Design

The ACA aimed to achieve bold goals through existing financing and delivery structures, and is neither as narrow nor as disruptive as often described. Much of the debate over the ACA has been laser focused on its coverage expansion and, even more specifically, its Health Insurance Marketplace (also called Exchanges). The law, however, includes numerous policies to lower health care costs for people already insured as well as improve the quality of coverage and care.^{Reference Obama1} Its provisions, among others, include expanding prevention and public health initiatives; promoting value-based purchasing; reducing fraud and abuse; strengthening clinic and workforce capacity; and reducing prescription drug prices. The cost of its new investments has been more than offset through reduced federal spending and new revenue.² Its significant and sweeping changes make it the most significant health legislation since Medicare and Medicaid were enacted.^{Reference Iglehart3}

In addition to not being as narrow as the debate suggests, it is neither as radical. The ACA is not a “government take-over” of the health system. Its authors went out of their way to sustain employer-based coverage. It includes a firewall to prevent people with access to affordable employer-based coverage from receiving premium tax credits in the Marketplace. Its employer responsibility provision (i.e., mandate) is designed to prevent employers from dropping their coverage due to federal financial assistance in the Marketplace. And it aims to lower private health care cost growth through policies such as a new pathway for approval of low-cost biosimilar drugs. Indeed, the fraction of non-elderly Americans with employer coverage remained steady between 2013 and 2016.^{Reference Barnett and Berchick4}

Similarly, instead of replacing Medicaid and the Children's Health Insurance Program (CHIP), the ACA extends them with relatively minor modifications. It offers generous federal funding in Medicaid to create a nationwide floor of eligibility for adults with income below 138 percent of the federal poverty level. It also extended and increased federal support for CHIP for 5 years (with subsequent legislation to reauthorize the program through 2027). People enrolled in these programs prior to and after the ACA's implementation experienced few changes in their coverage.

The coverage that was most profoundly changed was that in the individual health insurance market. Prior to the ACA, relatively few federal or state-level rules governed individual market plans.^{Reference Hushagen and Fish-Parcham5} This allowed insurers to limit their exposure to high-cost and high-risk people through medical underwriting: screening applicants for evidence of risk of a health problem or pre-existing condition and using that information to carve out related benefits, set higher premiums, or deny coverage altogether. They also would, where allowed by state policy, categorically exclude benefits that might attract high-cost applicants. Before 2014, 62 percent of individual market enrollees lacked maternity coverage, 34 percent lacked substance use disorder treatment coverage, 18 percent lacked coverage for mental health care, and 9 percent lacked prescription drug coverage.⁶ Another common insurer practice was to include annual and lifetime limits on payments. Some states attempted over the years to reform this market, but doing so without sufficient enrollment led to rising premiums which drove out an increasing number of low-cost enrollees, creating a “death spiral” in which there is insufficient pooling of costs to make the insurance market work.^{Reference Kirk7}

The ACA set out to improve the functioning of the individual market by making insurance higher quality, available to anyone, and more affordable. First, it sets federal minimum insurance reforms. It bans the discriminatory insurance practices that block enrollment of people with health needs and ensures ongoing access to coverage. It also sets standards for the quality of coverage, such as the requirement to cover ten “essential health benefit” and to calibrate enroll-ees' share of total costs to pre-set tiers: “bronze” plans charging enrollees the most and “platinum” plans charging them the least. Second, it provides financial assistance to lower the premiums and cost sharing in Marketplace plans to encourage uninsured low-income people to sign up. Premium tax credits limit the percent of income an enrollee pays for the second-lowest silver plan in an area for qualified individuals with income between 100 and 400 percent of the federal poverty level. Cost sharing “reductions” lower co-pays and deductibles for enrollees in Marketplace plans with income between 100 and 250 percent of the federal poverty level. Third, the ACA specified that if a person can afford coverage, that person must either buy it or pay a tax. The intent of this “individual mandate” is to prevent people from waiting until they need health care to buy insurance, spreading out the costs of care, both over time and at a point in time, across those who do and do not need health care.

