Despite the fact that the Internal Revenue Service (“IRS”) is overburdened and struggling to meet the needs of taxpayers, Congress continues to add to IRS responsibilities in areas that appear far removed from the agency’s core revenue raising function. The Affordable Care Act (“ACA”) is a commonly cited example of the non-revenue raising regulatory roles the IRS is increasingly asked to play, to the criticism of many. After providing an historical overview of the IRS’s involvement in health care regulation, the article provides a partial defense of the expanded role that the IRS has been given as a result of the ACA. The article concludes that much of the IRS’s involvement in health care regulation appears not only defensible, but efficient. For better or for worse, there is no better system for processing payments to or from a large number of taxpayers than the federal income tax system. Additionally, the use of excise taxes to shape taxpayer behavior appears to offer the best of both worlds: a powerful incentive that requires very few enforcement resources. The article concludes, however, that significant burden could be removed from the IRS by modifying or removing the IRS’s substantive rulemaking authority with respect to health care matters, deferring instead to other federal agencies, such as the Department of Health and Human Services, that have particular health policy expertise.
- I. Introduction
- II. How Treasury Initially Got Into the Employee Benefits Business
- III. The Passage of ERISA and the Introduction of Joint Authority for Employee Benefits
- IV. Pre-ACA Substantive Health Plan Regulation through Excise Taxes
- V. The ACA and the Future of the IRS’s Role
- VI. Conclusion
There is no doubt that the Internal Revenue Service (“IRS”) is overburdened and stretched thin, failing to provide even basic levels of responsiveness to taxpayers. Yet despite a low level of resources, Congress continues to add to the IRS’s responsibilities in areas that appear far removed from the agency’s core revenue raising mission. A prominent example of this is the deep involvement of the IRS in implementing the Affordable Care Act (“ACA”).
While the IRS’s involvement in ACA implementation and administration has drawn attention to the fact that the IRS is one of the primary regulators of employer-provided health care, this involvement is nothing new. It was, after all, favorable tax treatment of employer-provided health care that led to the employer-centric system we have today. The ACA, however, greatly expands the administrative burden on the IRS in this realm, to the extent that Professor Hickman in this issue questions whether health care regulation is something in which the IRS should have any role. Nina Olson, the National Taxpayer Advocate, labeled ACA administration one of the “most serious problems encountered by taxpayers” in 2014. Even the Chief Justice appears skeptical of the role of the IRS in ACA administration, stating in King v. Burwell that the IRS “has no expertise in crafting health insurance policy.”
This article wades into the debate concerning the proper scope of IRS administration in a time of restricted resources and deep unpopularity for the IRS. The article begins in Part II with a historical overview of how the IRS became a chief regulator of employer-provided benefits. Part III examines how the passage of the Employee Retirement Income Security Act of 1974 (“ERISA”) created a joint authority model for federal regulation of employee benefits, with the IRS and the Department of Labor sharing jurisdiction. The development of the IRS as a substantive regulator of employer-provided health plans is explored in Part IV, noting the distinction between the IRS as a rulemaker versus the IRS as an enforcer of those substantive rules. The article concludes in Part V by detailing how the ACA expands the IRS’s authority and administrative burden in health plan regulation, concluding that, while there are sound reasons to continue to use the tax code to achieve health policy goals, there are also steps that can be taken to reduce the associated administrative burdens.
For as long as there has been an Internal Revenue Code, there have been tax provisions addressing the taxation of employer-provided benefits. Initially, however, these provisions were limited to the taxability of pension trusts, and simply provided that amounts set aside in pension, profit sharing, or stock bonus plans would not be taxed until distribution, provided such plans were for the exclusive purpose of providing benefits to plan participants, and benefits could only be distributed in accordance with the terms of the plan. There were broadly stated nondiscrimination requirements, but nothing like the complex mathematical requirements that apply today. In addition, the Code provided that employers could deduct contributions made to such trusts. The Code was originally silent regarding any health or welfare benefits offered by employers. As of the 1939 Code, then, the employee benefits provisions could be characterized as “pure tax” provisions. The Code was simply codifying deferred compensation theory in delaying taxation until payment of benefits, while maintaining the employer’s compensation deduction when retirement plan contributions were made, rather than requiring such deduction to be delayed until benefit payout. Other than a slight tax incentive to offer such plans, and a generally stated requirement that the plan be nondiscriminatory, there was no significant social engineering of employee benefits through the Code, and very little required in the way of tax administration.
The 1954 Code marked the first appearance of tax provisions related to employer-provided health care. Sections 105 and 106 were added to the Code, excluding from gross income payments from employer-provided health plans that reimbursed medical expenses, as well as employer contributions to such plans. These Code provisions codified a World War II era revenue ruling that amounts paid by employers for employee health insurance did not constitute income to employees, which was subsequently reversed in 1953. Congress disagreed with this result, permanently adding the exclusion to the Code in 1954. This first appearance of health plan-related provisions could also be characterized as “pure tax.” It simply excluded two types of economic benefit from the definition of gross income and did not otherwise impose any substantive requirements on employer plans. Nor did it attempt to engage in any significant social engineering, other than an incentive for employers to offer health benefits, however structured. As with early pension provisions, these health plan provisions imposed nearly no administrative burden on the IRS.
The passage of ERISA dramatically expanded the role of Treasury in regulating employee benefit plans. As described at the time of passage, “[ERISA] adopts the most sweeping overhaul of pension and employee benefit rules in history. These new rules, which are both tax and non-tax in scope, will affect virtually every pension or other employee benefit plan, whether existent or future.” Rather than simply contain rules related to the taxation of retirement plans, ERISA amended the Code to include substantive participation, vesting, and funding requirements for retirement plans, and placed dollar amount limits on contributions and deductions. The Code was also amended to include an excise tax on certain violations of fiduciary duties. In addition to these tax changes, ERISA contained extensive reporting and disclosure requirements, fiduciary requirements, and remedial provisions that were not included in the Code amendments. ERISA was squarely aimed at pension plans and, while employer health plans were subject to ERISA and had to comply with its general requirements, ERISA did not in any way substantively regulate employer-provided health plans. As a result, no health plan provisions were added to the Code as part of ERISA’s passage.
ERISA did, however, put in place the current regulatory system for employee benefits, which involves joint rulemaking and enforcement authority shared by the IRS and Labor. The statutory text of ERISA provided for joint administration by Treasury and Labor, but also assigned areas of “primary” responsibility. In the early years of implementation, this did not go well. As President Carter declared, “ERISA has been a symbol of unnecessarily complex government regulation.” ERISA’s administrative provisions “have resulted in bureaucratic confusion and have been justifiably criticized by employers and unions alike. The biggest problem has been overlapping jurisdictional authority.” In Reorganization Plan No. 4, President Carter gave Treasury sole authority for enforcing minimum participation, vesting, and funding standards for retirement plans, while granting Labor authority to enforce fiduciary obligations. Treasury, however, continued to have authority to audit plans and levy tax penalties, while Labor continued to have authority to bring civil suits, thereby retaining “the special expertise of each Department.”
