Tuesday, June 27, 2017

Legislative Bulletin: CBO Analysis of Senate Republican “Repeal and Replace” Legislation

On June 26, the Congressional Budget Office (CBO) released its score of the Senate Republican Obamacare legislation. CBO found that the bill would:

  • Reduce deficits by about $321 billion over ten years—$202 billion more than the House-passed legislation.
  • Increase the number of uninsured by 15 million in 2018, rising to a total of 22 million by 2026—a slight short-term increase, and slight long-term decrease, of the uninsured numbers compared to the House bill.
  • Generally increase individual market insurance premiums between now and 2020, followed by a reduction in most parts of the country. However, impacts would vary based on states’ decisions regarding benefit structures, as listed below.
  • Reduce Medicaid spending by less than the House-passed measure ($772 billion vs. $834 billion), but have greater net savings with respect to insurance subsidies ($408 billion in deficit reduction vs. $276 billion for the House bill)—calculated as repeal of the Obamacare cost-sharing and premium subsidies, offset by the new spending on “replacement” subsidies.

In its analysis, CBO noted that it continues to use the March 2016 baseline to score the reconciliation legislation (as it did with the House bill). It has done so largely because 1) its updated January 2017 baseline was not available at the time Congress passed the budget resolution in early January and 2) the ten-year timeframe of the March 2016 baseline synchs with the timeframe of the current budget resolution. Had CBO used the January 2017 budget baseline to score the bill, coverage losses would likely have been smaller—CBO has reduced its estimates of Exchange coverage due to anemic enrollment. However, because premiums spiked in 2017, thus raising spending on subsidies, the fiscal effects likely would have been similar.

Premiums:    CBO believes premiums will rise by 20 percent compared to current law in 2018, and by about 10 percent compared to current law in 2019. The increases would stem largely from the effective repeal of the individual mandate (penalty set to $0), which would lead healthy individuals to drop coverage—offset in part by new “stability” funding to insurers.

In 2020, premiums would decline by about 30 percent compared to current law, and by 2026, premiums would be about 20 percent lower than current law (premium reductions declining slightly as “stability” funding declines in years after 2021). The premium reductions would come largely because of a decrease in the actuarial value (i.e., the average percentage of health expenses covered by insurance) of plans.

CBO believes that “few low-income people would purchase coverage” despite subsidies provided under the bill, because in its estimation, deductibles for low-premium plans would be prohibitively expensive for low-income individuals—and premiums for low-deductible plans would also be prohibitively expensive. In general, CBO believes out-of-pocket expenses would rise for most individuals purchasing coverage on the individual market.

Changes in Insurance Coverage:               CBO believes that under the bill, the number of uninsured would rise by 15 million in 2018, and 22 million in 2026. Moreover, “the increase [in the uninsured] would be disproportionately larger among older people with lower income—particularly people between 50 and 64 years old” with income under twice the poverty level. With respect to Medicaid, 15 million fewer people would have coverage than under current law; however, about five million of those individuals “would be among people who CBO projects would, under current law, become eligible in the future as additional states adopted” Medicaid expansion.

CBO believes that the individual insurance market would decline by 7 million in 2018, 9 million in 2020, and 7 million in 2026. The estimate notes CBO’s belief that “a small fraction of the population” will reside in areas where no insurers would participate. A reduction in subsidies would 1) make insurers’ fixed costs a higher percentage of revenues, discouraging them from participating, and 2) reduce the overall percentage of subsidized enrollees—giving some markets a disproportionate number of unsubsidized enrollees with higher health costs. However, in these cases, CBO believes that states could take steps to restore the markets within a few years, whether by obtaining waivers and/or “stability fund” dollars.

CBO believes that effectively repealing the individual mandate would, all things equal, increase premiums in the individual market; lead some employers not to offer employer-based coverage; and discourage individuals from enrolling in Medicaid. However, CBO “do[es] not expect that, with the [mandate] penalty eliminated under this legislation, people enrolled in Medicaid would disenroll.”

Waivers:         With respect to the state waivers for insurance regulations—including essential health benefits and other Obamacare requirements—CBO believes that “about half the population would be in states receiving substantial pass-through funding” under the Obamacare Section 1332 waiver provision, which the bill would revamp. States could receive pass-through funding to reflect savings to the federal government from lower spending on insurance subsidies from the waivers. Those pass-through funds could be used to lower premiums or cost-sharing for individuals.

While CBO believes that many states would apply for waivers with respect to insurance regulations or other requirements, few would “make significant changes” to the subsidy regime, to avoid administering said regime themselves—leaving this task to the Internal Revenue Service instead. However, CBO believes that about one-fifth of the total subsidy dollars available will be provided through the waiver pass-through, rather than directly to individuals.

CBO believes that, particularly in the first few years of the waiver regime, these waivers would actually increase the budget deficit—despite a requirement in the legislation that they not do so. CBO believes that states with waivers currently pending—who can choose whether their waiver would apply under the current regime or the “new” one created by the bill—would use this arbitrage opportunity to pick the more advantageous position for their state. Likewise, the agency notes that states would use overly optimistic data estimates when defining “budget-neutrality”—and that in the first few years of the bill, “the Administration would not have enough data about experience under this legislation to fully adjust [sic] for that incentive.”

