The primary goals of the Affordable Care Act (ACA) were to increase the availability and affordability of health insurance coverage and thereby improve access to needed health care services. Numerous studies have overwhelmingly confirmed that the law has reduced uninsurance and improved affordability of coverage and care for millions of Americans. Not everyone believed that the ACA would lead to positive outcomes, however. Critics raised numerous concerns in the years leading up to the law's passage and full implementation, including about its consequences for national health spending, labor supply, employer health insurance markets, provider capacity, and overall population health. This article considers five frequently heard worst-case scenarios related to the ACA and provides research evidence that these fears did not come to pass.
The primary goals of the Affordable Care Act (ACA) were to increase the availability and affordability of health insurance coverage and thereby improve access to needed health care services. The law included numerous provisions aimed at achieving these goals, but the major coverage components were an expansion of Medicaid to adults with incomes at or below 138% of the federal poverty level and the introduction of federal and state-based insurance marketplaces where low- to moderate-income Americans could purchase subsidized private coverage. Among other things, the highly regulated marketplaces increased risk sharing between healthy and sick consumers and eliminated explicit price discrimination based on health status. The law also included an extension of dependent coverage to young adults ages 19–25. Each of these provisions was reinforced by a requirement that most Americans have health insurance coverage or pay a penalty (Kaiser Family Foundation 2013).
The implementation of these provisions has evolved considerably in the 10 years since the law was passed. In 2012, the Supreme Court ruled the individual mandate to be constitutional and made adoption of the Medicaid expansion optional for states. As of January 2020, 35 states and the District of Columbia have implemented the ACA Medicaid expansion (Kaiser Family Foundation 2020). Nebraska also passed a ballot initiative supporting expansion but has thus far not implemented expansion, nor is it clear that they will do so. In 2017, the Tax Cuts and Jobs Act eliminated the penalty associated with the requirement to maintain health insurance coverage and thereby effectively eliminated the individual mandate. In addition, the marketplaces have experienced multiple threats to their stability but completed their seventh open enrollment period in December 2019, with 2020 insurer participation up and premiums modestly lower, on average, than in 2019 (Holahan, Wengle, and Elmendorf, 2020).
Numerous studies of the effects of the ACA have overwhelmingly confirmed that the law has reduced uninsurance and improved affordability of coverage and care for millions of Americans (Gruber and Sommers 2019; McMorrow and Polsky 2016). Approximately 20 million people gained health insurance under the law (NCHS 2019), and there is particularly strong evidence that the Medicaid expansion has improved access to and affordability of health care services for low-income adults (Antonisse et al. 2019). Moreover, the Medicaid expansion has improved both hospital and consumer finances (Blavin 2016; Caswell and Waidmann 2019; Hu et al. 2018), and evidence continues to roll in on the law's benefits for specific subgroups of the population, such as racial and ethnic minorities (Wehby and Lyu 2018; Yue, Rasmussen, and Ponce 2018), women and new mothers (Johnston et al. 2018; Daw and Sommers 2019; Gordon et al. 2020), as well as on nonhealth outcomes such as evictions (Zewde et al. 2019).
The mounting evidence on the positive impacts of the ACA stands in stark contrast to some of the predictions from a decade ago suggesting dramatic negative consequences of the law for US health care and the economy. Critics raised numerous concerns in the years leading up to its passage and full implementation, including about its consequences for national health spending, labor supply, employer health insurance markets, provider capacity, and overall population health. Those ideologically opposed to the law voiced these concerns most frequently, but some warnings also came from less partisan sources. In this article, we consider five frequently heard early warnings related to the ACA and provide research evidence that indicates these worst fears did not come to pass. We conclude by discussing where the ACA came up short and consider the lessons learned from a decade of debate, implementation, and evaluation of the law, as well as the implications of these lessons for current debates over appropriate next steps.
Scenario 1: The ACA Will Add to the Deficit and Make National Health Spending Growth Unsustainable
Often one of the biggest points of contention for any major legislative proposal is its cost, and the ACA was no exception. While the official projections by the Congressional Budget Office (CBO) and Joint Committee on Taxation in 2010 and 2011 predicted a net decrease in the deficit as a result of the ACA, they did project increases in federal health spending of $900 billion over 10 years (Elmendorf 2011). Moreover, Centers for Medicare and Medicaid Services (CMS) projections also predicted an overall increase in national health expenditures of $311 billion over 10 years (Foster 2010).