Given the difficulties that resulted from joint authority, it is interesting to review why such an unusual structure was adopted in the first place. What motivated Congress to explicitly create joint authority for two agencies over the administration of a single statute? The answer, it appears, lies not in any special theory of agency administration, but rather in turf wars between Congressional committees. As explained by one participant in ERISA’s legislative journey, “there is almost nothing more sacred in Congress than committee jurisdiction. If one committee encroaches on the jurisdiction of another, there is bound to be a great deal of pushback, sometimes very severe pushback. This is what happened in spades with ERISA. Much of the story of ERISA’s passage necessarily becomes the story of the clash between the Tax and Labor Committees and how that was resolved.” Joint Treasury and Labor authority was necessary in order to secure passage of ERISA because neither committee nor agency was willing to give up jurisdiction over pensions. Joint administration was the result of “interagency jealousy.”
Following passage of ERISA, some saw joint administration as a positive, based on the belief that the Department of Labor should not be left to enforce ERISA given its closeness with organized labor. In other words, there was a belief that Treasury would be a more neutral enforcer than Labor. However, this was considered to be a positive side effect of the joint administrative structure, not a motivation for the structure’s creation.
While the reasons behind the creation of the joint administrative structure of ERISA are not noble, scholars have put forth a variety of arguments concerning the benefits of this increasingly common regulatory structure. Professors Freeman and Rossi have declared interagency coordination as “one of the central challenges of modern governance.” One straightforward defense of multiple-agency jurisdiction is that complex policy problems can benefit from the “unique expertise and competencies of different agencies.” When we consider the IRS and Labor, this argument has some appeal. While Labor may be well-suited to tasks that touch on the interactions between employers and employees, the IRS may be better suited for highly technical tasks. Another argument put forward is that shared jurisdiction acts as a check on agency behavior, by reducing the risk of a single agency being captured by regulated parties.
While ERISA brought the IRS firmly into the business of employee benefits, that work was almost exclusively concerned with traditional, defined-benefit pension plans. The only Code provisions concerned with health plans were the uncomplicated sections 105 and 106. This relatively simple approach to the taxation of employer-sponsored health plans began to change in the 1980s, when substantive requirements for group health plans were first included in the Code.
Prior to the enactment of the ACA, there were three major health plan-related additions to the Code that meaningfully expanded the IRS’s role in health regulation. The first of these substantive requirements was passed as part of the Consolidated Omnibus Budget Reconciliation Act of 1985 (COBRA). COBRA gave participants the right to continue their employer-provided group coverage for a limited period of time if such coverage was terminated as a result of a “qualifying event.” These provisions were codified in the Code, ERISA, and the Public Health Services Act (“PHSA”). COBRA, then, followed the basic structure established by ERISA in that both Treasury and Labor remained involved in the regulation of employee benefits. The Department of Health & Human Services (“HHS”) was added to the legislative scheme in order to bring state and local government plans within COBRA’s reach, as ERISA does not apply to such plans. While three agencies were involved in administering virtually identical statutory provisions, COBRA was silent as to how this joint administration was to operate in practice. As a result, the agencies relied on a conference committee report, which stated that rulemaking authority would be split, with Labor authorized to issue regulations implementing the notice and disclosure requirements of COBRA, and the IRS authorized to issue regulations defining the required continuation coverage. HHS was not granted independent authority to interpret COBRA, instead being ordered to conform its implementing regulations to those promulgated by Labor and Treasury. COBRA does, however, contain separate enforcement provisions. On the tax side, the Code originally denied a deduction to employers for their group health plan contributions if they failed to comply with COBRA. Shortly after passage, however, the penalty for noncompliance with COBRA’s requirements was changed to an excise tax payable by the employer of $100 per day per affected individual. On the ERISA side, Labor has the ability to enforce COBRA by bringing a civil suit against a non-complying employer, which may include a penalty of up to $100 per day for each affected beneficiary or participant. Affected individuals may also bring civil suit under ERISA and seek the same remedies. Finally, individuals in state and local government plans are authorized to file civil suits for COBRA noncompliance under the PHSA.
The second major addition to the Code’s regulation of employer-provided health plans was the Medicare as Secondary Payer (MSP) rules. These rules regulate whether an employer plan or Medicare pays first in the case of individuals covered by both, a rule that has significant financial consequences for both the Medicare program and employer plans. The rules generally provide for the employer plan to pay first, thereby saving the Medicare program money. These substantive provisions were codified in the Social Security Act, while the Code was amended to provide for an excise tax on non-complying employers equal to 25% of a non-complying employer’s group health plan expenses in the relevant year. Unlike COBRA, which involved shared rulemaking and enforcement authority among three agencies, the MSP rules followed a simpler model. Because the MSP rules are primarily a cost-saving device for Medicare, the Department of Health & Human Services, specifically the Centers for Medicare & Medicaid Services (CMS), is responsible for rulemaking and primary enforcement, with CMS informing the IRS when the excise tax may be due. This structure can be characterized as “split authority,” rather than the overlapping jurisdiction model we see in other areas of employee benefit regulation.
The third and final regulation of health plans through the Code came with the passage of the Health Insurance Portability and Accountability Act of 1996 (“HIPAA”). HIPAA was a significant piece of legislation that regulated many aspects of health plans and health insurers. From an employer health plan perspective, the major changes made by HIPAA were: (1) limiting the ability of employers to exclude pre-existing conditions from their plans, (2) prohibiting employer plans from discriminating on the basis of health status with respect to eligibility, premiums, and benefits, (3) requiring plans to cover minimum hospital stays following childbirth, and (4) requiring plans that covered mental health treatment and services to do so on parity with other medical services. As with COBRA, HIPAA was codified in three different statutes: the Code, ERISA, and the PHSA. Unlike COBRA, however, HIPAA contained an explicit statutory provision requiring the three agencies to cooperate in rulemaking and enforcement:
SEC. 104. ASSURING COORDINATION.
The Secretary of the Treasury, the Secretary of Health and Human Services, and the Secretary of Labor shall ensure, through the execution of an interagency memorandum of understanding among such Secretaries, that—
(1) regulations, rulings, and interpretations issued by such Secretaries relating to the same matter over which two or more such Secretaries have responsibility under this subtitle (and the amendments made by this subtitle and section 401) are administered so as to have the same effect at all times; and
(2) coordination of policies relating to enforcing the same requirements through such Secretaries in order to have a coordinated enforcement strategy that avoids duplication of enforcement efforts and assigns priorities in enforcement.
The relevant departments entered into the required memorandum of understanding (MOU) in 1999, three years after the passage of HIPAA. The process described in the MOU is one of ongoing collaboration. Each department assigns a representative to “work closely to ensure that all Interpretations, Regulations and enforcement strategies related to shared provisions [of HIPAA] will be developed and implemented in a coordinated manner.” This “Coordinating Committee” is required to meet quarterly (or at other times as they mutually agree) to discuss pending interpretative guidance. The MOU sets a 45-day period to attempt to reach consensus, although it does not spell out what happens in the event that such consensus is not achieved. While there is little publicly available information regarding how the joint rulemaking structure has operated in practice, rulemaking under HIPAA has historically been slow, with officials attributing delays to both the complexity of the statute as well as “the protracted nature of developing policy and rules when multiple federal agencies are involved and the complexity of the statutory provisions.”