In its analysis, CBO concludes that “the additional waivers would have little effect on the number of people insured, on net, by 2026.” Most waivers would be used to narrow the essential health benefits, lowering premiums and giving savings to states as pass-through funds. While lower premiums would increase individual market coverage, it would in CBO’s estimate encourage some employers to drop coverage. Moreover, “people eligible for subsidies in the non-group market would receive little benefit from the lower premiums, and many would therefore decline to purchase a plan providing fewer benefits.” A small fraction of individuals might live in states that “substantially reduce the number of people insured,” either by re-directing subsidy assistance to those who would have purchased coverage even without a subsidy, or by taking pass-through funds and re-directing them for purposes other than health insurance coverage.

CBO believes that, in cases where states use waivers to narrow essential health benefits, “insurance covering certain services [could] become more expensive—in some cases, extremely expensive.” While states could use pass-through funding to subsidize coverage of these services, CBO “anticipate[s] that the funding available to help provide coverage for those high-cost services would be insufficient.”

Other Regulatory Changes:            CBO notes the two “stability funds”—the one short-term fund for insurers, and the second longer-term fund for states—and believes that about three-quarters of the $62 billion provided to states from 2019 through 2026 would go to arrangements with insurers to reduce premiums in the individual market—whether reinsurance, direct subsidies, or some other means.

CBO believes the six-month waiting period added to the legislation would “slightly increase the number of people with insurance, on net, throughout the 2018-2026 period—but not in 2019, when the incentives to obtain coverage would be weak because premiums would be relatively high.”

The changes in age-rating rules—allowing states to charge older applicants five times as much as younger ones, unless a state chooses another ratio—“would tend to reduce premiums for younger people and increase premiums for older people, resulting in a slight increase in insurance coverage, on net—mainly among people not eligible for subsidies,” as the subsidies would insulate most recipients from the effects of the age rating changes. However, net premiums for older individuals not eligible for subsidies would rise significantly.

CBO believes that about half the population will reside in states that will reduce or eliminate current medical loss ratio requirements. “In those states, in areas with little competition among insurers, the provision would cause insurers to raise premiums and would increase federal costs for subsidies,” CBO expects. However, this provision “would have little effect on the number of people coverage by health insurance.”

Insurance Subsidies:           In general, average subsidies under the bill “would be significantly lower than the average subsidy under current law,” despite some exceptions. For instance, while net premiums would be roughly equal for a 40-year-old with income of 175 percent of poverty, “the average share of the cost of medical services paid by the insurance purchased by that person would fall—from 87 percent to 58 percent,” thereby raising deductibles and out-of-pocket expenses. The changes “would contribute significantly to a reduction in the number of lower-income people” obtaining coverage under the bill when compared to current law.

CBO believes that the high cost of premiums and/or deductibles under the bill would discourage many low-income individuals eligible for Medicaid under current law, and who would instead be eligible for subsidies under the bill, from enrolling. “Some people with assets to protect or who expect to have high use of health care would” enroll, but many would not.

CBO also notes that “it is difficult to design plans” that might be “more attractive to people with low income” because of the mandated benefit requirements under Obamacare. For instance, it would be difficult to design plans that provide prescription drugs with low co-payments, or services below the plan’s high deductible, while meeting the 58 percent actuarial value benchmark in the bill. However, waivers could lessen these constraints somewhat, potentially yielding more attractive benefit designs.

While the bill eliminates eligibility for subsidies for individuals making between 351-400 percent of poverty, CBO believes that net premiums for individual (but not necessarily for family) coverage would be relatively similar under both current law and the bill. With respect to age, CBO believes that the addition of age as a factor in calculating subsidies, coupled with the changes to age rating in the bill, would mean that a larger share of individual market enrollees will be younger than under current law.

Medicaid Per Capita Caps and Block Grants:                         CBO believes that, in the short term (2017 through 2024), per capita caps would reduce outlays for non-disabled children and non-disabled adults, because spending would grow faster (4.9 percent) than the medical inflation index prescribe in the law (3.7 percent). However, spending on disabled adults or seniors would grow much more slowly (3.3 percent) than medical inflation plus one percent (4.7 percent). “In 2025 and beyond, the differences between spending growth for Medicaid under current law and the growth rate of the per capita caps for all groups would be substantial,” as CBO projects general inflation will average 2.4 percent.

With respect to the block grant option, CBO believes it “would be attractive to a few states that expect to decline in population (and not in most states experiencing population growth, as it would further constrain federal reimbursement).” Therefore, CBO considers the block grant to have little effect on Medicaid enrollment.

In CBO’s opinion, “states would not have substantial additional flexibility under the per capita caps. Under the block grant option, states would have additional flexibility to make changes to their Medicaid program—such as altering cost sharing and, to a limited degree, benefits.” In the absence of flexibility, CBO believes states facing the per capita caps would reduce provider reimbursements, eliminate optional services, restrict enrollment through work requirements, and/or deliver more efficient care. Specifically, “because caps on federal Medicaid spending would shift a greater share of the cost of Medicaid to state over time,” states would use work requirements to “reduce enrollment and the associated costs.”

Over the longer term, “CBO projects that the growth rate of Medicaid under current law would exceed the growth rate of the per capita caps for all groups covered by the caps starting in 2025.” As a result, CBO believes Medicaid enrollment would continue to decline after 2026 relative to current law.

Medicaid Expansion:           Currently, about half of the population resides in the 31 states (plus the District of Columbia) that have expanded Medicaid. CBO believes that, under current law, that percentage will rise to 80 percent of the newly eligible population by 2026. Under the bill, CBO believes that no additional states will expand Medicaid—resulting in coverage “losses” compared to current law, albeit without individuals actually losing coverage. Moreover, as the enhanced federal matching rate for the Medicaid expansion declines under the bill CBO believes the share of the newly eligible population in states that continue their Medicaid expansion will decline to 30 percent in 2026.