Opponents focused on these projected increases in federal health care spending, ignoring the offsetting spending cuts and tax increases, to claim that the law would add to the federal deficit. In 2011, for example, the Heritage Foundation reported that the marketplace subsidies would “harm the economy by increasing the national deficit” and that “since many of the beneficiaries will be in the upper middle class, Obamacare's subsidies represent a reckless addition to the welfare state” (Winfree 2011).
Despite these criticisms, the ACA included a number of cost-containment provisions, including pilot programs aimed at restructuring the way providers are paid, a tax on overly generous health plans (the so-called Cadillac tax), an Independent Payment Advisory Board intended to rein in Medicare costs, and incentives for nongroup market insurers to compete for enrollees on price. Even so, many observers believed that the increased use of health care services associated with more generous insurance coverage would lead to faster health spending growth and could ultimately crowd out spending on other important services.
Even without the implementation of the Independent Payment Advisory Board or the Cadillac tax, neither of which was ultimately put in place, critics' warnings about the cost of the law were overstated. The ACA did increase federal spending, but by 2017 the CBO and Joint Committee on Taxation estimated that the increases were lower than originally anticipated. For example, at the time the law was passed, the federal cost of the insurance coverage provisions was projected to be $214 billion in 2019, and in March 2017 that projection was revised downward to $148 billion, a reduction of about one-third (CBO 2017). Moreover, national health spending projections have also fallen since the ACA was passed. In 2010, CMS projected that national health spending from 2014 to 2019 would be $23.7 trillion under the ACA (fig. 1). By 2015, however, CMS had revised its projections and estimated that national health spending from 2014 to 2019 would be $21.1 trillion, or $2.6 trillion lower than in the ACA baseline forecast (McMorrow and Holahan 2016).
While some of this reduction in projected spending can be attributed to the Supreme Court decision making the Medicaid expansion optional and lower than anticipated take-up of marketplace coverage, the broader slowdown in health spending growth that began in 2008 and continued well past the end of the recession is a more important factor. The causes of slower health spending growth over this period are not fully understood but likely include the sluggish economic recovery, the rise of high deductible health plans, and strong state Medicaid cost containment efforts. The precise contribution of the ACA to slower spending growth has not been confirmed, but there are reasons to believe it played a role (Cutler and Sahni 2017; Emanuel 2016). Thus, with the federal cost of the law coming in lower than anticipated and no evidence that the ACA led to faster national health spending growth, early fears about rising deficits and insufficient cost containment provisions in the law have largely been alleviated. Real concerns remain, however, about the efficiency of spending in the health care system more generally.
Scenario 2: The ACA Will Suppress Labor Supply
Another widely held concern about the law surrounded its potential effects on the labor market. There were several provisions of the ACA that economists believed could suppress the labor supply. For example, the income eligibility thresholds to qualify for Medicaid expansion and marketplace subsidies create a potential scenario where working less could result in a net benefit for individuals if they became eligible for one of these subsidized programs. Moreover, the employer mandate created incentives for employers to shift their workforce toward greater use of part-time employees, and other tax increases under the law had the potential to suppress productivity.
Early estimates from CBO projected a loss of approximately 800,000 jobs under the ACA, mostly a result of workers choosing to supply less labor, and updated estimates in 2014 nearly tripled that number (CBO 2014b). This became one of the rallying cries of those opposed to the law. A Forbes article published in February 2014 concluded: “The bottom line is that the ACA will result in the equivalent of 2.9 million or more fewer working Americans. No amount of hand-waving by the law's proponents can avoid this inconvenient truth” (Conover 2014).
