Enrollment through HealthCare.gov continues to climb during the COVID-19 special enrollment period, bolstered by enhanced marketplace subsidies under the American Rescue Plan Act (ARP) through 2022. Millions of lower- and middle-income Americans are eligible for financial help under the ARP, including many who did not previously qualify based on income. The ARP did not, however, change other subsidy eligibility rules, or federal interpretations of those rules. As a result, gaps remain in who can access subsidized marketplace coverage, leaving some without relief even with the changes made by the ARP.
This post focuses on one of those long-standing gaps known as the “family glitch”; this renders not only an employee but also his or her family members ineligible for subsidized marketplace coverage if self-only coverage for the employee is affordable as defined by the ACA, even if family coverage is not. An estimated 5.1 million people—mostly children of low-income workers—fall into the family glitch and are thus barred from enrolling in subsidized marketplace coverage. Many of these families go on to purchase health insurance (either through a family member’s job or the marketplace) but pay high portions of their income towards premiums. Personal stories of those affected by the family glitch can be found in media outlets, Reddit threads, and even comments from readers on prior Health Affairs Blog posts.
Critics have argued that the family glitch is inconsistent with the goals of the Affordable Care Act (ACA) and unfairly penalizes family members of lower-income workers. There is also broad support among a range of health care stakeholders to fix the family glitch. An analysis for The Commonwealth Fund of recommendations from diverse health care stakeholders to the Biden-Harris presidential transition team showed that a majority recommended fixing the family glitch. And legal experts such as Tim Jost have argued that the Biden administration should fix this Obama-era mistake. Indeed, President Biden hinted at the possibility of fixing the family glitch in an executive order on the ACA and Medicaid that directed federal officials to examine “policies or practices that may reduce the affordability of coverage or financial assistance for coverage, including for dependents.”
This post discusses the history of the family glitch, recent data on its impact, and why a new interpretation based on the affordability of family coverage would be legally sound. An alternate interpretation that would “fix” the family glitch is supported by the ACA’s text and legislative intent. The Biden administration would have more than sufficient grounds to adopt a reasonable alternative interpretation of the ACA, as the Trump administration did in numerous instances.
What is the Family Glitch?
Under Section 36B of the Internal Revenue Code, individuals generally do not qualify for premium tax credits if they are eligible for another source of minimum essential coverage, including employer-sponsored plans. There is an exception, however, if an employer-sponsored plan is not “affordable” or of “minimum value.” An employer’s plan is not “affordable,” as defined under the ACA, if the employee must contribute more than 9.83 percent of household income towards premiums. (This percentage was initially set at 9.5 percent and is adjusted annually.) If an employer’s plan is not “affordable,” an employee may qualify for premium tax credits through the marketplace, and the employer may face penalties under the employer mandate.
The so-called “family glitch” stems from a 2013 interpretation by the Treasury Department and Internal Revenue Service (IRS) that an employer’s offer of coverage is “affordable” based on the cost of employee-only (rather than family) coverage. This means that an employee and their family members are ineligible for premium tax credits when the employee is offered affordable employee-only coverage. This is true even if the cost of employer-sponsored family coverage would otherwise be unaffordable (i.e., the employee’s contribution towards premiums for family coverage would exceed 9.83% of household income).
This situation—where employee-only coverage is affordable, but family coverage is not—is not uncommon. Most employers offer family coverage, but data from the Kaiser Family Foundation’s 2020 Employer Health Benefits Survey shows that average premiums and employee contributions have increased significantly over time.
In 2020, average premiums for employee-only coverage were $7,470 compared to $21,342 for family coverage. On average, employees contribute 17 percent of the premium for employee-only coverage compared to 27 percent of the premium for family coverage. The average employee contribution for self-only coverage was $1,243, while the average contribution for family coverage was $5,588 (an increase of 22 percent from 2015). The average employee contribution rate is even higher at employers with a large share of lower-wage workers: employees in these firms contribute 38 percent of the premium for family coverage. Six percent of covered employees, including 17 percent of those in small firms, are in a plan with an employee contribution of at least $12,000 for family coverage.