The ACA overlaid on this foundation — insurance reforms, subsidies, and the mandate — policies to promote choice of qualified health plans to increase satisfaction and lower costs. The word “Marketplace” is often considered synonymous with a website. Thirty-nine states use HealthCare.gov, while 11 states and the District of Columbia run their own. Tools on HealthCare.gov and most state websites allow plan comparisons by features such as providers included in each plan's network and potential out-of-pocket costs. The law also requires in-person assistance (e.g., Navigators) and transparency (e.g., summaries of benefits and coverage) to help match enrollees with health plans.

These consumer-facing policies to drive competition based on quality and price — rather than on the prior practice of avoiding high-cost enrollees — have been aided by behind-the-scenes premium stabilization programs, sometimes called the “three Rs”: the risk-adjustment, reinsurance, and risk-corridor programs. Risk adjustment shifts funding from plans with lower-risk to higher-risk enrollees, limiting the incentive for insurers to seek only healthy enrollees. Reinsurance pays for relatively rare, expensive claims, spreading their cost out over a broad base through premiums or general revenue. And, risk corridors protect plans from uncertainty in rate setting by having the federal government share in the risks of losses and gains. While such programs are permanent features of the Medicare Part D drug program, only the Marketplace's risk-adjustment program continued after the first three years of its operation.

The ACA's stitching together of coverage systems — bolstering employer-based coverage, making Medicaid a safety net for all low-income Americans, and reforming the individual market — required a complex set of policies. It would have been simpler technically to start from scratch or replace rather than interlace existing systems. Yet, health reform over the years has been thwarted by the fear of change and the formidable status quo. This was a factor in the failure of the Clinton health reform effort in 1994.^{Reference Oberlander8} Even the ACA's changes to the individual market, that caused insurers to cancel a relatively small 1.9 million plans that would have been out of compliance as they created new ones, led PolitiFact to name the President's claim, “You like your plan, you can keep it,” the lie of the year in 2013.^{Reference Gabel9} The U.S. health system is the largest in the world in terms of spending, offers eight of the top-ten best-paid occupations, and is a major employer — all of which fortify resistance to change.^{Reference Lambrew and Montz10} Anticipation of such challenges led the authors of the ACA to value minimal disruption above simplicity.

Implementation and Maintenance of the Marketplaces

The legislation that authorized the ACA set the time-line and parameters for implementation, but delegated its details to the executive branch. Its design assumed partnerships for implementation as well as ongoing modifications to calibrate its components, although, as discussed below, this did not play out as anticipated. Specifically, it created key roles for: states in setting up Marketplaces and enforcing the ACA's insurance reform; employers in adapting their plans to meet new standards; and private insurers in adopting new policies to compete for enrollees — to name a few. It also demanded detailed regulations that required regular titration based on experience. From enactment to the end of his term, President Obama implemented the law while being open to changes to improve the system's performance.¹¹

Yet, ongoing opposition to the law affected implementation. After 48 states applied for planning grants in 2010, only 11 states ultimately took full ownership of their Marketplaces. Four states choose not to enforce the ACA's insurance reforms, leaving the Centers for Medicare and Medicaid Services do so. On the day it was signed into law, litigation was launched that led to the ACA's constitutionality being decided by the Supreme Court in 2012. While the Supreme Court upheld the individual mandate, it ruled that the ACA could not require states to expand Medicaid; to this day, 17 states have turned down the significant federal financing associated with expanding this program — which raises Marketplace premiums.^{Reference Sen and DeLeire12} The resistance to making even technical corrections to the ACA contributed to another Supreme Court case (King v. Burwell) and a lawsuit over cost sharing subsidies (House v. Azar). An appropriations rider enacted first in 2014 prevented payments that exceeded collections in the risk corridor program, depriving health plans of $12.3 billion.^{Reference Livingston13} And the October 2013 government shutdown was largely attributed to Republican opposition to implementation of the ACA.^{Reference Weisman and Peters14} These actions were not always based on principles: for instance, the Republican House sued the Obama Administration over its delayed implementation of the employer mandate despite its own legislation to repeal such a mandate. As such, the assumption that, once passed, the implementation of the ACA would become non-partisan, as happened with the Medicare drug benefit and numerous other programs, proved (with some exceptions) wrong.

