Issue: Presidential candidate Hillary Clinton has proposed modifications to the Affordable Care Act to limit consumers’ out-of-pocket health spending. Goal: We analyzed four of these policies—cost-sharing tax credits to offset spending above 5 percent of income; reduced premium contributions for marketplace enrollees; a fix to the ACA’s “family glitch,” which leaves some families with expensive employer coverage; and the introduction of a public option on the marketplaces. Methods: RAND’s COMPARE microsimulation model. Key findings and conclusions: These policies would increase the number of insured individuals by 400,000 to 9.6 million, and decrease consumers’ health spending relative to current law. Cost-sharing tax credits have the biggest effect—increasing coverage by 9.6 million and decreasing average spending by up to 33 percent for those with moderately low incomes. However, the policies with the largest coverage gains also increase the federal deficit, with impacts ranging from –$0.7 billion to $90 billion.
OVERVIEW OF POLICY OPTIONS AND APPROACH
While the Affordable Care Act (ACA) has insured approximately 20 million people and extended subsidized coverage to millions of individuals,1 health care costs remain a significant concern for many Americans. As part of her presidential campaign, former Secretary of State Hillary Clinton has proposed several modifications to the ACA to make health care more affordable for consumers. In this analysis, we estimate the impact of four of Clinton’s proposed policies on families’ health care spending, health insurance enrollment, and the federal deficit. The policies we consider are:
HOW ARE MARKETPLACE PREMIUM CONTRIBUTIONS DETERMINED?
Under current law, people with incomes between 100 percent and 400 percent of the federal poverty level and no other affordable source of health insurance are eligible to receive advance premium tax credits (APTCs) on the ACA’s marketplaces. APTC-eligible individuals and families must contribute a percentage of their income toward coverage; the federal government then provides a tax credit to subsidize the additional cost of insurance, up to the cost of the second-lowest-cost silver plan available in the enrollee’s community. Enrollees’ required contributions vary with income, and increase slightly each year to account for health care cost inflation. In 2016, required contributions ranged from 2.01 percent of income for people with incomes between 100 percent and 138 percent of poverty to 9.66 percent of income for those with incomes between 300 percent and 400 percent of poverty.
Clinton’s plan would reduce the maximum amount of the required contribution from 9.66 percent of income to 8.5 percent of income, a factor of roughly 12 percent. Because the required premium contribution scales with income, we assume that the 12-percent reduction would be applied at all income levels. The required percentage contribution under Clinton’s plan would therefore range from 1.77 percent of income for those with incomes between 100 percent and 138 percent of poverty to 8.5 percent of income for those with incomes between 300 percent and 400 percent of poverty.
For example, under current law, a single individual with income at 350 percent of the federal poverty level ($41,580) would be required to contribute $4,017 (i.e., $41,580 x 0.0966) toward marketplace coverage. If the second-lowest-cost silver plan available to this individual cost $4,500, the individual’s APTC would be $483 (i.e., $4,500 – $4,017). Under Clinton’s plan, the individual’s contribution would be reduced to $3,534 (i.e., $41,580 x 0.085), and the APTC amount would increase to $966 (i.e., $4,500 – $3,534).
WHAT IS THE FAMILY GLITCH?
The Affordable Care Act permits individuals and families to receive APTCs if they have income between 100 percent and 400 percent of the federal poverty level and no other affordable source of health insurance coverage. Employer coverage is considered affordable if the worker’s premium contribution for self-only coverage is less than 9.66 percent of income. However, employers typically require workers to contribute more for family coverage than for individual coverage. For example, a worker’s annual premium contribution for individual coverage might be $1,000, while the premium contribution for family coverage is $4,000. In this situation, coverage for a worker with income at 150 percent of poverty ($36,450 for a family of four) would be considered affordable because the $1,000 premium contribution for individual coverage is less than 9.66 percent of income. However, the worker would need to spend 11 percent of income on health insurance to enroll in a family plan. Because of the “family glitch,” the worker and her family are ineligible for marketplace tax credits, even though they cannot obtain employer coverage without spending more than 9.66 percent of income on health insurance.
- Cost-sharing tax credit of up to $2,500 per individual or $5,000 per family to offset the cost of out-of-pocket spending that exceeds 5 percent of income. This would be available to all individuals enrolled in private coverage.2 In this scenario, out-of-pocket spending includes employee premium contributions for employer-sponsored coverage, premium payments for marketplace coverage after taking into account existing credits, and patient cost-sharing at the point of service. The tax credit is refundable and applied against the sum of premium contributions and out-of-pocket cost-sharing.