The interpretation of the ACA that created the family glitch was proposed in broader rulemaking from 2011. In the preamble to the proposed rule, the IRS justified this reading of the statute by noting that Section 36B(c)(2)(C)(i) referenced Section 5000A(e)(1)(B), which refers to an employee’s premium contribution for self-only coverage. This reference to Section 5000A(e)(1)(B), the IRS reasoned, meant that the affordability test for both employees and family members is based on the employee’s contribution for self-only (not family) coverage. The IRS also cited a 2011 analysis from the Joint Committee on Taxation (JCT) that interpreted the affordability test to be based on self-only coverage.
The family glitch policy was not finalized alongside other related provisions in a final 2012 rule. Instead, the IRS noted only that commenters opposed the proposed interpretation and that the policy would be finalized in future regulations. It was ultimately finalized as proposed in a short 2-page rule from 2013 that reiterated the prior rationale regarding the cross-reference to Section 5000A(e)(1)(B).
As Tim Jost notes, many comments on both the 2011 proposed rule and the 2012 final rule—found here and here, respectively—urged the IRS to adopt an affordability test for family members based on the cost of family coverage. That interpretation was supported by members of Congress, consumer advocates, labor unions, and others including the National Health Law Program, First Focus, and SEIU who argued that the interpretation was contrary to or, at a minimum, inconsistent with the ACA.
Who Is Affected By The Family Glitch?
The IRS’s interpretation has barred millions of Americans—largely the children of low-income workers—from receiving premium tax credits for marketplace coverage even when they would otherwise qualify based on household income. Two recent studies shed some light on who is affected by the policy and build on prior analyses from the Urban Institute, RAND, and the Agency for Healthcare Research and Quality, among others. The Commonwealth Fund also provided a recent illustrative example of how the family glitch results in higher premiums for families.
In April 2021, the Kaiser Family Foundation estimated that 5.1 million people currently fall into the family glitch, most of whom (2.8 million) are children and nearly half of whom (46 percent) are low-income, earning between 100 and 250 percent of the federal poverty level. Most of these individuals (85 percent) are enrolled in family coverage through an employer-sponsored plan, while about 9 percent are uninsured and 6 percent purchase unsubsidized individual coverage. The vast majority (94 percent) of those in the family glitch reported that they are in good health. Fixing the family glitch would thus increase marketplace enrollment among younger and healthier enrollees, which would improve the health of the overall risk pool and help lower marketplace premiums for all enrollees.
In May 2021, the Urban Institute found that 4.8 million people would be newly eligible for premium tax credits if the family glitch were fixed in the manner described below. Consistent with the Kaiser Family Foundation analysis, the Urban Institute found that nearly half of those affected by the family glitch are children. About 90 percent are insured but paying more than 9.83 percent of their family income towards that coverage. Although many more would be eligible, about 710,000 people would enroll in subsidized marketplace coverage, leading to 190,000 fewer uninsured people. Because these new enrollees would be younger and healthier, individual market premiums would be, on average, about 1 percent lower, according to the Urban Institute analysis.
The Urban Institute further quantified the impact of fixing the family glitch on affordability. Families that switched from employer-sponsored plans would save about $400 per person in premiums on average, with even greater gains ($580 per person) for those with incomes under 200 percent of the federal poverty level. (The Urban Institute had previously estimated that those who fell in the family glitch were spending, on average, nearly 16 percent of their income on premiums alone.) Increased premium tax credits and cost-sharing reductions to cover those who newly enrolled in subsidized marketplace coverage would cost about $2.6 billion annually. There would be no disruptions to the employer market even as hundreds of thousands of families gained access to affordable marketplace coverage.