Since then, numerous studies have examined the effects of the dependent coverage expansion and the Medicaid expansion on employment, hours worked, wages, and earnings and have found no evidence to support the predictions described above (Abraham and Royalty 2017; Garrett, Kaestner, and Gangopadhyaya 2017; Kaestner et al. 2017). At least two studies have also considered the effects of marketplace subsidies on labor supply and have found no significant effects (Cucko, Rinz, and Solow 2017; Duggan, Goda, and Jackson 2019). Furthermore, studies examining the potential shift to part-time employment or a trend toward early retirement under the ACA found no evidence to support these claims (Levy, Buchmueller, and Nikpay 2018; Moriya, Selden, and Simon 2016). Overall, the evidence is remarkably consistent across studies looking at the various policy mechanisms and using different outcomes and data sources. While the economic theory supporting a concern about labor supply effects was sound, the evidence suggests that the assumptions used to predict the magnitude of the effects did not adequately reflect the full range of factors beyond health insurance that affect employer and employee preferences.
Scenario 3: The ACA Will Destroy the Employer Health Insurance Market
The components of the law that strengthened and subsidized the private nongroup insurance market contributed to fears about the future of the employer-sponsored health insurance market. When the law was passed in 2010, approximately 57% of nonelderly Americans received health insurance from their employer or that of a family member (fig. 2). The subsidized insurance marketplaces under the ACA presented an alternative to employer coverage and created potential incentives for employers to stop offering coverage to their employees. These incentives were strongest for employers of lower-wage workers who would be eligible for Medicaid or the most generous premium and cost-sharing subsidies under the law. If many employers were to stop offering insurance to their workers, many people would find their source of coverage disrupted and the government cost of the law could increase due to larger numbers of workers and their dependents enrolling in federally subsidized marketplace coverage or Medicaid. As a result, the ACA included a provision to encourage employers to continue to offer coverage. Employers with more than 50 full-time-equivalent employees would pay a $2,000 penalty per worker if they did not offer coverage and one or more employees received a premium subsidy in the marketplace (there is no penalty for workers enrolling in Medicaid). In addition, the ACA did not alter the significant tax advantage of employer-sponsored insurance that predated the law.
The most dramatic predictions suggested extremely widespread dropping of employer coverage. A study by the American Action Forum predicted that as many as 35 million Americans would lose their employer-sponsored coverage, with an associated increase in federal premium subsidies of $1.4 trillion (Holtz-Eakin and Smith 2010). In July 2013, leaders of three major unions seized on this message and sent a letter to Congress claiming that the law's incentives were causing “nightmare scenarios” and would “destroy the very health and wellbeing of our members along with millions of other hardworking Americans” (Gara 2013). Official projections from the CBO (2014a) were much more modest but still predicted employer coverage losses of approximately 6 million people by 2016. In contrast, in 2012, Urban Institute researchers estimated that employer-sponsored insurance would increase modestly under the ACA (by less than 3%), due in large part to the individual mandate and persistent tax incentives (Blumberg et al. 2012).
Far from the nightmare scenario predictions, both early and sustained evidence suggests that the employer-sponsored insurance market has remained large and strong under the ACA. Blavin et al. (2015) found no evidence of declines in employer offer, take-up, or coverage rates through 2014, and consistent evidence since then has found stable if not rising rates of employer coverage (Shartzer, Blavin, and Holahan 2018). By 2018, the rate of employer coverage among the nonelderly population was 58.1% compared to 56.6% in 2010 (fig. 2). A strong economy, the law's individual mandate, the continued tax advantage of employer coverage, and turbulence in the marketplaces have likely contributed to the strength and stability of employer coverage in recent years, but there is certainly no evidence to support widespread dropping of coverage under the law.
In addition to fears of employers dropping their health insurance plans, there were also concerns that the Medicaid expansion would cause newly eligible individuals to drop their employer coverage in favor of the more affordable public plan. This “crowd-out” of existing employer coverage is a cause for concern to some because, were it to occur, the federal cost of the law would increase without covering additional individuals. There would be value, however, to low-income people obtaining very comprehensive coverage at little to no cost to the household, since it would improve their health insurance affordability and access to care, potentially improving their downstream health outcomes as well. There is some evidence of displacement of private insurance under the ACA's Medicaid eligibility expansion, but the magnitude of the estimates varies widely, and the results are often imprecise. The smallest estimates indicate virtually no evidence of crowd-out (Frean, Gruber, and Sommers 2017), while the largest indicate significant crowd-out in 2015 that then declined over time (Miller and Wherry 2019). Thus, with no evidence of net declines in employer coverage under the ACA and modest evidence of crowd-out due to the Medicaid expansion, it is clear that the employer-sponsored insurance market remains strong, despite the critics' worst fears.