- Reduction in the maximum premium contribution individuals must make to enroll in a benchmark plan on the ACA’s marketplaces. Under current law, eligible marketplace enrollees receive an advance premium tax credit (APTC) equal to the premium of a benchmark health plan in their geographic area, minus a maximum premium contribution that currently ranges from 2.01 percent to 9.66 percent of income. The APTC will grow over time at the rate by which health care spending growth exceeds inflation (see the “How Are Marketplace Premium Contributions Determined?” text box). The APTCs effectively limit premium contributions to between 2.01 percent and 9.66 percent of income for those who qualify, if they choose a benchmark plan. APTCs are available to those with incomes between 100 percent and 400 percent of the federal poverty level (i.e., $24,300 to $97,200 for a family of four) and no other source of affordable coverage. The maximum premium contribution scales with income; those at the lower end of the income eligibility range contribute the lowest percentage. Clinton’s plan would reduce the maximum premium contribution from 9.66 percent to 8.5 percent of income for those with incomes at 400 percent of poverty, with proportional reductions for those at lower income levels.3
- Elimination of the so-called “family glitch” and reduction in maximum premium contribution.4 Under current law, families with access to employer coverage are eligible for APTCs only if the worker’s premium contribution for single enrollee coverage exceeds an affordability threshold of 9.66 percent of income. Because family contributions often exceed single contributions, many families with unaffordable employer coverage are precluded from receiving APTCs (see the “What Is the Family Glitch?” text box). Clinton’s proposal would “fix” this issue by giving families access to APTCs if the enrollee contribution for family employer coverage exceeded 8.5 percent of income. Because the affordability threshold that triggers APTC eligibility for those with employer coverage is tied to the maximum premium contribution on the marketplaces (9.66 percent of income under current law versus 8.5 percent of income under Clinton’s plan), we modeled the elimination of the family glitch in combination with the reduction in the maximum premium contribution.
- Introduction of a public health insurance option into the ACA’s marketplaces. This idea was frequently discussed before the ACA was passed.5 We assume that the public plan would reimburse hospitals and physicians at Medicare rates and could achieve administrative savings relative to private plans. Although these cost-saving strategies would make the public option cheaper than a comparable private plan, we assume the option would be slightly less popular, depending on its price, because of potential access constraints introduced by lower reimbursement. It is possible people could prefer the public plan if, for example, it has larger networks or fewer restrictions on service use. However, we have limited basis to assume that the public plan would be preferred to the private option; prior research has found that doctors are less likely to accept new Medicare patients than patients with private insurance.6 We further assume that the public option would put downward pressure on private plan premiums, as a result of increased competition.
We modeled the cost-sharing tax credit, the reduction in the maximum premium contribution, and the public option as separate policies. For reasons described above, we modeled the elimination of the family glitch in combination with the reduction in the maximum premium contribution. In prior work, we estimated the impact of eliminating the family glitch given the maximum premium contributions specified by the ACA.7 Our analysis focuses on several critical policy options that the Clinton campaign announced as of May 2016. The campaign offered several additional policy options in July 2016; we added the public option scenario in response to these changes. We plan to model Clinton’s proposed Medicare buy-in, which was also announced in July, at a later date.
We do not present a combined policy scenario in this brief because, even if we combined all of the options considered, this would still not represent the entirety of Clinton’s health reform proposals. For example, Clinton would also impose spending limits for those with high pharmaceutical costs, change negotiation strategies with drug companies, extend 100 percent federal matching rates to encourage additional states to expand Medicaid, and allow people ages 55 and over to buy into the Medicare program. Some of her proposals that would affect health care in the United States are not directly linked to insurance expansions; for example she would invest in research and development to improve treatments for Alzheimer’s disease, make public health investments to reduce exposure to lead and other environmental toxins, and increase funding for primary care services at community health centers.8
In our technical appendix, we report the effects of combining all four policies addressed in this brief, recognizing that this is still just a subset of Clinton’s proposals.
We conducted the analysis using the RAND COMPARE microsimulation model,9 an analytic tool that uses economic theory and data to estimate the effect of health reform proposals. For this issue brief, we updated the model to ensure consistency with the most recent estimates of the ACA’s impact on coverage. We provide an overview of the model, along with a discussion of the updates, in the technical appendix. (Also see a To the Point blog post and a digital feature on both candidates' proposals and their implications.)