Fixing The Family Glitch
There are several policy options to fix the family glitch. The most comprehensive approaches would address the family glitch as part of a broader effort to eliminate or reduce the ACA’s current “firewall.” As discussed above, the firewall bars workers with an offer of employer-sponsored coverage from accessing subsidized marketplace coverage. Eliminating the firewall would allow workers to choose to enroll in subsidized marketplace coverage over an employer’s plan, thereby eliminating the restriction that resulted in the family glitch in the first place.
Alternatively, policymakers could focus more narrowly on the family glitch and allow family members (or the employee and family members) to enroll in subsidized marketplace coverage if an employer fails to offer affordable family coverage.
Congress has considered, but not yet adopted, these types of fixes before. Bills to fix the family glitch were introduced as early as 2014, passed by the U.S. House of Representatives in 2020, and included in public option proposals and ACA enhancement bills in 2021. As noted above, the ARP extended subsidies to millions more Americans but did not address the family glitch.
Although Congress has not yet acted, the IRS could adopt an alternative interpretation relative to the 2013 rule and revise the family glitch policy. Members of Congress have acknowledged as much. The 2014 legislation, sponsored by former Sen. Al Franken (D-MN), included a “sense of Congress” that the Secretaries of the Department of Health and Human Services (HHS) and the Treasury “have the administrative authority necessary to apply the affordability provision … to expand access to affordable health insurance coverage for working families without further legislation.”
The Legal Rationale
Although this would not be the most comprehensive “fix” to the family glitch, the Biden administration could revise the affordability test to account for the cost of coverage for family members. This would separate the affordability determination for employee-only and family coverage and enable family members to access subsidized marketplace coverage if an employer offered family coverage that was not “affordable.” Under this policy, employees with an offer of affordable employee-only coverage would still be barred from receiving subsidies, but their family members would not be. Because employees would not be newly eligible for premium tax credits, there would be no impact on liability under the employer mandate (which is only triggered by statute if an employee receives subsidies).
Fixing the family glitch in this manner would be a permissible interpretation of the ACA. Indeed, it would arguably be more faithful to the statute and consistent with the text, structure, and goals of the ACA than the current interpretation that has resulted in the family glitch. This section highlights just some of the legal arguments in support of a revised interpretation.
First, by concluding that the affordability test is the same for family members as for employees, the current interpretation ignores the ACA’s special rules for family coverage in Sections 36B(c)(2)(C)(i) and 5000A(e)(1)(C). In adopting the family glitch policy, the IRS emphasized the cross-reference to Section 5000A(e)(1)(B) but ignored an adjacent provision, Section 5000A(e)(1)(C), which lays out special rules for individuals related to employees. Notably, the IRS reached a different conclusion regarding these special rules when it came to the individual mandate. In interpreting Section 5000A(e)(1)(C), the IRS concluded that an employee’s required contribution towards family coverage, not employee-only coverage, dictated affordability for purposes of an exemption to the individual mandate.
The IRS thus adopted inconsistent interpretations for adjacent statutory provisions, by considering the cost of employee-only coverage for subsidy eligibility but the cost of family coverage for exemptions to the individual mandate. The IRS noted this inconsistency but never reconciled its interpretation with the fact that Congress, had it intended the family glitch, would not have included special rules in the ACA for related individuals. Had Congress intended the special rules for dependents to use the same measure as for employees, it could have used similar language—or it could have omitted the special rules altogether. Instead, in requiring the use of the same test for affordability as for the individual mandate, Congress intended that the entire provision be applied, including the special rule in (e)(1)(C) that qualifies the application of the affordability test for employees in (e)(1)(B).
This alternative reading—that the reference to Section 5000A(e)(1)(B) in Section 36B(c)(2)(C)(i) is a reference to Section 5000A(e)(1)(B) as clarified by the special rule in Section 5000A(e)(1)(C)—is a more than reasonable (and arguably better) construction of the statute that the IRS already adopted in rules regarding the individual mandate. Some commenters suggested as much at the time, noting that a “more cohesive and logical reading” is that Congress referenced Section 5000A(e)(1)(B) because it intended for the entire rule (including the special rule for related individuals) in Section 5000A(e)(1) to be applied to PTC eligibility.
