In the first installment in this series, I explained that “[e]ven though the [Affordable Care Act’s] penalty is [now] set to $0, the individual mandate still plays an important social function to reduce adverse selection, and plays a role in the operation of employer-based coverage.” Indeed, according to a November 2017 report from Congressional Budget Office and the Joint Committee on Taxation, “with no penalty at all, only a small number of people who enroll in insurance because of the mandate under current law would continue to do so solely because of a willingness to comply with the law.” The number is no doubt “small,” but it is not zero. The mandate still has force, even if no penalty accompanies it.

As promised, this second installment “will focus on how the recent tax reform legislation affects Chief Justice Roberts’s saving construction in NFIB v. Sebelius.” It will also discuss the Attorney General’s decision not to defend the constitutionality of the individual mandate. The third installment will address the issue of severability: if the mandate is now unconstitutional, should other portions of the law also be set aside? Part III will also discuss the Attorney General’s decision concerning severability.

The Saving Construction, Revisited

Pursuant to the recent tax legislation, starting in 2019, the Affordable Care Act’s penalty (codified at 26 U.S.C. § 5000A(c)) will be set to $0. It did not amend the individual mandate (codified at § 5000A(a)). My prior post illustrated that the Affordable Care Act distinguishes between the mandate and the penalty. Merely zeroing out the penalty does not rescind the mandate. Yet, Congress’s decision to zero out the mandate does change the predicate facts that gave rise to Chief Justice Roberts’s saving construction. This post will carefully parse Part III of Chief Justice Roberts’s controlling opinion in NFIB v. Sebelius.

In Part III.A.1, the Chief Justice concludes that the individual mandate “cannot be sustained under a clause authorizing Congress to ‘regulate Commerce.’” In Part III.A.2, the Chief Justice concludes that the mandate cannot be “upheld as a ‘necessary and proper’ component of the insurance reforms.” However, “[t]hat is not the end of the matter.” In Part III.B, the Chief Justice considers if “the mandate may be upheld as within Congress’s enumerated power to ‘lay and collect Taxes.’” He posits that “if the mandate is in effect just a tax hike on certain taxpayers who do not have health insurance, it may be within Congress’s constitutional power to tax.” Yet, “[t]he most straightforward reading of the mandate is that it commands individuals to purchase insurance,” and the penalty is not a tax. Still, that observation was not the end of the matter.

In Part III.C., the Chief Justice develops the so-called “saving construction.” First, he notes that “[t]he exaction the Affordable Care Act imposes on those without health insurance looks like a tax in many respects.” The Chief Justice then lists three ways in which the “exaction”—that is, the penalty—resembles a tax.

1. “The ‘[s]hared responsibility payment,’ as the statute entitles it, is paid into the Treasury by ‘taxpayer[s]’ when they file their tax returns. 26 U.S.C. § 5000A(b).”
2. “For taxpayers who do owe the payment, its amount is determined by such familiar factors as taxable income, number of dependents, and joint filing status. §§ 5000A(b)(3), (c)(2), (c)(4).
3. “This process yields the essential feature of any tax: It produces at least some revenue for the Government. . . . Indeed, the payment is expected to raise about $4 billion per year by 2017.”

Starting in 2019, the “shared responsibility” will be set to $0. As a result, these three elements can no longer be satisfied. First, because no future penalties will be assessed, no new money will have to be paid into the treasury. Second, the formulas used to calculate the penalty will become irrelevant. This is true even if some taxpayers defer payment of any penalties assessed in 2018 to 2019, or later. Third, the imposition of the mandate can no longer generate new revenue for the government. Again, this is true, even if some taxpayers defer payment of any penalties assessed in 2018 to 2019, or later. Because these three conditions are not satisfied, the predicate for the saving construction topples. Starting in 2019, none of the factors identified by the Chief Justice will be satisfied. Therefore, the remainder of Part III.C is no longer relevant. Or, to put it differently, Part III.C. has now been severed from the opinion.

At the conclusion of Part III.D, the Chief Justice succinctly explains why the saving construction can no longer hold:

The Federal Government does not have the power to order people to buy health insurance. Section 5000A would therefore be unconstitutional if read as a command. The Federal Government does have the power to impose a tax on those without health insurance. Section 5000A is therefore constitutional, because it can reasonably be read as a tax.