This hostility toward the ACA has colored the Trump Administration's management of the Marketplace. Despite early claims of support for lowering premiums and preserving protections for people with preexisting conditions, President Trump's actions suggest his intent of making “implosion” of the ACA a self-fulfilling prophesy.^{Reference Fabian15} On his first day in office, he signed an executive order (E.O. 13765) to “take all actions consistent with the law to minimize the unwarranted economic and regulatory burdens of the Act, and prepare to afford the States more flexibility and control to create a more free and open healthcare market.” In October 2017, he stopped making $9 billion a year in cost sharing reduction payments to insurers despite insurers' ongoing obligation to lower cost sharing for eligible consumers.^{Reference Pear, Haberman and Abelson16} His administration cut spending on education and outreach about existing coverage options, reduced in-person assistance support, reduced tax credits, and repeatedly and vocally criticized the law which could discourage enrollment.^{Reference Park17} In part responsive to a second executive order (E.O. 13813), the administration proposed and finalized rules for 2019 and beyond that would expand access to types of health plans (i.e., association health plans, short-term limited duration plans) that are subject to lower regulatory standards than other individual and small group market plans, reduce federal essential health benefit standards, and scale back requirements for Navigators (one type of in-person consumer assister).^{Reference Jost18} Significantly, in its tax cut legislation (Public Law No. 115-97), the Congress zeroed out the penalty associated with the individual mandate. This came after a year-long effort to repeal the law's mandates, taxes, Medicaid expansion, and Marketplace structure and partly replace them in various ways. The impact of these actions has begun to show in the Marketplace, as described below.

Effects on Coverage, Access, and Satisfaction

Evidence suggests that the ACA's goals of improving access, quality, and affordability have largely been met, despite aforementioned challenges. An estimated 20.0 million adults gained health insurance as a result of all of the ACA's coverage provisions, with the uninsured rate dropping by 43 percent from October 2013 to early 2016.^{Reference Uberoi, Finegold and Gee19} Access on a number of measures has improved; for example, between September 2013 and March 2015, an estimated 9.4 million fewer adults had problems paying family medical bills.^{Reference Karpman and Long20} Similarly, quality of care has also improved on some key measures. In part thanks to the ACA's payment incentives and its Partnership for Patients initiative, an estimated 125,000 fewer patients died in the hospital as a result of hospital-acquired conditions in 2015 compared to 2010, saving approximately $28 billion in health care costs over this period.²¹ Since the law was enacted, health care cost growth has also slowed, as evidenced in lower national health expenditures, spending per enrollee, and prices.²² While other factors contributed to this slowdown, most experts agree that the ACA was a catalyst.

The Health Insurance Marketplace contributed an estimated 40 to 50 percent of the coverage gain explained by the ACA.^{Reference Frean, Gruber and Sommers23} After a rocky start in 2013, enrollment, as measured by people signed up during open enrollment, grew from 8.0 million in 2014 to a peak of 12.7 million in 2016, falling to an estimated 11.8 million for 2018.²⁴ The uninsured rate dropped by more than 30 percent in all income groups, including the Marketplace's target population of middle- and high-income uninsured.²⁵ The law's provisions also opened up individual insurance to people previously denied coverage: the number of people with preexisting conditions insured through the individual market rose by 64 percent between 2010 and 2014 — although, even with this increase, the share of people with pre-existing conditions in the individual market (4.5 percent) was less the share of all non-elderly people in this market (7 percent).²⁶

These coverage gains were accompanied by increases in access and satisfaction with health coverage and care. The law has reduced “job lock” or staying with a suboptimal job to keep health benefits; an estimated 1.5 million more people became self-employed because of the ACA's individual market reforms and financial assistance.^{Reference Blumberg, Corlette and Lucia27} The percent of people seeking individual market coverage that found it very difficult or impossible to find affordable care fell from 60 percent in 2010 to 34 percent in 2016.^{Reference Collins, Gunja, Doty and Beutel28} People newly enrolled in a Marketplace plan reported a significant increase in having a personal health care provider (from 59 to 73 percent) and receiving a flu shot in the previous year (from 41 to 52 percent).^{Reference Schmittdiel, Barrow, Wiley, Ma, Sam, Chau and Shetterly29} Satisfaction among enrollees was roughly the same in Marketplace plans as employer plans in 2016.³⁰