While both the statute and the MOU require some degree of coordination with respect to enforcement, it is important to note that different enforcement tools are available to the various agencies. As with COBRA, the IRS’s sole enforcement tool is the imposition of an excise tax of $100 per day per affected individual during the period of noncompliance. Affected individuals may bring civil lawsuits under ERISA to enforce HIPAA, although there is no monetary fine prescribed by the statute. The Department of Labor is not granted authority to commence civil action against a HIPAA non-complying employer; only individuals may commence such actions. HHS, which has authority over health insurers and nonfederal governmental plans with respect to HIPAA, may impose a civil monetary penalty on non-complying health insurers. HIPAA specifically requires Treasury, Labor and HHS to coordinate enforcement activities, and the MOU discussed above lays out the basic framework for such coordination in a highly generalized manner. The MOU provides that the departments will coordinate, share information, and develop a “written operational agreement” that can address various enumerated enforcement topics. The only specific requirement in the MOU with respect to enforcement is that the Departments are required to notify each other in writing prior to commencing any administrative or judicial proceeding on HIPAA matters within the shared regulatory space.
The three pieces of legislation described above were significant additions to the IRS’s role in employer health plan regulation, involving both rulemaking tasks and enforcement tasks. With respect to rulemaking, however, Congress initially appeared interested in shielding the IRS from significant substantive involvement. The IRS was given just a single piece of COBRA rulemaking: the task of defining the required continuation coverage. While the IRS COBRA regulations are extensive, they do not force the IRS to stray too far from typical tax administration. For example, the regulations involve topics such as how corporate transactions affect COBRA rights, how to count employees, how to identify the “employer” and how to calculate plan costs, all topics that are comfortably within the expertise of the IRS. Yet IRS responsibility for COBRA did begin to push the IRS into the health policy arena. For example, the IRS defined in regulations the meaning of “health care” and issued regulations outlining the interaction between COBRA and participant rights under the federal Family and Medical Leave Act. The IRS was completely shielded from any responsibility for MSP rulemaking (leaving them only to collect penalties for noncompliance). With HIPAA, the IRS was given a greatly expanded role in substantive health plan rulemaking (on paper, at least). The IRS was given an equal role at the table with Labor and HHS when it came to any rulemaking that involved employer-provided coverage, and jointly promulgated regulations that had a significant health policy impact. In this way, HIPAA looks like a tipping point at which the IRS really became a health policy agency. What is unclear, however, is what exactly the role of the IRS was in developing these health policy regulations: which agency, if any, took the lead in formulating the regulations, what the actual administrative burden was on the IRS, and what internal expertise, if any, the IRS subsequently developed in the health policy arena.
The sole method for the IRS to enforce the health plan provisions contained in the Code is through the imposition of an excise tax. An excise tax can be imposed through one of two methods: (1) self-reporting by the tax payer, or (2) direct assessment by the IRS, typically accomplished through audit or third-party reporting.
The IRS’s enforcement of COBRA’s requirements has been historically almost non-existent. Until 2010, employers did not even have a duty to self-report COBRA violations and pay the excise tax. Even worse, the dual enforcement structure shared between the IRS and Labor appears to confuse individuals about where they should go for help. A 1990 GAO report found that thousands of individuals contacted the IRS about COBRA rights and violations but, because the Code does not provide individuals with a right of action, such individuals were simply referred to Labor. While it is unknown how many of those individuals subsequently followed up with Labor, IRS very rarely examined the employers of individuals who made inquiries. There is no publicly available data regarding how many employers have self-reported the excise tax, how many employers have been sanctioned on audit, nor how much revenue the excise tax has raised.
With respect to HIPAA compliance, the IRS takes an explicitly passive role, relying solely on “voluntary employer compliance” and referrals from the Department of Labor. Until 2010, however, there was no obligation for noncomplying employers to self-report the excise tax. As of 2001, Treasury officials told the GAO that “they did not believe the agency has assessed, nor has any employer voluntarily paid, an excise tax associated with noncompliance.”
MSP excise tax compliance appears better organized. While there is no requirement for noncomplying taxpayers to self-report the MSP excise tax, CMS has a robust enforcement regime in place, and it has a procedure in place for referring employers that fail to satisfy CMS’s MSP demands for compliance within a specific timeframe to Treasury for imposition of the excise tax. There is no publicly available information regarding the number or such referrals, or the frequency or amount of any excise tax paid as a result. The health plan-related excise taxes, then, do not appear to place a significant administrative burden on the IRS.
The story thus far presents a rather nonsensical role for the IRS in the substantive regulation of group health plans. The IRS has been given what should be a very powerful enforcement mechanism, yet available data suggest that it is not used. While this is positive from an IRS resource perspective, it raises legitimate questions regarding whether these provisions should be in the Code in the first place. As this section will describe, however, the excise tax might not need to be enforced in order to achieve its goals.
Before the passage of the ACA, each of the health plan provisions that was added to the Code was also contained elsewhere in federal law and had litigation-based remedies available to harmed individuals under those provisions. Why, then, might Congress desire to add the additional layer of enforcement through an excise tax? The answer lies in the fact that excise taxes (at least in the health plan context) are not designed to raise revenue. They are designed to strongly encourage ex ante compliance. For the reasons explained below, litigation does not have the same compliance effect.
Participants who are harmed by a lack of health plan compliance may seek redress in state or federal court. Because such lawsuits are brought under ERISA, remedies would typically be limited to requiring the plan to cover the required service or otherwise comply with the legal requirements and to potentially pay the plaintiff’s attorneys’ fees, along with interest if applicable. Punitive damages are not available to ERISA litigants. It is likely that there will be relatively few participant lawsuits seeking to enforce group health plan requirements because of the limited remedies, barriers caused by a lack of knowledge of highly complex legal requirements, lack of easy access to legal services, and a disinclination to sue one’s employer.
In addition to participant lawsuits, however, the Department of Labor has investigative authority that enables it to work with employers to voluntarily remedy any compliance defects that it discovers. The Department of Labor may also commence a civil action against a non-complying employer, although the remedies available are highly limited. As a result, the threat of Department of Labor action, either on audit or through civil lawsuits, is similarly unlikely to create a significant compliance incentive.
The bottom line is that while litigation may serve an important remedial function for harmed participants, it is likely to have little ex ante compliance effect. Lawsuits and Department of Labor action are neither frequent nor costly enough to be effective in making the cost-benefit analysis favor compliance over noncompliance.
While we normally rely on the cost-benefit analysis associated with potential litigation to create the optimal level of legal compliance, ERISA’s highly limited remedial provisions prevent this from occurring in the employer health plan context. Excise taxes, however, have the potential to significantly change the cost-benefit analysis to one that strongly favors compliance.
Excise taxes occupy an unusual role in the federal tax system. They are not typically based on any theory of ideal taxation, nor do they have as their primary goal the raising of revenue. Rather, they are almost always devised to discourage behavior that Congress deems to be undesirable. Excise taxes, it should be noted, provide a specific type of deterrence that is otherwise hard to achieve within the tax system. The most common method that is used in the Code to discourage certain behavior is to deny a deduction for an expense associated with the behavior. For example, the Code denies a deduction for bribes and kickbacks paid by a business, even though they might otherwise be permitted under the general deduction for business expenses. In fact, when COBRA was first added to the Code, noncomplying businesses would lose their deduction for group health plan expenses, rather than face an excise tax. But the loss of a deduction is imperfect in its compliance effect because the value of a lost deduction can vary significantly with the employer’s effective tax rate. A lost deduction might be a strong deterrent for some taxpayers, yet have no effect on taxpayers who have little to no tax liability. In addition, lost deductions are not highly visible to taxpayers, because they are simply folded into the general income tax calculation. An excise tax solves both of these problems by applying uniformly across taxpayers and with high visibility. As a result, excise taxes should factor into the cost-benefit analysis of all taxpayers, thereby improving front-end compliance.
