out their children are eligible for Medicaid or CHIP. (See box below for comparison with CBO estimates.)

Health Insurance Coverage by Income, 2019

For people with incomes below 138 percent of the federal poverty level,5 the number of uninsured would fall by 582,000 people, or 3.8 percent, with the state mandates in place (Exhibit 2). The relatively small effect in this group, a 0.7 percentage-point drop in the share of nonelderly people uninsured, occurs because most people in this cohort are eligible for Medicaid or large marketplace subsidies, depending on where they live. Since they are eligible for free or very-low-cost insurance with minor or no out-of-pocket requirements and most are exempt from the individual mandate because of income level, they are the least likely to drop coverage when the federal penalties end and the least likely to take it up when a state penalty is put in place.

Among people with incomes between 138 percent and 400 percent of poverty, 1.8 million fewer people would be uninsured with the state mandates in place, a reduction of 12.7 percent. People in this income group are eligible for marketplace subsidies in every state if no one in the family has access to affordable employer coverage, they are not eligible for other public health insurance, and they are legal residents. About two-thirds of the 1.8 million additional insured would take up nongroup insurance coverage. The remainder would be roughly split between people enrolling in Medicaid or CHIP (mostly children who would enroll in CHIP) and employer-sponsored insurance. People who are only eligible for smaller marketplace subsidies — that is, those at the higher end of the income scale — or ineligible for subsidies are the most likely to be affected by a mandate, meaning they are most likely to become uninsured or face significantly larger premiums to retain coverage.

For those with incomes above 400 percent of poverty, 1.5 million fewer people would be uninsured with the state mandates in place, a decrease of 33.4 percent. In this income group, about 78 percent of the otherwise uninsured would take up nongroup insurance coverage with the state mandates. Almost all of the remainder would enroll in employer-sponsored insurance coverage.

Health Insurance Coverage by State, 2019

With its own mandate in place, California would see a reduction of 389,000 uninsured (10.3%) in 2019 (Exhibits 3 and 4). About 60 percent of this decrease is attributable to otherwise uninsured people enrolling in nongroup coverage. There also would be an increase in Medicaid and CHIP coverage of 167,000 people.

In New York, the number of uninsured would fall by 142,000, with the bulk coming from people taking up nongroup coverage. The effect in New York is smaller in percentage terms than in many other states because the state offers the Essential Plan (a Basic Health Program), with lower premiums and cost-sharing for people between 138 percent and 200 percent of poverty. This plan already encourages greater retention of coverage, regardless of mandates.

In Texas, the number of uninsured people would fall by 483,000. We estimate that 410,000 more people would enroll in nongroup coverage, and 48,000 more would enroll in employer coverage. Texas has not expanded Medicaid eligibility under the ACA and has not aggressively undertaken marketplace outreach and enrollment assistance. As a result, coverage gains have been smaller than average. The effects of reinstating the mandate would be relatively small as well.

Beyond Massachusetts and New Jersey, two states — Hawaii and Vermont — and the District of Columbia have explored implementing their own individual mandates. They would see reductions in the number of uninsured residents of 8,000, 4,000, and 5,000, respectively.

Marketplace Premiums, 2019

Exhibit 5 shows the changes in marketplace premiums by state that would result from each state implementing an individual mandate. We use monthly benchmark single premiums for a 40-year-old to illustrate the effect, although the percentage change in premiums would be the same for any age and any coverage level because of the ACA’s fixed age-rating curves and uniform risk pool. On average, the state mandates would reduce marketplace premiums by 11.8 percent if all states adopted the ACA’s federal individual mandate structure. The impact of the mandate varies somewhat across states (Exhibit 6). States with larger shares of healthy people who have enrolled in coverage under the ACA because of the mandate will experience larger premium declines if it is reinstituted at the state level. For example, states with more people who either receive small tax credits or no credits (based on higher income levels) will tend to have large declines. This is because enrollees who pay the full premium themselves tend to have lower health care expenses than those getting tax credits. In turn, people in better health and those who have to pay more for coverage are most likely to go uninsured without a mandate.