Impact on Affordability

The Marketplace's record on affordability is more complicated. For enrollees with income below 400 percent of the federal poverty level who receive premium tax credits, affordability has remained high from the start — by design. Eligible individuals who choose the second-lowest cost silver plan pay a statutorily set amount, regardless of the unsubsidized premium for that plan. During the 2018 Marketplace open enrollment period, as in previous years, 83 percent of enroll-ees qualified for premium tax credits. The average premium tax credit was 75 percent of the average premium among all HealthCare.gov enrollees in 2018.³¹ Taking into account these tax credits, 80 percent of enrollees in 2018 had a choice of a health plan costing less than $75 per month.³² In 2016, most (70 percent) of Marketplace enrollees reported no difficulty paying out-of-pocket costs in the previous year, slightly lower than those in employer plans (75 percent).³³

In contrast, the affordability of unsubsidized coverage in the Marketplace has been variable. Premium growth and levels are lower in areas with competing insurance companies, consistent with expectations.^{Reference Sheingold, Nyguyen and Chappel34} Additionally, premium growth has differed in three periods of time. First, during 2014 to 2016, both premium levels and growth were relatively (and, as discussed below, potentially too) low. The actual benchmark premium in 2014 was $900 or nearly 20 percent below what had been projected at the time of passage.^{Reference Glied, Arora and Solís-Román35} Premiums for the benchmark plan increased by a relatively modest 3 percent in 2015 and 8 percent in 2016.³⁶ One study found that the Marketplace benchmark premium in 2016 was 10 percent lower than the average employer sponsored insurance premium.^{Reference Blumberg, Holahan and Wengle37}

Second, policy changes and one-time adjustments drove up premiums for 2017. The average premium for the benchmark plan rose by 24 percent in HealthCare.gov states.³⁸ One clear cause was the end of the temporary reinsurance program, which added an estimated 4 to 7 percent to premiums.³⁹ Additionally, insurers implemented corrections to their pricing based on their experience in the Marketplace. As predicted and confirmed by the Council of Economic Advisers, insurers generally underpriced their plans relative to their enrollees' health risks initially and, with rate corrections in 2017, returned to pre-ACA levels of profitability.⁴⁰

Third, actions taken by the Trump Administration, previously described, have significantly increased individual market premiums. Evidence increasingly shows that premiums would be increasing in single digits^{Reference Fiedler41} and perhaps even falling in some states. However, because of policy actions, the Marketplace benchmark premium rose by 37 percent between 2017 and 2018.⁴² About 19 percent of this can be attributed to the administration's decision to stop federal payments for reduced cost sharing.⁴³ Insurers also raised rates for 2018 in anticipation of cuts in marketing and lack of enforcement of the individual mandate, both of which discourage enrollment of relatively healthy people. In 2019, premiums are expected to climb by an estimated 10 percent due to fewer healthy enrollees due to the tax bill's zeroing out the individual mandate penalty.⁴⁵ Additionally, the administration's proposal to expand short-term limited duration plans, which can offer lower premiums to healthier people due to medical underwriting and reduced benefits, is projected to drive up premiums for plans that abide by the ACA rules because their enrollees will disproportionately be people with higher average health care spending.^{Reference Blumberg, Buettgens and Wang46}

These premium increases reduced Marketplace enrollment by about 4 percent over the previous year in both 2017 and 2018 — which is much less than expected because of the premium tax credit design. When unsubsidized premiums rise in the Marketplace so too do the premium tax credits. This design feature was used in 2018 by insurers and insurance commissioners to offset the loss of federal payments for cost sharing reductions, a practice called “silver loading.” Rather than spread out the unsubsidized cost of reduced cost sharing across all types of plans, they added it to silver-plan premiums — and in some states, only those silver plans offered through the Marketplace. This resulted in higher premium increases in silver than bronze and gold plans (32, 17, and 18 percent respectively) and premium tax credits rose even faster: by 45 percent.^{Reference Semanskee, Claxton and Levitt47} This higher premium tax credits improved affordability for low-income enrollees who are not otherwise eligible for cost sharing reductions or who could choose a non-silver plan using their higher tax credit for less (e.g., a gold plan with a lower deductible). CBO projects that, on net, 1 million more people may gain coverage as a result of silver-loading.⁴⁸ If Congress does not restore cost sharing reduction payments and the Trump Administration does not limit silver loading, higher premium tax credits will continue to be offered in, and sustain, the Marketplace.