Excise taxes are in fact a form of “announced” penalty for certain behaviors. They are very similar to statutorily specified damages, but instead of being enforced through the courts, they are enforced by the IRS. Because excise taxes are due regardless of participant harm or participant legal action, they can be thought of as a type of precommitment device with respect to compliance. Indeed, if the excise tax is set high enough, it should function to deter nearly all violations of a legal rule. Its effect can be to “front-load” compliance so that further remedies are unnecessary. As Professor Bray summarizes this compliance function of high, announced penalties, “[t]he law is announcing in terrorem.”
While there is a strong case to favor the use of high, fixed penalties in order to achieve compliance, it is not obvious that the best or only way to do so is through excise taxes. For example, COBRA gives the Department of Labor the ability to file suit in the event of noncompliance and seek a monetary penalty of up to $100 per day. However, this provision is unlikely to have the same effect as an excise tax because it both requires litigation and it is discretionary in amount, in contrast to an excise tax. Theoretically, however, Congress could remedy both of these defects and give the Department of Labor the authority to impose a civil monetary fine equal to the excise tax through administrative procedures. Interestingly, Congress appears to have favored IRS enforcement in this area because the IRS is viewed as much more fearsome than Labor, particularly given Labor’s close historic ties to organized labor. So while other agencies could be given the ability to impose high, fixed penalties for noncompliance, the compliance effect might not be as great as the imposition of these penalties by the IRS.
On a theoretical level, then, there is much to support the excise tax model as a compliance mechanism. Given current IRS funding levels and resources, however, it is not clear that the excise tax will actually achieve its goals. If it becomes apparent to employers that the threat of excise tax enforcement is empty, it may have little effect. But this is speculation; the mere threat of outsized excise tax penalties may be enough to force compliance, even if regulated parties believe the probability of being caught is low.
It seems reasonable to believe that the success of excise taxes in ensuring front-end compliance will depend both on actual enforcement and on perceptions of enforcement by regulated parties. Publicly available information, however, suggests that there is very little enforcement of the excise tax provisions. Until 2010, there was not even an affirmative obligation for taxpayers to self-report either the COBRA- or HIPAA-related excise taxes, decades after the statutes became law, and available information suggests that the IRS does little to no independent enforcement of these provisions.
The good news from an IRS resource perspective is that these excise taxes appear to require little to no effort with respect to enforcement (in contrast to rulemaking), but that raises the question of whether unenforced excise taxes have the compliance effect theorized above. Usually, tax compliance is concerned with whether taxpayers voluntarily report their tax due accurately. Compliance is different in the case of the excise taxes. The compliance we are concerned with is not what is reported on a tax return (indeed, as noted above, there were no requirements to self-report any of the health plan-related excise taxes until 2010), but rather whether the excise taxes result in employers complying with the underlying health plan regulations. Nevertheless, there is likely still something to be learned from the tax compliance literature. There are two general theories regarding tax compliance: (1) taxpayers comply with tax laws because they engage in a cost-benefit analysis of compliance versus noncompliance, and find compliance to be the least costly, and (2) taxpayers comply with tax laws in order to conform to personal or social norms.
Applying these compliance theories to excise taxes, we can guess that both contribute to compliance with the substantive health plan rules. With respect to the cost-benefit theory of tax compliance, a large penalty, even if rarely enforced, is likely to have the same deterrence effect as a smaller penalty more consistently enforced. And the excise taxes at issue here are large indeed. Recall that most are $100 per day per participant. In general, it would be highly unlikely that noncompliance would occur for only a short period of time, as the structure of health plans does not frequently change (typically no more than once per year). An employer with 50 employees, who failed to comply with a group health plan requirement for a single year, would face a potential excise tax of $1.825 million. A large, national retailer with 125,000 employees would face a potential excise tax of $3.65 billion. Even if the perceived probability of enforcement is incredibly low, it is not hard to imagine that the cost-benefit analysis for employers of all sizes favors compliance.
On top of the cost-benefit analysis, it also appears to be the case that many taxpayers will comply with the law out of adherence to social norms. As a result, while we have no hard data on this issue, it appears likely that the excise tax strongly encourages front-end compliance even with very low levels of enforcement. From both an IRS resources and compliance perspective, then, the excise taxes appear to be a worthwhile addition to the Code.
Prior to the ACA’s enactment, the IRS was already firmly entrenched in the business of employer health plan regulation. While the reasons behind that involvement may have been less than compelling, the use of the excise tax as a compliance mechanism appears sound. Excise taxes should greatly improve front-end compliance with rules that might otherwise be too easy to ignore, and, even better given current IRS resources, can achieve their goals with relatively little investment by the government. The IRS’s substantive health plan rulemaking is more troubling, but closer examination reveals that such rulemaking often involves tasks comfortably within the IRS’s expertise, and that shared rulemaking authority may have placed little burden on the IRS if, as one would expect, agencies with greater substantive expertise took the lead in drafting such rules.
The ACA continues the pattern of the IRS’s involvement in the substantive regulation of employer-provided health plans. Along with Labor and HHS, the IRS will be significantly involved in administering the ACA’s individual mandate, employer mandate, premium tax credit, excise tax on high-cost employer plans, and substantive group health plan requirements. The role of the IRS in both rulemaking and enforcement around these ACA requirements is explored in more detail below.
The scope of the ACA is vast, and the corresponding rulemaking required to implement the law is similarly staggering. One of the best known tax provisions in the ACA is the so-called individual mandate. The individual mandate imposes a financial penalty on uninsured individuals who have affordable health insurance available to them. Because this provision is contained exclusively in the Code, the IRS has primary rulemaking authority. The rulemaking authority is not, however, exclusive. HHS has the authority to determine which individuals qualify for hardship exceptions to the individual mandate, and HHS and Treasury are required to coordinate efforts to further define what “other coverage” might satisfy the individual mandate requirements.
A similar rulemaking structure is used with respect to the employer mandate penalty, which is also contained exclusively in the Code. While the IRS has primary rulemaking authority, Labor is relied on to define a “seasonal worker” for purposes of the penalty, and Labor and IRS together must define “hours of service” for purposes of determining who is a full-time employee for purposes of the penalty.
The premium tax credit falls almost entirely within the IRS’s authority, although the statute gives HHS the authority to determine which part of the premium, if any, is attributable to state-required benefits that are not permitted to be included for purposes of the premium tax credit calculation. In addition, the statute instructs the IRS and HHS to cooperate regarding necessary regulations on family size and household income for purposes of the premium tax credits. The excise tax on high-cost employer plans, known colloquially as the “Cadillac Tax,” falls entirely within the IRS’s rulemaking authority.