States with the highest marketplace enrollment rates tended to attract healthier enrollees even among those eligible for tax credits. Therefore, the average health care risk of the subsidized populations varies by state and will lead to differential individual mandate effects. Because of their Basic Health Programs, New York and Minnesota can be expected to see less of an effect in their marketplaces if a mandate were implemented. States with small nongroup insurance markets are likely to experience large effects from changes in the number of enrollees. Premiums would decrease by 21.1 percent in New Mexico and by 15 percent or more in Colorado, the District of Columbia, Kentucky, Nevada, North Dakota, Washington, and West Virginia. Premiums would fall by less than 10 percent in Alaska, Hawaii, Minnesota, Mississippi, New York, and Wisconsin.

Federal and State Health Care Spending, 2019

The flow of federal dollars would increase to most states as more people enrolled in Medicaid or took advantage of marketplace premium tax credits (Exhibit 7). In general, with more coverage, there is more federal spending. However, in 21 states, federal spending actually declines if there is a mandate in place because of a decrease in premiums due to healthier people enrolling in nongroup coverage. With a mandate, average premiums would decrease with the entrance of healthier people into the market, and since the premium subsidies are computed based on a standard premium, the average subsidy would fall at the same time. In these 21 states, the lower average premium subsidies offset the fact that larger numbers of subsidized people enroll and federal spending drops somewhat. As a result, relative to current law, federal health care spending in California would decrease by $356 million, or 0.7 percent, in Florida by $690 million or 3 percent, and in Michigan by $137 million, or 1 percent. On the other hand, Louisiana would see an increase in federal health care spending of $92 million or 1.3 percent, and Texas would see an increase of $396 million or 1.4 percent.

Spending for the state-financed portion of Medicaid and CHIP would increase by $1.1 billion nationally in 2019 (Exhibit 8). The changes in state spending are small in percentage terms across all the states, with 41 states and the District of Columbia experiencing an increase of
1 percent or less.

Individual Mandate Penalties, 2019

Exhibit 9 shows the number of tax units (i.e., families) that would pay individual mandate penalties in each state and the revenue states would collect. While the penalty structure is assumed to be the same in each state, low-income states have fewer residents as a percentage of their total populations who would be subject to the mandate. In addition, low-income families who are subject to the mandate are assessed smaller penalties, so poorer states will collect less. Also, residents in some states have shown they are less likely to enroll in coverage when the mandate is in place. Nationally, in 2019, 8.8 million families would pay individual mandate penalties; the aggregate penalties would amount to $7.4 billion if every state instituted its own mandate. This reflects an average mandate penalty across all states of $830 per family. Average penalties per family range from a high of $1,270 in Delaware to a low of $630 in West Virginia. The largest states will collect the most revenue from the penalties.

Demand for Uncompensated Care, 2019

Demand for uncompensated care would fall by $11.4 billion nationally with the implementation of state mandates (Exhibit 10). Uncompensated care is paid for by federal and state governments as well as through in-kind donations of care by providers. The effect of the mandate on uncompensated care is directly related to the decrease in the number of uninsured people and the health status of the people getting coverage.

National Distribution of Health Insurance Coverage, 2022

We also estimate the changes in health insurance coverage that would occur in 2022 if all states adopted individual mandates (Exhibit 11). Restoration of the mandate at the state level would increase insurance coverage nationally by an estimated 7.5 million people in 2022. We estimate that the number of people with employer-sponsored insurance would increase by 2.3 million people compared to there being no mandates in place (other than Massachusetts and New Jersey). An additional 1.5 million people would enroll in marketplace nongroup coverage with tax credits, 2.7 million more would enroll in nongroup coverage without tax credits, and 1 million more would enroll in Medicaid.

In 2022, the largest absolute decreases in rates of uninsured are in large states like California, Florida, and Texas (Exhibit 12). As health care costs get more expensive relative to incomes over time, fewer people tend to purchase insurance and the number of uninsured rises. However, with an individual mandate in place, the effect of health care cost growth is lessened because more people hold on to their insurance to comply with the mandate. As a result, the effect of the individual mandate on reducing the number of people without insurance increases over time in percentage terms.