Scenario 4: Provider Capacity Will Be Insufficient to Meet Demand under the ACA
Despite criticisms on many other fronts, most observers seemed to agree that the ACA would succeed in its effort to increase the number of insured Americans. There was considerable disagreement, however, on whether that coverage would actually provide access to care or improved health. In particular, there was a great deal of concern that the supply and distribution of health care providers would be insufficient to meet the increased demand created by the newly insured. A 2013 Forbes article, for example, warned that “America is suffering from a doctor shortage. An influx of millions of new patients into the healthcare system will only exacerbate that shortage—driving up the demand for care without doing anything about its supply” (Pipes 2013). Also in 2013, Joseph Antos argued in this journal that this issue would be particularly problematic for the Medicaid population. Low provider payment rates have historically kept many providers from accepting Medicaid patients, so “putting millions of additional people into a program that has been struggling with access to care for the past forty-five years is likely to result in worsening access for those who are currently enrolled in Medicaid” (Antos 2013).
The ACA included several provisions aimed at shoring up provider access, including a temporary increase in Medicaid fees and increased federal funding for community health centers. A study using a “secret shopper” analysis in 10 states found increased appointment availability for Medicaid patients and stable appointment availability for privately insured patients following the ACA (Polsky et al. 2017). The researchers in this study posed as Medicaid or privately insured patients new to the practice at which they sought an appointment but did not provide further details about their coverage. Moreover, when considering the potential spillover effects of the eligibility expansion on those who were insured prior to the ACA, Carey, Miller, and Wherry (2018) found that Medicare enrollees did not experience any adverse effects of the ACA Medicaid expansion on their access to care. To our knowledge, there is currently no evidence on how the expansion affected access for those who were already enrolled in Medicaid prior to the expansion.
A few studies have found evidence consistent with provider capacity constraints. The same secret shopper study that found increased appointment availability overall found that wait times increased for both Medicaid and privately insured patients (Polsky et al. 2017). Two studies using data from the National Health Interview Survey also found evidence among low-income adults that the Medicaid expansion increased reports of delaying care due to wait time for an appointment and problems finding a provider who could see them (Miller and Wherry 2017; Selden, Lipton, and Decker 2017). These access problems were not trivial in magnitude through 2015, with the share reporting care delays due to appointment wait times increasing by almost one-third (Miller and Wherry 2017), but when data through 2017 were added these access problems had diminished (Miller and Wherry 2019). Moreover, considerable evidence has shown that the ACA Medicaid expansion had a positive effect on having a usual source of care and use of preventive services so any delays or wait times due to capacity constraints did not appear to prevent the law from improving access to and utilization of care for those it targeted (Antonisse et al. 2019; Simon, Soni, and Cawley 2017; Sommers et al. 2015).
Scenario 5: The ACA Will Not Improve, and May Actually Be Harmful to, Population Health
At the end of the day, insurance coverage and even access to and use of health care services do not guarantee good health outcomes, and concerns about the ACA's impact on population health were widespread. Those generally supportive of the law feared that, despite its best intentions, the ACA would not actually move the needle on health outcomes. Some of these concerns were reinforced by the lack of clinical improvement found in a study of the Oregon Health Insurance Experiment (Baicker et al. 2013), as well as growing evidence on the social determinants of health. Recognizing that there are many and varied contributors to good health, even supporters believed that the ACA coverage expansion was necessary but potentially not sufficient to meaningfully improve health status (McMorrow 2010).
In contrast, some of the law's fiercest critics believed that the law would actually harm people's health. Perhaps the most extreme example was Sarah Palin's claim that the law would effectively institute “death panels,” or groups of bureaucrats tasked with deciding who was worthy of care (Kessler 2012). But the more insidious claims focused on Medicaid and misinterpreted research evidence to claim that Medicaid was worse than no coverage at all (Carroll and Frakt 2017). Unlike Palin's obviously sensationalist comments, these claims have been made by respected scholars and were used to argue against the ACA more generally (Gottlieb 2011), and then to argue against expanding Medicaid once the Supreme Court made the expansion optional (Antos 2013).