Because Section 5000A can no longer “reasonably be read as a tax,” then Section 5000A is “unconstitutional” because it “read[s] as a command.” We are left with “[t]he most straightforward reading of the mandate . . . [which] commands individuals to purchase insurance.”

New Jersey’s Defense of the Constitutionality of the Individual Mandate

New Jersey and a host of other states intervened in the litigation to defend the constitutionality of the mandate, and to oppose Texas’s severability argument. (I will address the latter point in my next post.) The brief argues that “[c]ontinuous production of revenue is not a constitutional requirement for a tax, and the minimum coverage requirement will continue to produce revenue for years to come.” In other words, the states contend that the saving construction can still save the mandate, even though the penalty has been set to $0.

First, New Jersey asserts that Section 5000A “still maintains the tax-like features identified in NFIB.” New Jersey does not dispute the three factors I noted above that allowed the Chief Justice to construe the penalty as a tax. Rather, New Jersey cites a fourth criterion:

The fact that the shared responsibility payment raised revenue was just one of several factors that caused it to resemble a tax, and the generation of revenue was not central to the Court’s constitutional determination. The Court noted that “[a]lthough the payment will raise considerable revenue, it is plainly designed to expand health insurance coverage,” which is a perfectly valid exercise of Congress’s taxing powers.

Respectfully, the quoted sentence in this passage did not concern whether a penalty could be construed as a tax. Rather, the Chief Justice was addressing whether a payment intended to “affect individual conduct,” rather than raise revenue, could still be considered a tax. The controlling opinion answered that it could. But this argument doesn’t help New Jersey. All of the exactions cited by the Chief Justice raised revenue as the means to “affect individual conduct.” In other words, people modified their conduct to avoid having to pay extra money to the government. For example, “federal and state taxes can compose more than half the retail price of cigarettes, not just to raise more money, but to encourage people to quit smoking.” Some people will quit smoking to avoid having to pay the taxes, but even those who continue smoking will pay the tax. The Chief does not discuss whether a penalty of $0 could still constitute a tax, for such an exaction is without any precedent.  (Inadvertently, perhaps, New Jersey makes an argument that I advanced in my first part in this series: the mandate, even without a penalty, could still “expand health insurance coverage.”)

Second, New Jersey contends that “if all non-exempt taxpayers made the ‘financial decision’ to purchase insurance, the provision would not raise any revenue whatsoever.” That is, if 100% of taxpayers who were subject to the mandate purchased qualifying insurance, then the penalty would generate $0 in revenue. This argument is inconsistent with the Chief Justice’s saving construction, which was premised on a forecast that “the payment is expected to raise about $4 billion per year by 2017.” Had the Chief Justice concluded that the penalty would generate $0 in revenue, because everyone would buy insurance, then his saving construction would have prematurely toppled.

Third, New Jersey posits that an exaction can still be a tax, even if it does not raise revenue right away.

The production of revenue at all times is a not a constitutional requirement for a tax to be lawful. Congress routinely enacts taxes with delayed effective dates and/or taxes that may not raise revenue in all calendar years, including numerous examples found in the ACA itself such as the so called “Cadillac Tax,” the Medical Device Tax, and the Health Insurance Providers Tax. The shared responsibility payment has now joined that list of ACA taxes for which Congress has suspended collection, but retains the option of increasing in future years. The shared responsibility payment has not been rendered unconstitutional merely because it will be $0 in 2019.

New Jersey did not mention the most obvious example of an exaction that failed to raise revenue right away: the shared responsibility payment! Though enacted in 2010, the Affordable Care Act’s penalty would not be assessed until 2014. Indeed, when NFIB was decided in 2012, no one had yet paid the penalty. Only through a strained application of the Tax Anti-Injunction Act was the Court able to adjudicate the case in 2012. Yet, the Chief’s analysis was premised on the fact that under the law, “[b]eginning in 2014, those who do not comply with the mandate must make a ‘[s]hared responsibility payment’ to the Federal Government.” The same analysis applies to the Cadillac tax or the Medical Device Tax. Under current law, payments for these taxes will come due at a date certain in the future. For example, CBO forecasts concerning the ACA rely on expected revenue to be generated by the Cadillac Tax and Medical Device Tax.  With respect to the mandate, however, no new revenue can be generated in 2019, 2020, 2021, and so on.