Discussion

If they implement their own individual mandates, states could mitigate the negative impact the elimination of the ACA penalties will have on coverage and premiums. Massachusetts legislated its own individual mandate as part of its 2006 broad-based health reforms; New Jersey did so this year. This approach does pose significant challenges. For example, Alaska, Florida, Nevada, New Hampshire, South Dakota, Texas, Washington, and Wyoming do not have state income taxes, and thus new structures would have to be developed to collect individual mandate penalties, making the arrangement far less feasible.

In addition, the political environment in some states has been actively hostile to the ACA, making the adoption of state mandates extremely unlikely. Even states that have governors and state legislators who are generally supportive of the ACA are likely to find it politically challenging to impose mandate penalties. Still, some states are considering such a move. In addition to the individual mandate law passed by New Jersey this year,6 Vermont has passed a bill into law but must work out penalty amounts and enforcement mechanisms through a working group, with implementation requiring further legislation. D.C.’s bill is still pending, but may be resolved soon. Connecticut, Hawaii, Maryland, and Washington considered bills. These are currently inactive, although there is a chance that other bills may be considered in the future. Other states may consider such a step after the consequences of elimination of the federal penalties become evident in 2019.


Methods

This analysis estimates the coverage and health care spending effects that would occur if each state implemented an individual mandate to replace the federal penalties that will be eliminated in 2019 under the 2017 Tax Cuts and Jobs Act. We assume that each state would implement a mandate with the same structure as the ACA’s federal mandate. Massachusetts and New Jersey are the only states that currently have their own individual mandates. Massachusetts’s requirements and penalties associated with its mandate are different than the federal requirements in the ACA. The New Jersey mandate structure and penalties are very similar to the ACA. Consequently, we exclude Massachusetts and New Jersey from the state-specific tables. Our analysis relies upon the Urban Institute’s Health Insurance Policy Simulation Model (HIPSM).

HIPSM is a detailed microsimulation model of the health care system designed to estimate the cost and coverage effects of proposed health care policy options. HIPSM is based on two years of the American Community Survey (ACS), which provides a representative sample of families that is large enough to produce estimates for individual states. The population is aged to future years using projections from the Urban Institute’s Mapping America’s Futures program. HIPSM is designed to incorporate timely, real-world data when they are available. As described below, we regularly update the model to reflect published Medicaid and marketplace enrollment and costs in each state. The enrollment experience in each state under current law affects how the model simulates policy alternatives.

HIPSM is unique among microsimulation models of health coverage and costs because individual and family decisions combine the two most common types of microsimulation decision-making: elasticity and expected utility. Decision-making follows an expected-utility framework that captures factors such as individual health risk, but we add a term for each observation that represents factors involved in their observed choices that the expected-utility approach alone could not capture. These terms are set so the model leads to each person in the data making the choice they reported in the survey, and the distribution of the terms is set so the model replicates premium elasticity targets from the literature. In this way, the model has the overall population change insurance enrollment decisions at a rate consistent with the research literature. Still, individuals within the model respond to changes in prices in a way that is consistent with their characteristics and their decisions observed in the data. This approach makes it easier to simulate novel policies consistently while calibrating the model to a wide range of real-world data, such as Medicaid and marketplace enrollment.

In this analysis, we provide all results in 2019 and a subset of results for 2022. Our current law — or baseline — scenario implicitly takes into account policy changes made since early 2017 that affected health insurance coverage for the 2018 open-enrollment period; our model is calibrated to 2018 state marketplace enrollment figures7 and the most recent state-specific estimates of Medicaid enrollment.8 We also use state average marketplace premiums for the 2018 plan year. While estimates of nonmarketplace nongroup insurance enrollment are not currently available, HIPSM uses premium growth in marketplace bronze plans between 2017 and 2018 to estimate enrollment in unsubsidized nonmarketplace plans. The current-law scenario assumes the elimination of the federal mandate penalties but does not assume the expansion of the short-term, limited-duration plans in proposed regulations as they have yet to be made final.