States are also in a position to counter some if not all of the actions taken by Congress and the administration that could increase premiums. For 2018, premiums were 17 percent lower in states that run their own Marketplace compared to those that use HealthCare.gov — greater than the differential in previous years.^{Reference Hall and McCue49} State-based Marketplaces conduct their own marketing, set their own open enrollment periods (most of which were longer than that of HealthCare.gov in 2018), and work with insurers to manage rates. A number of states have also adopted policies to counteract those occurring at the federal level, such as limiting unregulated short-term plans.^{Reference Lucia, Giovannelli, Corlette, Volk, Palanker, Kona and Curran50} In fact, Massachusetts, which has long-standing policies that supersede all the recent changes, had the second-lowest benchmark plan premiums of all states in 2017 and 2018, even though its health care costs are higher than the national average. Some states like Idaho and Iowa, in contrast, are following the administration's lead in seeking to expand access to less regulated plans which would lower premiums for healthy enrollees but would raise premiums for people with pre-existing conditions left behind in the regulated market.^{Reference Goldstein51}

Premiums and access to plans in 2019 may also be better than expected due to overcorrections for prices in 2017 and 2018. In 2017 and the first quarter of 2018, insurers' financial results were better than they were in any year from 2011 to 2016.^{Reference Semanskee, Cox and Levitt52} The Council of Economic Advisers similarly has documented profits for insurers in the Marketplace.⁵³ North Carolina Blue Cross Blue Shield and with Health Care Services Corporation, the Blues plan in Illinois, Montana, New Mexico, Oklahoma, and Texas, suggest premiums for 2019 would not significantly increase but for additional policy changes.^{Reference Keith54} This suggests that Marketplace and insurer adaptation are counteracting premium increases due to recent policy fluctuations. Early evidence suggests insurers may expand their offerings in the Marketplace in 2019 for the first time in several years.^{Reference Luhby55}

Do Nothing

Policy makers (including the administration) could decide to leave Marketplace policy as is, taking a “hands off” approach. Opponents of the law assume its inherent flaws would erode its performance, causing it to ultimately fail. Proponents of the law assume this strategy would prove the Marketplaces are viable, supporting its comparative advantages over proposed alternatives. This could, alternatively, be a placeholder posture for assessment or political purposes. Analytically, what is a “normal” and steady-state Marketplace is hard to discern given rapid cycles of change since its start, which complicates efforts to change it. Politically, doing nothing allows policy officials time to explore options or wait for a more propitious balance of power.

Fix It

Policy makers could also endeavor to fix the Marketplace. A number of legislative proposals have been introduced (and enacted in some states) to maintain and build upon its structure. Bipartisan support is widespread for providing reinsurance for the Marketplace to limit the risk of extremely expensive claims (e.g., Collins (R-ME)-Nelson (D-FL) bill, S. 1835; Costello (R-PA)-Peterson (D-MN) bill, H.R. 4666). Reinsurance programs have proven effective in Medicare Part D, the early years of the Marketplace, and states like Alaska, Minnesota, and Oregon that implemented them through State Innovation Waivers. Streamlining the State Innovation Waiver process while maintaining its “guardrails” for approval was included in the bipartisan Alexander (R-TN)-Murray (D-WA) bill.^{Reference Alexander and Murray57} This bill also included a “trade”: liberals' increased funding for outreach in return for conservatives' expanded access to “catastrophic plans.” Some proposals have attracted support only from their own parties such as the bills to expand versus limit access to unregulated short-term plans (Barrasso (R-WY) bill, S. 2507, and Baldwin (D-WI) bill, S. 2494). Historically, bipartisan incremental proposals have been the vehicle through which long-established public programs like Medicare and Medicaid have been modernized.

