After King v. Burwell: Next Steps for the Affordable Care Act

Image result for the urban institute

In this paper we have argued that the ACA has already achieved some major milestones. The law has
reduced the number of uninsured Americans by about 15 million people. It has reformed the nongroup
insurance market, no longer allowing insurers to discriminate against high-risk individuals.

Furthermore, the marketplace has been structured to assure considerable competition and has resulted
in surprisingly moderate premiums in 2014 and 2015. Health care growth has been slow by historical
standards, in part because of policies adopted in the ACA. In contrast to fears of widespread employer
dropping of insurance coverage, there appears to have been no loss in employer-sponsored insurance.
Finally, there have been no adverse effects on employment.

But at the same time, there are many reasons to believe the law is underfunded. The original
budgetary cost for the ACA’s coverage expansion was under $1 trillion, with financing coming from a
combination of cuts in Medicare and Medicaid and new taxes. The amount that many individuals are
expected to pay in premiums is still relatively high as a percentage of income. Further, premium
subsidies were tied to silver-level (70 percent AV) plans, a metal tier that has relatively high deductibles
and other forms of cost-sharing. The high premiums coupled with high cost-sharing not only can lead to
substantial financial burdens for some people, but also may have an adverse effect on enrollment.

Further, the premium tax credit caps are indexed to increase over time as medical costs grow faster
than general inflation, meaning that household financial burdens will increase over time as well.
Another problem is that families that include a worker who has an affordable employer offer are
typically not eligible for financial assistance in the marketplaces, even if the cost of family coverage
through the employer is very high. Finally, as of this date, 21 states are not participating in Medicaid,
leaving large numbers of very low income individuals without coverage.

In addition, the administrative functions in the law have been underfunded considerably. IT systems
continue to need upgrades and ongoing operational support. Efforts at education, outreach, and
enrollment assistance are in need of more federal financial support. Finally, increased support is needed
at the federal and state levels for oversight and enforcement of insurance regulations; the premise of
the law is that we can build upon a regulated private insurance market and doing so requires adequate

Given this set of problems, we propose reducing the amount of nongroup insurance premiums that
individuals are expected to pay at each income level to make coverage more affordable. We would tie
premium tax credits to gold plans rather than to silver (80 percent AV rather than 70 percent) and
improve cost-sharing subsidies for low-income people. Further, we propose eliminating the indexing of
premiums tax credits so that their value does not erode over time. We would fix the family glitch by
allowing family members to obtain subsidized coverage through the marketplaces even if one of the
adults has an affordable offer of single coverage. We would modify the ACA’s affordability standard to
make it consistent with the highest nongroup premium tax cap that we propose and the employersponsored
insurance firewall exemption level.

Next, we would address the reluctance of the 21 states to expand Medicaid up to 138 percent of
FPL by giving all states the option of extending coverage up to 100 percent of FPL. Many states that
have ideological reasons for opposing expansion of Medicaid are more comfortable covering individuals
below the poverty level in public insurance programs, and thus this option may induce many of the
remaining states to participate. It may also result in many states that are already covering individuals up
to 138 percent of FPL reducing coverage levels to those technically in poverty. Moving some current
Medicaid enrollees into marketplace plans clearly comes with trade-offs. For example, some consumers
would have modest increases in out-of-pocket costs, although our improved subsidy schedule would
limit that exposure. All enrollees would be subject to open enrollment period requirements, which
Medicaid does not have. Some states provide additional services through Medicaid (e.g., transportation
to providers) that may not be covered in marketplace plans, but some people would gain access to a
broader set of providers than they have in the Medicaid program. States with Medicaid expansions are
clearly experiencing larger increases in coverage than nonexpansion states (Long et al. 2015), and if the
approach moves more states to participate, it would go a long way toward redressing the indefensible
inequity of subsidizing higher-income individuals while providing no assistance to many of the nation’s
poorest residents.

Taken together, these measures designed to improve affordability would increase enrollment to
levels at least commensurate with original projections and likely to even higher levels. We also propose
additional funds to support IT system development and ongoing improvements, support for state
education, outreach and enrollment assistance efforts, and for increased oversight and enforcement of
federal and state insurance regulations.