The most significant rulemaking burden imposed on the IRS by the ACA may be the oft-overlooked section 9815. Section 9815 incorporates into the Code all of the ACA’s group health plan requirements that were added to the PHSA. These include at least twenty substantive requirements, and were also codified in section 715 of ERISA. Plans that fail to comply with the group health plan requirements are subject to the $100 per participant per day excise tax in section 4980D. This tri-statute amendment scheme mirrors HIPAA’s structure, and was presumably based on the same rationale. Amending both the PHSA and ERISA is necessary to cover all group health plans, because ERISA does not apply to governmental or church plans, and amending the Code through the addition of an excise tax is an important tool to encourage front-end compliance.
The result of codifying the requirements in three statutes is that all three agencies have a role in rulemaking. While the ACA does not contain any specific requirement for coordinated rulemaking for the group health plan requirements, the MOU the Departments signed to coordinate HIPAA rulemaking and enforcement was also intended by the Departments to apply to any future federal legislation “concerning health care which result in two or more . . . Departments having shared jurisdiction.” The three agencies have in fact been following the MOU coordination provisions, effectively giving the three agencies an equal role in the regulatory process, given the requirement to reach a consensus. Imposing such a rulemaking burden on the IRS, however, is difficult to justify.
The group health plan requirements are far removed from nearly any traditional tax issues or concerns, and may be well outside the competence of even those in the employee benefits division of the IRS. The substantive group health plan requirements include detailed provisions regarding what treatments and services a plan must cover, how plans can impose cost-sharing for those services, and how coverage can be priced, among many others. Perhaps the most well-known of these provisions is the requirement that group health plans cover preventive health services. Preventive health services were defined to include contraception, which some private employers objected to on religious grounds, resulting in the Hobby Lobby litigation. The IRS, as a result of the Hobby Lobby decision, helped draft a solution that both provided the required contraceptives to employees, but did not interfere with a closely-held private firm’s religious beliefs. It is hard to imagine a task that is further removed from the core tasks of a revenue agency. Of course, the overlapping jurisdiction model makes it difficult to determine exactly what role the IRS played in drafting these regulations.
Prior to the ACA, IRS enforcement of group health plan regulations was accomplished through imposition of an excise tax on non-compliant plans. The ACA continues to build on this model, but it also moves IRS enforcement into new areas.
Several of the major tax provisions in the ACA are non-excise taxes, including the individual mandate penalty and the employer mandate penalty. In terms of enforcement, these non-excise taxes fit relatively easily into the existing structure of our tax system. They require individual taxpayers to self-report and pay any tax due as a result of failing to purchase affordable health insurance and require large employers to self-report and pay any tax due as a result of failing to offer employees affordable health insurance coverage. While tax forms need to be modified accordingly, these taxes look and feel like many others already in our system. And in terms of how the IRS ensures compliance with these provisions, the standard auditing mechanism can easily be utilized, requiring few extra resources. Individual mandate enforcement does, however, play by slightly different rules. The statute specifies that the individual mandate tax shall be treated as an “assessable penalty under subchapter B of chapter 68.” However, the Code specifies that the IRS may not utilize liens or levies where there has been a failure to pay the individual mandate tax due. The employer mandate follows the same “assessable penalty” treatment, although there is no limitation of liens or levies in its provisions.
The premium tax credits are not “enforced” in the way that we typically think of enforcement, and therefore do not result in typical enforcement costs. Instead, it is more helpful to think of premium tax credits as requiring administration. Here we see a significant burden being placed on IRS resources. While the IRS is no stranger to administering tax credits, the structure of the premium tax credits is highly unique because the credit is advanceable. Tax credits are normally administered through the standard tax return filing process. When tax payers file their return for a given year, they will claim any applicable credits on their return and the amount of the tax they owe will be correspondingly reduced. While this works well for many types of credits, there was concern that requiring taxpayers to wait until the end of the year (when they filed their tax returns) to receive the benefit of the premium tax credit would defeat the purpose of providing the cash flow necessary to pay for health insurance for that year. Those who are eligible to receive premium tax credits do not often have the ability to pay the cost of health insurance upfront and then wait a year or more for reimbursement through the tax filing process. As a result, the ACA allows individuals to receive an advance of the premium tax credit, using a best estimate of eligibility. Individuals who apply for coverage through an exchange are screened for tax credit eligibility. The IRS shares data with the exchanges to verify income from the most recently available year. The exchange then informs the IRS of the estimated tax credit, and the IRS pays the relevant insurance companies directly on the individual’s behalf. When that individual files her tax return for the year, the premium tax credit will be reconciled based on the actual amount of that individual’s household income, compared to the estimate on which the advanced credit was based. If additional credit amounts are due, those will be credited to the taxpayer and will either reduce the tax that is owed or be refunded. If the advanced credit was larger than the taxpayer was actually entitled to, the taxpayer may be required to repay the excess amount. The administrative burden of estimating and reconciling tax credits is significant and has both confused and angered taxpayers. While it may make sense to administer widespread financial subsidies through the Code, the complex system of advancing and reconciling tax credits imposes a serious burden on the IRS, not only in terms of workload, but also with respect to the reputation of the IRS.
The Cadillac Tax is an excise tax, although it differs in key respects from the type of excise tax we typically see in the health plan context. First, the amount of the excise tax is calculated in a different manner than the typical $100 per day per individual. Instead, the tax is equal to 40% of the cost of employer-provided health care coverage above a certain dollar threshold that can vary based on an individual’s age, type of profession, or union status. As a result, it is much harder to determine easily, and in advance, how much would be due. In addition, the tax is spread among all health plan providers, which may include multiple taxpayers, such as the employer, and one or more health insurers. Perhaps the most important difference is that it is not designed to encourage compliance with substantive health plan regulations, but is instead aimed at encouraging employers to offer less generous health plans, with the intent of lowering overall health care expenditures. In this sense, the Cadillac Tax is similar to the other health-related excise taxes in that its ultimate goal is to change behavior, not raise revenue. As a result, while the Cadillac Tax’s provisions are quite complicated, there may not be a significant administrative cost associated with them, since the strong incentive it provides to avoid the excise tax in the first place largely negates the need for IRS administration and enforcement.
Finally, the IRS has a potentially enormous compliance burden imposed by section 9815, which incorporates all of the ACA’s substantive group health plan requirements into the Code. Section 9815, together with section 4980D, subjects non-complying employers to the same $100 per participant per day excise tax as that imposed for HIPAA noncompliance. The MOU signed between Treasury, Labor, and HHS by its terms also applies to the ACA. All that the MOU requires in terms of enforcement, however, is coordination between the Departments. It does not grant a single agency primary authority, or in any way limit enforcement. What is not publicly known is how the agencies will handle enforcement in practice. Recall that with HIPAA, Treasury does not independently enforce HIPAA, relying instead on voluntary employer compliance and referrals from Labor. If the same model is followed here, there will be little to no administrative burden placed on the IRS. The risk, however, is that if employers begin to believe that noncompliance will never be punished via the excise tax, the excise tax may fail to achieve its compliance goal. However, as discussed above, the mere possibility of a very large fine, imposed by the “fearsome” IRS, may be sufficient to achieve the desired employer behavior.