The 2018 ACA penalty for being uninsured for a full year is equal to the maximum of 1) $695 per adult; half that amount for children and 2) 2.5 percent of household income. The penalty is capped at the national average premium of a marketplace bronze plan, and it is prorated for people uninsured for fewer than 12 months. There are a number of penalty exemptions.9 We assume that each state’s own penalty would use the state average premium of a marketplace bronze plan as the penalty cap, instead of the national average. We also assume that the state mandates would be as effective as the federal mandate.

The IRS has released state specific data on individual mandate penalty payments through 2015.10 HIPSM estimates of the number of households paying penalties by household adjusted gross income (AGI) level in each state correspond well with IRS data. However, HIPSM does not simulate monthly coverage decisions, so the model computes the amounts households would pay if members were uninsured for a full year. The IRS reports actual penalty collections and, as such, it reflects that some people are uninsured for only part of a year (and thus pay proportional penalties), the fact that some people receive hardship exemptions unrelated to individual data collected by household surveys like the ACS, as well as idiosyncrasies in the way that the law is being implemented. Consequently, we make adjustments to the level of our revenue estimates that reflect the differences between IRS and HIPSM full-year penalties per household for 2015 at each AGI level and state. These adjustments are applied to penalties computed using the tax brackets enacted by the Tax Cuts and Jobs Act of 2017.

Demand for uncompensated care for the uninsured is estimated in our model based on data from the Medical Expenditure Panel Survey–Household Component adjusted to the results of a detailed analysis of uncompensated care in 2013.11 The authors of that analysis found that the uninsured pay for about 30 percent of their health care out-of-pocket, with the remainder becoming uncompensated care. About 45 percent of uncompensated care is funded by the federal government through programs such as Medicaid Disproportionate Share Hospital (DSH) funding, Medicare DSH, and the Veterans Administration. About 24 percent is funded through state and local governments. The remainder is funded by health care providers themselves.

Comparison of Urban Institute and Congressional Budget Office Estimates

Our estimates are consistent with CBO’s 2019 estimates of the effect of eliminating the federal individual mandate.12 The CBO estimated that in 2019 eliminating the individual mandate would decrease Medicaid coverage by approximately 1 million people and nongroup insurance coverage (marketplace and nonmarketplace) by 3 million people, and would increase the number of uninsured by 4 million. Our estimates represent an inverse effect of the same size.

The estimates presented in our analysis are essentially the mirror image of the types of estimates made by CBO. Our estimates differ somewhat from CBO’s in terms of the effect of eliminating the individual mandate in 2022. They diverge most with regard to the effect on Medicaid enrollment; CBO rounds coverage effects to the nearest 1 million people, which makes precise comparisons difficult. CBO estimates that elimination of the individual mandate would decrease Medicaid coverage by 4 million people in 2022, whereas we estimate a smaller inverse effect of 1 million additional people enrolling in Medicaid with state mandates in place. Our smaller estimate may reflect the fact that CBO’s baseline assumes that some states that had not yet expanded Medicaid under the ACA by 2017 would do so in future years; we make no such assumptions. CBO estimates a 2022 nongroup coverage effect from eliminating the individual mandate of 5 million people, compared to our estimate of 4.2 million people gaining coverage nationally under state mandates. CBO estimates that 2 million fewer people would have employer coverage without a mandate, compared to our estimate of 2.3 million people gaining employer coverage with state mandates. Taken together, CBO estimates that the number of uninsured would be 12 million people higher in 2022 absent mandate penalties, compared to our estimate that 7.5 million fewer people would be uninsured with state mandates introduced across the country.

Acknowledgments

In addition to Commonwealth Fund support for this research, the Robert Wood Johnson Foundation provided substantial funding for the development of the Health Insurance Policy Simulation Model, which was used in this analysis. The authors are grateful for comments and suggestions from Kevin Lucia, Justin Giovanelli, and Sara Collins, and for research assistance from Robin Wang and Erik Wengle.

Notes

1. Congressional Budget Office, Federal Subsidies for Health Insurance Coverage for People Under Age 65: 2018 to 2028 (CBO, May 2018), https://www.cbo.gov/system/files/115th-congress-2017-2018/reports/53826-healthinsurancecoverage.pdf.