Our preliminary estimate of the total cost of these reforms is $453–559 billion over the 10-year
period 2016–2025, with $78 billion of this amount not requiring additional revenues to finance it as
noted previously. We estimate that improving the premium and cost-sharing subsidies would cost $221
billion. Fixing the family glitch would add another $117 billion, although fixing this problem through
federal regulations means having to raise revenue for only a fraction of that cost. The option to extend
Medicaid to 100 percent of FPL would cost $100–200 billion in new Medicaid and subsidy costs,
depending on how nonexpanding and expanded states respond to the new option. A rough estimate for
increasing the financing of administrative functions (IT, outreach and enrollment, oversight and
enforcement of insurance regulations) is an additional $15–21 billion. Although a large sum taken
together, these additional investments would add only 0.20 to 0.24 percent of GDP to the cost of the
program. The current costs of the coverage expansions in the program have been estimated to be 0.74
percent of GDP. Even expenditures of 0.94–0.98 percent of GDP to solve a major national problem do
not seem excessive. As we have pointed out, national health expenditures over the period 2014–2019
were projected in 2014 to be $2.5 trillion less than originally projected in 2010, and thus these
proposed investments would cost substantially less than national savings resulting from lower than
expected national health expenditures.

We propose several ways in which these costs could be paid for. The first option is to extend
Medicaid drug rebates to all dual eligibles, providing $103 billion over 10 years. Increases in cigarette
and alcohol taxes, a second option, have been estimated by CBO to result in $34–66 billion,
respectively, over 10 years. Increasing the Medicare hospital insurance tax on wages by 0.2 percent
would yield another $160 billion, a third financing option. Finally, eliminating the excise or “Cadillac” tax
and replacing it with a cap on the employer-sponsored insurance tax exclusion for health care costs
above a certain threshold would yield a large sum of money. For example, a cap at the 50th percentile of
employer-based insurance costs would yield $537 billion over 10 years, even after accounting for
added Medicaid and subsidy costs resulting from some employer dropping of coverage. If a mix of the
other aforementioned revenue sources or others not mentioned here were used, nowhere near this
much money would be required from the employer exclusion. Setting the cap, for example, somewhere
between the 70th–75th percentile of employer-based insurance costs and combining that revenue with
some of the other possible revenue sources would yield sufficient funds.

We have not attempted to address all of the important issues related to the ACA and health
insurance affordability here. For example, low-income workers with access to employer-based
coverage deemed affordable under the ACA are not currently provided financial assistance, yet many
face high cost-sharing requirements that could limit their access to necessary care. Providing costsharing
subsidies to this population is another area worthy of analysis and policy development. Some
controversial components of the ACA which do not play a fundamental role in the coverage expansions
could be debated as possible trade-offs for further investments like those proposed here. Such
components include the employer mandate and the Independent Payment Advisory Board (IPAB). As
we have shown elsewhere, the employer mandate contributes little to coverage but has resulted in
considerable business opposition to the law overall. Given IPAB’s limited authority to control Medicare
costs and the slowdown in Medicare cost growth to levels below the targets which would trigger action
by the IPAB, it may be another candidate for tradeoffs.

However, it is essential that policymakers preserve the structural pillars of the ACA while taking
steps necessary to redress the underinvestment in the commitments it represents. Affordability of
insurance coverage remains a significant barrier for many of the remaining uninsured and some of those
already covered. Both premiums and out-of-pocket costs for the entire family unit must be considered
in combination to ensure effective access to necessary medical care. Although the ACA has made
substantial advances in this regard, we have further to go to ensure that the law meets its objectives of
providing access to adequate and affordable coverage for all Americans. Failing to do so will likely
inhibit the law from meeting its insurance coverage goals over time and will leave many low-income and
middle-income Americans with heavy health care financial burdens.

And while affordability remains a substantial barrier to coverage, one cannot overestimate the
importance of a sufficiently funded administrative structure to support the processes of enrolling
individuals in coverage and ensuring that the consumer protections promised by the ACA are
implemented effectively. Private insurance markets provide choices in cost-sharing options, provider
networks, and benefit design that many consumers value. However, these options require sufficient
numbers of well-trained assisters to ensure the health insurance programs reach the intended
populations and allow them to make effective insurance decisions; a smoothly operating IT system with
an easily managed consumer interface and an underlying set of complex functions serving government,
insurers, and assisters of different types; and an effective level of oversight and enforcement such that
competition between insurers flourishes on quality and efficiency instead of on the history of enrolling
individuals with the best possible health care risks.

It is too much to expect that a single piece of legislation could address the many challenges of our
health care system. All developed countries continue to modify their health care policies over time,
addressing issues and concerns as they are identified. The ACA has been a critical first step in improving
the US system. The proposals outlined here represent important subsequent steps that can be
implemented well within the national health expenditures originally envisioned when the ACA was
passed. The hard work of reform has begun and it has accomplished much in a short period of time, but
there is more to do.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s