The ACA significantly expands the IRS’s role in health care regulation, and it is far from obvious that placing such a burden on the agency is a wise policy decision. This section will explore whether it might be possible to achieve the ACA’s policy goals in an efficient manner while lessening the role the IRS must play.
The clearest justification for IRS involvement in ACA administration is based on the fact that the IRS is the most obvious choice for processing payments to or from large numbers of taxpayers. Starting with the ACA’s payments to taxpayers, the ACA’s tax credits are calculated based on income, and the IRS obviously has the most readily available information on household income, along with pre-existing definitions of income that can be used for this purpose. The scale of the premium tax credits also supports that argument that the IRS is better positioned than other agencies to handle administration, as many more individuals will be eligible for premium tax credits than for other pre-existing social welfare programs. The IRS also has experience in making the types of calculations required for premium tax credit eligibility. While calculation and payment of premium tax credits could clearly be given to a different agency, with HHS being the most obvious choice, the IRS would necessarily have to be involved in order to provide accurate income information. But adding an additional agency would not seem to offer any efficiencies (other than perhaps making budget allocations easier to come by given the politically disfavored status the IRS currently enjoys).
Similar reasoning supports IRS involvement in those areas of ACA administration that require payments from taxpayers to the federal government as part of the individual and employer mandates. The amount of the individual mandate varies based on family size and income, both of which are readily available to the IRS. And the employer mandate is based in part on employer size, again something the IRS already tracks. In addition, the IRS obviously already has a procedure in place for collecting amounts due from the same population, something no other federal agency can claim. And finally, it is important to note that the IRS is required by constitutional limitations to be involved in the administration of the individual mandate. The constitutionality of the individual mandate was, after all, upheld as a valid exercise of Congress’s taxing powers.
But what about the excise taxes imposed by the ACA? The previous parts have established that excise taxes cannot be thought of as “pure tax” provisions. They are distinct from our general income tax system and are not designed to raise revenue. They function as compliance mechanisms in the case of section 9815 and as a cost-control mechanism in the case of the Cadillac tax. But these provisions are likely to accomplish their goals without any active IRS enforcement, and in this sense appear to be highly efficient.
The troublesome part of the excise taxes, from a tax administrative standpoint, is the rulemaking burden that comes along with the excise taxes. As Professor Hickman has documented, nearly ten percent of IRS rulemaking projects between 2008 and 2012 were ACA-related tasks. However, many of these regulations do involve tasks or topics that are squarely within IRS expertise. For example, rulemaking on how employees are to be counted and how employers are to be aggregated for purposes of the employer mandate are subjects with which the IRS has vast experience. Similarly, rulemaking regarding the computation of individual mandate penalties, reconciliation and reporting requirements for the individual mandate are also standard IRS functions. However, many other regulations appear to have no obvious relationship to either the tax system or IRS expertise. IRS regulations have covered such diverse topics as the requirements for group health plans to (1) cover preventive services, including religious exemptions therefrom resulting from the Hobby Lobby decision, (2) comply with specific procedures when processing claims, (3) cover pre-existing health conditions, (4) not include certain lifetime and annual limits on benefits, (5) cover children through age 26, and (6) not rescind coverage except in specific circumstances. It is in this health policy arena that I believe the strongest arguments can be made against IRS involvement. Yet, the health policy regulations just mentioned were not promulgated solely by the IRS. Because the group health plan requirements are codified in three statutes, they were jointly promulgated pursuant to the 1999 MOU by IRS, HHS, and Labor. It is possible, therefore, that the IRS’s actual involvement in these health policy oriented regulations was minimal, with the IRS deferring to HHS’s lead—but no publicly available information confirms whether that intuition is accurate.
The IRS is overburdened, and coming dangerously close to being unable to competently perform its core revenue-raising function. Yet Congress continues to add to the IRS’s workload, most notably in recently years through the many tax-related provisions of the ACA. While the IRS has always had a role to play in regulated employer-provided benefits, the ACA substantially expands that role. This raises an obvious question of whether it is wise to give the IRS what appears to be a primary role in health plan regulation.
As this article has argued, however, much of the IRS’s involvement appears not only defensible but efficient. For better or for worse, there is no better system for processing payments to or from a large number of taxpayers than the federal income tax system. Additionally, using excise taxes to shape taxpayer behavior appears to offer the best of both worlds: a powerful incentive that requires very few enforcement resources. The only troubling aspect of the IRS’s involvement in ACA administration stems from the IRS’s broad rulemaking authority across a number of substantive health plan provisions. Why, after all, should it fall to the IRS to determine how to accommodate the religious freedoms of closely held corporations as they apply to the provision of contraceptives? But we do not actually know the extent of IRS involvement in these and other health plan regulations, because three agencies jointly promulgate such rules. There are, however, multiple steps we could take to ease our concerns both about the IRS’s expertise in such matters and the burden it imposes on an already struggling agency. HHS, Labor, and IRS could renegotiate their MOU to make the IRS’s role explicitly more limited. Or Congress could potentially take rulemaking authority away from the IRS, instead requiring IRS to follow the regulations issued by HHS and Labor.
We may not like the role the IRS has come to play in health plan regulation, and we may fear that it makes little sense. But the fault here lies not with any recent additions to the Code, but rather America’s long-standing tradition of having employers as the primary providers of health insurance and the provision of a tax benefit associated with such coverage. Because the ACA was an attempt to achieve universal coverage through the existing employer-based system, the statute requires significant reliance on the tax system to achieve its goals. And upon closer examination of the specifics of the ACA, there appears to be strong justification for most (but certainly not all) of its tax provisions.
 See National Taxpayer Advocate, 2014 Annual Report to Congress vii (2014) (noting, among other things, “a devastating erosion of taxpayer service”).
 Id. at 67.
 King v. Burwell, 135 S. Ct. 2480, 2489 (2015).
 See I.R.C. of 1939 § 165 (providing tax benefits to employer-sponsored retirement plans). Section 165 in the 1939 Code was the predecessor to § 401(a) in the current Code.
 The nondiscrimination requirements prohibited discrimination in favor of highly-compensated employees. Such provisions were not in the 1939 Code as originally enacted, but were added as amendments on Oct 21, 1942 by Session Law 56 Stat 798 (77th Congress, 2nd Session, Ch. 619).
 I.R.C. of 1939 § 23(p) (2015).
 The greatest administrative burden imposed by the employee benefits tax provisions would likely have been enforcing the nondiscrimination requirements. However, employers had a duty to self-report their compliance with such provisions, leaving the IRS only with the discretionary task of verifying such reports. See Internal Revenue Service, Form 4848 (1973), https://www.irs.gov/pub/irs-prior/f4848–1973.pdf [https:// perma.cc/85CQ-WPXX] (illustrating the minimal reporting requirements related to pension plan nondiscrimination rules).
See I.R.C. of 1954 §§ 105, 106 (2015).
 David A. Hyman & Mark Hall, Two Cheers for Employment-Based Health Insurance, 2 Yale J. Health Pol’y L. & Ethics 23, 25 (2001).
 Commerce Clearing House, Inc., Pension Reform Act of 1974 with Explanation, Preface (1974).
 See, e.g., I.R.C. §§ 401(a)(3); 404(a)(3)(A); 415(a), (b) (2015).
 See I.R.C. §§ 4975(a), 4975(c)(1)(E) (2015).