We estimate that each of the policies under consideration would lead to an increase in the number of people with insurance. We estimate that 251.6 million people would be insured in 2018 under the ACA as currently enacted (Exhibit 1). Adding a cost-sharing tax credit would increase the number insured by approximately 9.6 million. We estimate that approximately 25 million people would be uninsured in 2018 under the ACA (see Appendix Table A.2), so a 9.6 million person increase in insurance represents a 39 percent decline in the share of people without coverage. Reducing premium contributions for marketplace coverage would lead to a 1.7 million person increase in insurance enrollment relative to the ACA; addressing the family glitch in combination with this policy insures an additional 1.1 million people. Adding a public option insures an additional 400,000 people relative to the ACA alone.
Our analysis suggests that—of the four policies considered—the cost-sharing tax credit would have the largest effect on coverage. This is in part because the cost-sharing tax credit applies to all individuals who enroll in private coverage, regardless of income or program eligibility. The other policies are more narrowly targeted. For example, the reduction in maximum premium contribution applies only to individuals with incomes between 100 percent and 400 percent of poverty and no access to affordable employer coverage or public insurance. Further, many of the policies are designed to reduce out-of-pocket costs for those who already have insurance; expanding coverage may not be the primary goal of the policy.
Exhibit 2 shows the population of individuals who could potentially benefit from each of the policies considered. Roughly 178 million people are targeted by the cost-sharing tax credits, compared to only 20 million people who would be eligible for reduced marketplace premium contributions. The family glitch fix affects only 5 million, a very small segment of the population. While most people would be eligible to enroll in the public option, the plan is of greatest value to those who do not have access to employer coverage or Medicaid, an estimated total of 42.7 million people, including 22.6 million who would enroll in private nongroup coverage or marketplace plans under the ACA. For all policies, a large segment of the eligible population is estimated to be already insured under the ACA, but would gain access to new or enhanced tax credits or additional insurance options with the Clinton plan.
Consumer Out-of-Pocket Spending
In Exhibit 3, we show the effect on total out-of-pocket health care spending (i.e., premium contributions plus out-of-pocket cost-sharing) for everyone with insurance, by family income. We excluded the uninsured from these analyses; uninsured individuals tend to have very low health spending, reduced access to health care, and higher risk of forgoing necessary care or experiencing catastrophic health expenses.
There are three striking findings. First, all the policies reduce the insured populations’ out-of-pocket spending on health care relative to the ACA. However, the magnitude of the effect varies depending on the policy and the result of differences in the size of the population eligible for the policy. For example, the cost-sharing tax credits affect people with employer coverage and low-income individuals in states that did not expand Medicaid. In contrast, the reduction in premium contributions affects only those who are eligible for APTCs on the ACA’s marketplaces. Fixing the family glitch, which we modeled in combination with the reduction in marketplace premium contributions, affects an even smaller segment of the population—those with access to employer coverage where a family contribution is in excess of 8.5 percent of income. While the public plan introduces a relatively low-cost option on the marketplaces, this policy generally has a modest effect on consumer spending because many people who would enroll in the public plan already receive APTCs. However, in the public-option scenario, spending declines by approximately 9 percent for those with incomes between 139 percent and 250 percent of poverty ($33,534 to $60,750 for a family of four). In some cases, individuals in this income range can enroll in the public option at no cost. This occurs when the APTC amount, which can be based on the price of a private plan, is large enough to cover the full cost of the public plan.10
Second, in all scenarios, total out-of-pocket health spending increases as income moves up to 400 percent of poverty, but then falls or remains in a close range for individuals with incomes above 400 percent of the federal poverty level. The decline in spending for individuals above 400 percent of poverty reflects the fact that most people in this income range are insured through employer coverage and typically face relatively low out-of-pocket premium contributions. Insured individuals with incomes between 250 percent and 400 percent of poverty ($60,750 to $97,200 for a family of four) have the highest health spending of any income group; they may be enrolled in marketplace plans with low actuarial values11 and they are also eligible for fewer subsides than lower-income individuals.12 Even when enrolled in employer coverage, individuals in this income range may receive a less-generous employer policy than higher-income workers.
Third, relative to the ACA, the cost-sharing tax credit leads to significant reductions in out-of-pocket spending particularly for low- and moderate-income individuals. Lower-income individuals who are not otherwise enrolled in public coverage are more likely than higher-income individuals to be eligible for the tax credit, because even a small health expense can lead to health spending in excess of 5 percent of their income. On a proportional basis, the effect is particularly large among insured people with incomes between 139 percent and 250 percent of poverty. With the tax credit, these people experience a 33 percent reduction in spending. The estimated spending reduction is lower (23%) for those with incomes under 139 percent of poverty, partly because many are enrolled in Medicaid and therefore unlikely to spend more than 5 percent of their incomes on health care.