Second, a revised interpretation is more consistent with other parts of the ACA. The Secretary of HHS must, for instance, determine whether an individual’s coverage (rather than an employee’s coverage) under an employer-sponsored plan is unaffordable. Under the ACA, marketplace enrollees who qualify for premium tax credits because “the enrollee’s (or related individual’s) employer” fails to offer affordable coverage must report certain information to HHS. This information includes “the lowest cost option for the enrollee’s or individual’s enrollment status and the enrollee’s or individual’s required contribution” towards the employer-sponsored plan. There would be no reason for the statute to repeatedly distinguish between required contributions for an “enrollee” and an “individual,” or for HHS to require a related individual to report their required contribution for a plan, if the only relevant factor for the affordability test is the cost of employee-only coverage.
Third, the IRS’s reliance on the JCT analysis from 2011 is misplaced. Indeed, JCT reached the opposite conclusion while the ACA was being debated in Congress. A JCT analysis released on March 21, 2010 (the day the bill was passed by the U.S. House of Representatives) noted that the affordability test would be based on the type of coverage applicable (e.g., individual or family coverage). While the JCT adopted a different interpretation later in 2010 and 2011, commenters such as Avik Roy noted that “the JCT’s narrow point of view wasn’t apparent at the time that PPACA was being voted upon, because on the day the final vote took place in the House, the JCT told Congress something different.” An early Senate Finance Committee report also suggested that the affordability test would be based on family coverage.
Finally, eliminating the family glitch is more consistent with the ACA’s goal of expanding access to affordable health insurance than the current interpretation, which unfairly penalizes low-income workers and their families and creates barriers to affordable coverage. Observers and commenters have long raised these concerns, noting that the current interpretation “excludes people Congress intended to cover” and is “simply incongruent” with Congress’s intent. A review of the publicly available comments on the prior rule showed little support for the current interpretation. Indeed, the IRS’s interpretation was attributed more to policy and political concerns that existed in the early years of ACA implementation, such as a desire to avoid higher federal outlays.
Policy (But Not Legal) Criticisms
These are certainly some valid policy criticisms of such an administrative fix. Some will argue that the policy would have little impact on the overall uninsured rate or point to the policy’s price tag of about $2.6 billion annually, as estimated by the Urban Institute. Others will argue that the policy would not go far enough: If the worker cannot join the family members in subsidized marketplace coverage, families will be split between two plans—job-based coverage for the worker and marketplace coverage for the rest of the family—thus paying separate premiums, deductibles, and out-of-pocket maximums. The cost of coverage in this “premium stacking” scenario may still be too high for some families.
These are valid policy criticisms but not legal criticisms. Despite suggestions that agencies cannot change their minds, the Supreme Court has made clear that policy changes are permissible and expected, and agencies can reconsider prior interpretations to reflect new circumstances or a change in policy preferences. As discussed above, the IRS may ultimately read the statute differently or take the position that the ACA is ambiguous on the question of how an offer of employer-sponsored family coverage should be treated for purposes of premium tax credit eligibility. This view, that the statute is ambiguous, is supported by many diverse observers who have long noted that the statute is unclear on the question of the affordability test as it relates to family coverage.
In considering whether to reinterpret the family glitch, the Biden administration will surely weigh these factors as well as the need to expand access to affordable coverage, promote financial security, and respond to the COVID-19 crisis. The fix described above could bring significant financial relief to many working families who are currently paying more than 10 percent of their income towards premiums alone. Indeed, the disproportionate impact of the family glitch on low-income children, in particular, is among the reasons why the U.S. Government Accountability Office previously recommended that the IRS explore an alternative interpretation. The change would also be beneficial for the ACA risk pool, reducing overall marketplace premiums by, on average, 1 percent as new younger, healthier people enroll.