 See 29 U.S.C. §§ 1001 et seq. (2015).
 ERISA created the office of Assistant Commissioner, Employee Plans and Exempt Organizations within the Internal Revenue Service. See Commerce Clearing House, Inc., supra note 11, at 904.
 Employee Retirement Income Security Act of 1974, Pub. L. 93-406, § 3004.
 For example, primary administrative responsibility for participation, vesting, and funding provisions was assigned to the IRS. Commerce Clearing House, Inc., supra note 11, at 901. The IRS was also granted authority to audit qualified plans. Id. at 902.
 Reorganization Plan No. 4 of 1978, 5 U.S.C. app. (2006).
 Panel 2: Making Sausage – The Ninety-Third Congress and ERISA, 6 Drexel L. Rev. 291, 300 (2014).
 Id. at 304.
 Panel 3: Negotiating the Agency Peace Treaty: Reorganization Plan No. 4, 6 Drexel L. Rev. 319, 324 (2014).
 Jody Freeman & Jim Rossi, Agency Coordination in Shared Regulatory Space, 125 Harv. L. Rev. 1131, 1134 (2012).
 Id. at 1142.
 Id. at 1142-43.
 Section 105 did become slightly more complicated when paragraph (h) was added as part of 1978 Revenue Act, Pub. L. 95-600 § 366, which taxes discriminatory health benefits provided to highly-compensated employees.
 I omit from this article any discussion of section 89 of the Code, which was added to the Code as part of the 1986 Tax Reform Act. Section 89 imposed non-discrimination requirements on employer provided health plans, similar to the requirements imposed on retirement plans, and for similar reasons. For plans that were considered discriminatory under section 89, highly-compensated employees covered by such plans had to include in gross income the amount of the discriminatory benefit. Given the significant cost of the tax expenditure for employer-provided health plans, Congress was attempting to ensure that the tax benefit was not concentrated among highly-compensated individuals. Section 89, however, was not long-lived. It was repealed in 1989 following significant backlash against its complexity and administrative impracticability. For further information on section 89’s history, see Rosina B. Barker, Lessons from a Legislative Disaster, 47 Tax Notes 843 (1990), http://www.ipbtax.com/assets/pdf/publication_129.pdf.
 Consolidated Omnibus Budget Reconciliation Act of 1985 (COBRA), Pub. L. 99-272, 100 Stat. 82 (1986).
 I.R.C. § 4980B (2015).
 COBRA, Pub. L. 99-272 § 10002.
 Id. § 10003.
 H.R. Conf. Rep. No. 99-453, 99th Cong., 1st Sess., at 562-63 (1985).
 Id. at 563.
 COBRA § 10001(a).
 I.R.C. § 4980B(b) (2015).
 ERISA § 502(a) (authorizing civil suits by participants or the Department of Labor) and §502(c)(1) (authorizing courts to award up to $100 per individual per day, along with other relief the court deems proper).
 ERISA § 502(a).
 COBRA § 10003 (adding § 2207 to the PHSA).
 Omnibus Budget Reconciliation Act of 1986, Pub. L. 99-509 § 9319.
 Id. § 9319(a).
 Id. § 9319(d)(1) (adding § 5000 to the I.R.C.).
 As the CMS enforcement manual explains:
If CMS Central Office determines that a plan has been a nonconforming [group health plan] in a particular year, it refers its determination, including the identity of the contributors that it has identified, to the IRS, but only after the parties have exhausted all appeal rights with respect to the determination. Section 5000 of the Internal Revenue Code of 1986 imposes an excise tax penalty on employers and employee organizations that contribute to nonconforming GHPs. They are taxed 25 percent of the employer’s or employee organization’s expenses incurred during the calendar year for each [group health plan] (conforming as well as nonconforming) to which they contribute. This tax penalty does not apply to Federal and other governmental employers. The IRS administers Section 5000 of the IRC, which imposes the tax on employers (other than governmental entities) or employee organizations that contribute to a nonconforming [group health plan] mentioned in § 80.
Centers for Medicare and Medicaid Services, Medicare Secondary Payer (MSP) Manual 36 (2014), https://www.cms.gov/Regulations-and-Guidance/Guidance/Manuals/downloads/msp105c01.pdf.
 Health Insurance Portability and Accountability Act of 1996 (HIPAA), Pub. L. 104-191, 110 Stat. 1936 (codified as amended in scattered sections of the U.S. Code).
 Id. § 101 (adding §§ 701 et seq. to ERISA).
 See generally HIPAA, Pub. L. 104-191 (1996).
 42 U.S.C.A. § 300gg‑92 note (2015).
 Notice of Signing of a Memorandum of Understanding among the Department of the Treasury, the Department of Labor, and the Department of Health and Human Services, 64 Fed. Reg. 70164 (Dec. 15, 1999) [hereinafter “HIPAA MOU”].
 Id. at 70165.
 U.S. Gen. Accounting Office, Implementation of HIPAA: Progress Slow in Enforcing Federal Standards in Nonconforming States 15-16 (2000).
 I.R.C. § 4980D(b) (2012).
 ERISA § 502(a)(1).
 Id.; HIPAA MOU, supra note 49, at 70165.
 42 U.S.C. § 300gg‑22 (2012).
 HIPAA MOU, supra note 49, at 70166.
 Treas. Reg. § 54.4980B‑9 (2012).
 Treas. Reg. § 54.4980B‑2 (2012).
 Treas. Reg. § 54.4980B‑8 (2012).
 Treas. Reg. § 54.4980B‑10 (2012).
 Treas. Reg. §§ 54.9801‑1 to 54.9801‑3, 54.9811‑1 (2012); I.R.C. § 9812 (2012).
 U.S. Gen. Accounting Office, Improvements Needed in Enforcing Health Insurance Continuation Requirements 5 (1990) (noting one completed examination of a potential COBRA violation).
 74 Fed. Reg. 45,994 (Sept. 8, 2009).
 U.S. Gen. Accounting Office, supra note 66, at 5.
 U.S. Gen. Accounting Office, Private Health Insurance: Federal Role in Enforcing New Standards Continues to Evolve 11 (2001) [hereinafter “HIPAA GAO report”]. In practice, HHS appeared to be overburdened by its HIPAA enforcement activities, and therefore struggled to effectively enforce HIPAA’s requirements. Labor, at least in the early years, relied largely on consumer complaints to identify noncompliance. A random sample of employer HIPAA compliance by Labor, conducted in 1999, revealed a 21% noncompliance rate for certain HIPAA standards—although the DOL characterized many of the violations as “technical” in nature. Where DOL discovered violations, it worked with the employer to voluntarily correct the default and had not (as of 2001) initiated any legal action to force compliance. Id. at 10-11.
 74 Fed. Reg. 45,994 (Sept. 8, 2009).
 HIPAA GAO report, supra note 69, at 11.
See Ctr. for Medicare & Medicaid Serv., Group Health Plan Recovery, https://www.cms.gov /Medicare/Coordination-of-Benefits-and-Recovery/Coordination-of-Benefits-and-Recovery-Overview /Group-Health-Plan-Recovery/Group-Health-Plan-Recovery.html, for an overview of the CMS enforcement process and referrals to Treasury.
 ERISA § 502.
 Mass. Mut. Life Ins. v. Russell, 473 U.S. 134 (1985).