2. Sabrina Corlette et al., Insurers Remaining in Affordable Care Act Markets Prepare for Continued Uncertainty in 2018, 2019 (Urban Institute, March 2018), https://www.urban.org/sites/default/files/publication/97326/moni_insurercanvas2018_2001756.pdf.

3. New Jersey and Vermont have both recently passed legislation; Vermont’s legislation requires specification of the penalties during the course of 2019 with implementation in 2020. Connecticut, Hawaii, Maryland, Washington, and the District of Columbia have all considered or are continuing to consider their own legislation. See Dania Palanker, Rachel Schwab, and Justin Giovannelli, “State Efforts to Pass Individual Mandate Requirements Aim to Stabilize Markets and Protect Consumers,” To the Point (blog), Commonwealth Fund, June 14, 2018, https://www.commonwealthfund.org/blog/2018/state-efforts-pass-individual-mandate-requirements-aim-stabilize-markets-and-protect. As this report was going to press, an individual mandate was passed by the Council of the District of Columbia as part of a larger budget bill. It has not yet been signed by the mayor, although she is expected to do so. In addition, there are riders to the D.C. budget in the U.S. House of Representatives intended to inhibit implementation, although it is not clear they would be successful.

4. Katie Jennings, “New Jersey Will Become Second State to Enact Individual Health Insurance Mandate,” Politico, May 30, 2018, https://www.politico.com/states/new-jersey/story/2018/05/30/new-jersey-becomes-second-state-to-adopt-individual-health-insurance-mandate-442183.

5. In 2019, 138 percent of the federal poverty level in the 48 contiguous states will be $16,753 for an individual and $34,638 for a four-person household; 400 percent of poverty will be $48,560 for an individual and $100,400 for four-person household.

6. Palanker, Schwab, and Giovannelli, “State Efforts,” 2018; and New Jersey State Legislature, “A3380 Aca (1R) — New Jersey Health Insurance Market Preservation Act” (State of New Jersey, May 30, 2018), http://www.njleg.state.nj.us/bills/BillView.asp?BillNumber=A3380.

7. Centers for Medicare and Medicaid Services, 2018 Marketplace Open Enrollment Period Public Use Files (CMS, updated Apr. 4, 2018), https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Marketplace-Products/2018_Open_Enrollment.html.

8. We used Centers for Medicare and Medicaid Services Monthly Enrollment Snapshots to determine the change in Medicaid enrollment in each state since 2013. See Centers for Medicare and Medicaid Services, MMCO Statistical & Analytic Reports (CMS, updated Mar. 16, 2018), https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medicaid-Coordination-Office/Analytics.html.

9. Exemptions from the individual mandate penalties include: income below the tax filing threshold, religious conscience, members of health care sharing ministries, people not lawfully present in the United States, incarcerated individuals, people uninsured for less than three months in the year, people for whom the cost of coverage exceeds 8 percent of household income (with the 8 percent indexed over time), members of Indian tribes, a person who would be eligible for Medicaid but who lives in a state that had not expanded Medicaid eligibility under the ACA, people receiving a hardship exemption from the Secretary of Health and Human Services.

10. Internal Revenue Service, SOI Tax Stats — Historic Table 2 (IRS, updated Oct. 11, 2017), https://www.irs.gov/statistics/soi-tax-stats-historic-table-2.

11. Teresa A. Coughlin et al., Uncompensated Care for the Uninsured in 2013: A Detailed Examination (Urban Institute, May 2014), http://www.urban.org/research/publication/uncompensated-care-uninsured-2013.

12. CBO, Federal Subsidies, 2018.

Publication Details

Publication Date: July 20, 2018
Citation:

Linda J. Blumberg, Matthew Buettgens, and John Holahan, How Would State-Based Individual Mandates Affect Health Insurance Coverage and Premium Costs? (Commonwealth Fund, July 2018).

Experts

Institute Fellow, Urban Institute Health Policy Center
Senior Fellow, Urban Institute Health Policy Center
Institute Fellow, Urban Institute Health Policy Center