Thus far, the evidence for positive health effects of the ACA Medicaid expansion is weaker than that for coverage or affordability of care, but several important findings have emerged (Allen and Sommers 2019). Several early studies found small or no effects on self-reported general or mental health status (Courtemanche et al. 2018; Miller and Wherry 2017), while others have found improvements in self-reported health and reductions in psychological distress (McMorrow et al. 2017; Simon, Soni, and Cawley 2017). With respect to improved clinical outcomes, researchers found improved blood pressure control among community health center patients, but no improvement for diabetes (Cole et al. 2017). There is also recent evidence that Medicaid expansion improved surgical outcomes for several common conditions, seemingly driven by earlier presentation for care (Loehrer et al. 2018).
When considering mortality, perhaps the ultimate health outcome, there is emerging evidence that the ACA has, in fact, saved lives. Recent work using survey data linked with death records has attributed a 9% reduction in mortality among low-income adults to the Medicaid expansion (Miller et al. 2019), and another study capitalizing on an Internal Revenue Service experiment found that coverage gains for middle-aged adults under the ACA reduced their mortality (Goldin, Lurie, and McCubbin 2019). Thus, despite somewhat inconsistent evidence across a variety of populations and health outcomes, there is little or no evidence to support the claim that the ACA Medicaid expansion has harmed health, and the weight of the evidence appears to indicate health improvements.
We have argued above that the ACA failed to live up to its critics' worst fears when it came to cost, labor market effects, the demise of employer coverage, provider capacity constraints, and population health. We must also acknowledge, however, where the ACA underperformed the expectations set by some of its advocates. First, a commonly heard argument for expanding coverage was that it would actually save money by reducing reliance on emergency departments (ED) and improving access to preventive care. While insurance coverage does generally increase use of preventive services, it does not necessarily reduce use of the ED or reduce costs (Russell 2010). Like other services, ED visits typically become less expensive to the consumer after they gain insurance, and when a service becomes less expensive, people tend to use more of it. This pattern was seen following the Oregon health insurance expansion (Taubman et al. 2014), and there is evidence that ED use increased under the ACA Medicaid expansion as well (Garthwaite et al. 2019; Nikpay et al. 2017). This should not be interpreted as a failure of the ACA but, rather, a failure of those promoting an unrealistic outcome.
Second, and more important, despite the ACA's many successes, coverage and care remain unaffordable for too many Americans, particularly many of those who rely on the individual health insurance market. The ACA's marketplaces were intended to provide an affordable option for individuals and families with low to moderate incomes and no access to employer or public coverage. For many marketplace enrollees, this option has improved access to and affordability of care (Kirby and Vistnes 2016; McMorrow et al. 2016). But for those eligible for small premium subsidies or none at all, coverage in the marketplace can be quite expensive (Holahan, Blumberg, and Wengle 2017). Moreover, when individuals seek lower premiums to increase affordability, they generally face high deductibles and other cost-sharing requirements (Gunja et al. 2016). The fundamental driver of high insurance costs is, of course, the high cost of health care itself, so truly delivering on the promise to make health care affordable for all Americans will likely require tackling this head on, as well as funding additional subsidies.
As the 2020 presidential campaign heats up and we begin to debate the merits of various health reform proposals, we should take at least two lessons from over 10 years of ACA debate, implementation, and evaluation. First, one should not take individual projections of costs or other outcomes too literally but, rather, consider the range of estimates and potential outcomes. As we have seen with the ACA, not all predictions will be realized, so it is important not to focus too much attention on any one predicted benefit or cost of a specific proposal. Second, while predictions and projections are an important part of the process of developing and implementing reform proposals, the only way to truly know the effects of a particular policy is to wait for the evidence. And on that front, researchers have delivered a tremendous body of work that should allow anyone interested to form a robust and nuanced opinion of the ACA's successes and failures, and with time, we should expect the same of any future reforms.
The authors are grateful to all of the researchers who have contributed to such a strong evidence base on the effects of the Affordable Care Act, and to Caroline Elmendorf for excellent research assistance. This work was supported by the Robert Wood Johnson Foundation. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funder.