 ERISA § 502.
 This is not to suggest that employers are relying solely on a cost-benefit analysis when deciding whether to comply with the law. Many other forces are clearly at play. See, e.g., Robert Cooter, Do Good Laws Make Good Citizens? An Economic Analysis of Internalized Norms, 86 Va. L. Rev. 1577 (2000); Cass R. Sunstein, Social Norms and Social Roles, 96 Colum. L. Rev. 903 (1996).
 See Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. Pol. Econ. 169 (1968); A. Mitchell Polinsky & Steven Shavell, Punitive Damages: An Economic Analysis, 111 Harv. L. Rev. 869, 887-96 (1998).
 For example, the IRS has explicitly stated that COBRA excise tax is “designed as a deterrent against noncompliance.” Internal Revenue Serv., Audit Techniques and Tax Law to Examine COBRA Cases (Continuation of Employee Health Care Coverage), http://www.irs.gov/Businesses/Small-Businesses-&-Self-Employed/Audit-Techniques-and-Tax-Law-to-Examine-COBRA-Cases-Continuation-of-Employee-Health-Care-Coverage.
 I.R.C. § 162(c) (2012).
 See John D. Colombo, Paying for Sins of the Master: An Analysis of the Tax Effects of Pension Plan Disqualification and a Proposal for Reform, 34 Ariz. L. Rev. 53, 84 (1992) (referring to excise taxes as the “obvious choice for a targeted deterrent sanction”). See also Internal Revenue Service Commissioner’s Penalty Study, Report On Civil Tax Penalties IX-16 (1989) (noting that excise taxes “generally have been effective” as deterrents).
 See generally Samuel L. Bray, Announcing Remedies, 97 Cornell L. Rev. 753 (2012) (discussing the term in the context of civil remedies).
 Id. at 781.
 Id. at 784-85.
 Id. at 785.
 Panel 3: Negotiating the Agency Peace Treaty: Reorganization Plan No. 4, 6 Drexel L. Rev. 319, 324, 335 (2014).
 74 Fed. Reg. 45994 (promulgating Treas. Reg. § 54.6011‑2).
 See Michael Doran, Tax Penalties and Tax Compliance, 46 Harv. J. on Legis. 111, 124-38 (2009); Alex Raskolnikov, Revealing Choices: Using Taxpayer Choice to Target Tax Enforcement, 109 Colum. L. Rev. 689, 694-710 (2009).
 See Curtis W. Copeland, Cong. Research Serv., R41180, Rulemaking Requirements and Authorities in the Patient Protection and Affordable Care Act (PPACA) 1-2 (2011), http://perma.cc/XB5K-TJJS, for an overview of the regulatory tasks involved in ACA implementation.
 I.R.C. § 5000A.
 I.R.C. § 5000A(e)(5).
 I.R.C. § 5000A(f)(1)(E).
 PPACA § 1513 (adding § 4980H to the I.R.C.).
 I.R.C. § 4980H(c)(2)(B)(ii).
 I.R.C. § 4980H(c)(4)(B).
 I.R.C. § 36B(b)(3)(D)(ii).
 I.R.C. § 36B(e)(3).
 See generally I.R.C. § 4980I.
 See Patient Protection and Affordable Care Act, Pub. L. No. 111-148, § 1001 (2010) (adding §§ 2711-2719A to the PHSA).
 HIPAA MOU, supra note 49, at 70164-65.
 The one exception to this statement is the provision in section 2716 of the PHSA that prohibits discrimination in favor of highly compensated employees. While not necessarily a traditional tax provision, it is a well-established concept within the employee benefits tax community.
 Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751 (2014).
 I.R.C. § 5000A(g) (2010). Pursuant to I.R.C. § 6671 (2012), assessable penalties are “assessed and collected in the same manner as taxes.”
 I.R.C. § 5000A(g) (2010). This provision was likely included for policy reasons. Politicians may not have wanted individuals who failed to purchase health insurance to be subject to the full array of IRS enforcement in the event they failed to pay the mandate penalty.
 I.R.C. § 4980H(d)(1) (2003).
 42 U.S.C. § 18082 (2011).
 42 U.S.C. § 18081 (2011).
 42 U.S.C. §§ 18081-18082 (2011).
 I.R.C. § 36B(f) (2011).
 I.R.C. § 36B(f)(2) (2011). Full repayment is required if the individual has household income equal to or greater than 400% of the federal poverty limit for the relevant year, but the Code places limitations on the amount that must be repaid for individuals who earn less than 400% of the federal poverty limit. Id.
 See Ted Griggs, Thousands in La. To Repay Portion of Health Subsidies, Advocate (April 15, 2015), http://theadvocate.com/news/12097896-123/thousands-in-la-will-have [http://perma.cc/5WXP-WAAX]; Jayne O’Donnell, ACA Could Trip Up Some at Tax Time, USA Today (Jan. 9, 2015), http://www .usatoday.com/story/news/nation/2015/01/08/obamacare-health-insurance-taxes-exemptions-check-box /21440171/ [http://perma.cc/3WPS-DDDJ]; Pamela Yip, Confusion Likely for Many Filing Taxes, Dallas Morning News (Feb. 23, 2015), http://www.dallasnews.com/business/personal-finance/headlines/20150222-health-insurance-confusion-likely-for-many-filing-taxes.ece [http://perma.cc/B6ZF-8YQT].
 See Timothy Jost, Implementing Health Reform: Complicated ACA Tax Forms Could Cause Problems, Health Affairs Blog (Sept. 21, 2014), http://healthaffairs.org/blog/2014/09/21/implementing-health-reform-complicated-aca-tax-forms-could-cause-problems/ [http://perma.cc/M6WC-LNH4] (stating that it is likely that taxpayers receiving premium tax credits are “likely to be confused, frustrated, even angry, and certainly bewildered, completing [the relevant tax] forms”). See also Press Release, Senator Orrin Hatch, Five Years Later: Obamacare Subsidies Cause Confusion and Frustration for American Taxpayers (Mar. 18, 2015), http://www.hatch.senate.gov/public/index.cfm/2015/3/five-years-later-obamacare-subsidies-cause-confusion-and-frustration-for-american-taxpayers [http://perma.cc/7PXF-Q5VC].
 I.R.C. § 4980I(b) (2011).
 I.R.C. § 4980I(c) (2011).
 For a detailed discussion of the behavioral goals of the Cadillac Tax, see Amy B. Monahan, Why Tax High-Cost Employer Health Plans?, 65 Tax L. Rev. 749 (2012).
 I.R.C. §§ 9815, 4980D (2012).
 HIPAA MOU, supra note 49, at 70164-65.
 See notes 69-71 and accompanying text.
 Panel 3: Negotiating the Agency Peace Treaty: Reorganization Plan No. 4, 6 Drexel L. Rev. 319, 335 (2014).
 Nat’l Fed’n of Indep. Bus. v. Sebelius, 132 S. Ct. 2566, 2593-2601 (2012).
 See generally Kristin E. Hickman, Administering the Tax System We Have, 63 Duke L. J. 1717 (2014).
 Id. at 1748. The percentage of ACA-related rulemaking climbs to 15.8% if pages of rulemaking are counted instead of rulemaking projects. Id. at 1750.