In Exhibit 4, we consider the effect of each policy on the deficit. The cost-sharing tax-credit scenario has the largest effect on the deficit, increasing the estimated impact by $90.4 billion, relative to the ACA. This larger impact reflects the fact that this policy targets a larger share of the population than other reforms. The tax credits extend to everyone with private insurance (including employer-sponsored coverage) and people with incomes above 400 percent of poverty. But despite the large estimated increase in the deficit, the effect is mitigated to some extent by a reduction in Medicaid enrollment. In some cases, Medicaid-eligible individuals would enroll in employer-sponsored or other private coverage to take advantage of the cost-sharing credits, thereby reducing Medicaid spending while increasing outlays related to the credit.
Reducing the maximum premium contribution alone leads to a $3.5 billion dollar increase in the deficit, primarily because the government would spend more money on APTCs. Adopting this policy in combination with the family glitch fix increases the deficit more because additional people become eligible for tax credits. The public option marginally reduces the deficit, despite slightly higher insurance enrollment under this policy relative to the ACA. When a public option is introduced, the federal government reaps two forms of savings. First, we assume that private premiums fall slightly because of competitive pressures, reducing APTC outlays. Second, we assume that in some areas APTCs will be tied to the public option, which is less expensive than a private plan and less costly to the federal government. (A full discussion of how we model the public option and its effects on government spending can be found in the technical appendix.)
All of the policies considered increase the number of insured people and reduce consumers’ out-of-pocket spending on health care. The cost-sharing tax credit, which affects the largest segment of the population, increases insurance coverage by nearly 10 million, decreases average consumer spending by as much as 33 percent, and increases the federal deficit by $90 billion in 2018. For the currently uninsured population, the cost-sharing tax credit acts as an alternative to the APTCs, potentially reaching low-income uninsured people in states that did not expand Medicaid. Specifically, for people ineligible for APTCs, the cost-sharing tax credit subsidizes insurance premium contributions that exceed 5 percent of income, up to a maximum of $2,500 annually for an individual or $5,000 for a family. Similarly, APTCs subsidize premiums in excess of a required percentage contribution, which ranges from 2.01 percent to 9.66 percent of income, up to the cost of the second-lowest-cost silver plan in an individual’s community.
The cost-sharing tax credits reduce consumer out-of-pocket spending for all groups; people with incomes above 400 percent of poverty will see a 7 percent reduction in spending. The reduction in spending even for those with high incomes suggests an opportunity for targeting the tax credit. For example, it would be possible to reduce or eliminate the credit for higher-income individuals, either to reduce the impact on the deficit or to increase the credit amount for people with very low incomes.
Relative to the cost-sharing tax credits, the other policies have more modest effects on coverage, out-of-pocket-spending, and the federal deficit, primarily because these policies are more narrowly targeted than the cost-sharing tax credits. However, these policies have a greater effect on the small subset of people to whom they are targeted.13 For example, marketplace enrollees who switch from private coverage to the public plan experience an average 17 percent decline in out-of-pocket spending.
Three of the four policies considered here increase the federal deficit. The cost-sharing tax credits, which have the biggest impact on coverage and spending, have the largest impact on federal outlays. We have not modeled how these credits would be financed, but this policy would likely require new taxes or offsetting savings from other proposals, like reductions in Medicare drug spending. If new taxes are required to finance the cost of the proposed options, the net impact to individuals’ pocketbooks might change. For some, the savings generated by the tax credit could be reduced or even outweighed by additional spending in the form of tax payments.
It is also unclear how the proposed policies would affect long-term growth in health care spending or how this growth would affect the federal deficit. Because new tax credits shield consumers from the effects of higher costs, consumers may opt to use more care, providers may increase prices, or insurers may relax utilization management processes. Such changes may ultimately cause national health spending and the federal deficit to increase. At the same time, competitive pressures created by the public plan and other Clinton policies, such as leveraging Medicare’s bargaining power to lower prescription drug costs, may reduce the rate of health care cost growth and the deficit. Our analysis does not consider how Clinton’s proposals may affect the long-term trajectory of health spending in the United States.
Clinton’s plan includes numerous additional policies, such as new protections for prescription drug users, extending 100 percent federal matching for the first three years to states that newly expand their Medicaid programs, offering a Medicare buy-in for individuals ages 55 to 64, and allowing undocumented immigrants to buy into the marketplaces without federal subsidies. Because we did not consider all these policies, we cannot estimate the full effect of Clinton’s health reform proposals in combination. We present the combined effect of the four policies considered in this brief in the technical appendix. The effects on coverage and the federal deficit under the combined scenario are similar to the effects of the cost-sharing tax credits implemented individually. However, combining the four options adds additional cost-sharing protections for low- and moderate-income individuals with insurance.