Because of the underlying complexity and history of the ACA, successfully fixing it may be more difficult than doing nothing or replacing the Marketplace. On policy, this was exemplified by challenge related to federal payments for cost sharing reductions. In most of 2017, a bipartisan coalition of policy makers supported restoring such payments, fearful of the consequences should they stop. Yet, after they stopped and insurers used silver loading to adapt to change, CBO explained that restoring such funding would result in low-income enrollees paying more, high-income enrollees paying less, and more people becoming uninsured.⁵⁸ This contributed to the coalition breaking apart.

The politics as well as the policy of fixing the ACA's Marketplace are fraught. Conservatives who have long sought to eradicate the law are often vehemently against propping it up. Liberals are loath to use additional capital on a system from which private insurance companies have reaped increasingly rich profits. There is not a nature political constituency for Governor Mitt Romney's Marketplace brought to scale by President Obama.

Replace It

The third approach to the Marketplace is to replace it altogether. Efforts to do so, with the background chant that the ACA was “imploding” or “exploding,” dominated the Congressional debate in 2017. The Republican House-passed American Health Care Act (AHCA) would have, among other provisions, replaced the ACA's premium-based tax credits with flat-dollar tax credits which were delinked from premiums and set to increase at rates below medical cost growth. CBO estimated that this bill would result in premiums that are lower for some but higher for others, fewer people insured in the individual market, and unstable markets in areas where one-sixth of Americans live.⁵⁹ While the Senate version that failed to pass in July 2018 was more incremental than the AHCA, its successor, the Graham-Cassidy bill (also called the Graham-Cassidy-Johnson-Heller amendment), was more radical in its approach to the Marketplace. This proposal would replace premium tax credits with a combined Marketplace-Medicaid federal funding block grant for states. States could reduce or eliminate benefit requirements, rating rules (e.g., prohibition on medical underwriting and gender rating), and other ACA requirements for private plans receiving block grant funding. CBO assumed that states would choose to spend little of their block grant to shore up the individual market which would have significantly fewer enrollees.⁶⁰ The Trump administration included this proposal in its fiscal year 2019 budget, which means it could be the Republicans' health plan for 2020.^{Reference Lambrew61}

On the same day as Graham-Cassidy bill was released, Senator Sanders (I-VT) and 16 co-sponsors introduced the Medicare-for-All Act (S. 1804) that would entirely replace the Marketplace with Medicare. Rather than financing benefits through premiums, this proposal would do so through taxes (although the bill itself does not include a specific tax rate or method). Proponents claim people will pay less for this coverage than Marketplace coverage because of its statutory limits on provider payment rates and lower administrative costs. Whether those savings are sufficient to fully offset the increased generosity of coverage (e.g., no deductibles, no co-pays for most services) is unknown.

Such replacement proposals have the advantage of ideological purity. The Graham-Cassidy plan to limit federal entitlement spending and return decisions to states resonates with core conservative values. The Medicare-for-All proposal's emphasis on equity, simplicity, and ending a reliance on profit-oriented insurance companies attracts liberals. It is easier for replacement than incremental proposals to generate popular support given their clean lines and bold — if often untested — assertions about results. They also avoid the difficulty of legislating second-best proposals: they offer a north star to guide health system change.

Yet, as previously described, change is hard — not just from the perspective of stakeholders and people whose fears of change may exceed their hopes for it. They pose real-world implementation challenges, especially given the large share of jobs and the economy devoted to heath care. Such proposals are also most vulnerable to demonization and caricatures. A Tom Toles cartoon illustrated the Graham-Cassidy bill as an elephant pointing a patient to a chopping block, saying, “My plan is for you to place your neck here, on our block grants.” An often-used cliché for government-run health bills is that they have the efficiency of the Post Office with the compassion of the Internal Revenue Service (IRS). The leap of faith inherent in bold health reform plans allows for speculation about their results, which may both attract supporters and be hard to refute.