Yet, this argument does present something of a riddle. If the Supreme Court finds the mandate unconstitutional because it no longer collects revenue, and Congress subsequently restores the penalty, is the individual mandate resurrected as constitutional? We would have something akin to Schrödinger’s cat, where the mandate is simultaneously constitutional and unconstitutional. This situation doesn’t trouble me for reasons discussed in Jonathan Mitchell’s new article, The Writ of Erasure Fallacy. The Supreme Court cannot, and does not actually strike down laws. The mandate would remain on the books, and if Congress restored the penalty, the Chief’s saving construction would once again apply.

New Jersey’s fourth fares no better. The states contends that even though the tax will not be assessed in 2019, because some people defer their payments, the mandate will in fact continue to generate revenue in 2019 and beyond:

The shared responsibility payment will yield revenue for the federal government in the range of $3 to $5 billion for 2018, based on the most recent data available. And much of that revenue will flow into the federal government’s coffers after April 15, 2019. Like other taxes, the IRS may collect on any unpaid penalty from 2018 (or prior years) via offsets under 26 U.S.C. § 6402(a). And approximately 26% of individuals do not file their taxes on time, underreport their assets, or pay too little tax when they initially file.21 Accordingly, the federal government will likely continue to collect shared responsibility payments owed from 2018 until 2020 or beyond. The shared responsibility payment will thus “produce at least some revenue for the Government” long after January 1, 2019. NFIB, 567 U.S. 564. Therefore, even if Plaintiffs’ theory were legally sound, the Court could not enjoin the minimum coverage requirement until it ceased producing any revenue for the government several years down the road.

Penalties assessed in 2018 may not be paid in 2019. Indeed, that money could trickle into federal coffers in 2020, 2021, 2022, and so on, depending on how individuals structure their returns. Furthermore, the Affordable Care Act is somewhat unique in that the government cannot use traditional collection devices, like a notice of lien or a levy. As a general matter, penalties that are due can only be deducted from a tax refund. If taxpayers who are owe the penalty are never due a refund, the government will never collect the penalty. And this theory is not limited to penalties assessed in 2018. Any penalty assessed from 2014 through 2018 could remain outstanding in perpetuity. Under New Jersey’s theory, every repealed tax—not just the ACA—could “produce at least some revenue for the Government” indefinitely.

This argument is not consistent with the Chief’s opinion. First, the saving construction is premised on when the “shared responsibility” is “paid into the Treasury by ‘taxpayer[s]’ when they file their tax returns.” It is a fair reading of Roberts’s opinion that he did not have in mind, as New Jersey speculates, “individuals [who] do not file their taxes on time, underreport their assets, or pay too little tax when they initially file.” Second, Roberts referenced the fairly complicated formula used to calculate the penalty in Section 5000A(b)(3). That provision turns on whether a taxpayer does not have qualified insurance “for any month” in a “taxable year.” It is a fair reading of Roberts’s opinion that he did not have in mind people who are assessed a penalty in 2018, but choose to pay it in 2019, 2020, 2021, etc. Third, Roberts states “[t]his process yields the essential feature of any tax: It produces at least some revenue for the Government.” And that “process” is premised on how the payment is “assess[ed] and collect[ed],” not when  (if ever) it is ultimately paid. No tax has perfect enforcement rates. In his opinion, Roberts cites a CBO report, which forecasted that “the payment is expected to raise about $4 billion per year by 2017.” Page 71 of the report notes that some “individuals will try to avoid making payments.” The fact that some people may never pay the penalty is not enough to save the mandate.

New Jersey buries its fifth, and most potent argument in a footnote:

Although still a lawful tax, in the alternative, the minimum coverage provision may now be sustained under the Commerce Clause. In NFIB, the Court held that the minimum coverage provision exceeded Congress’s Commerce Clause powers because it “compels individuals to become active in commerce by purchasing a product.” NFIB, 567 U.S. at 552. But with a tax of zero dollars, there is no compulsion. The constitutional problem—compelling the purchase of insurance—is no longer present absent any penalty for failing to do so.

I discussed in Part I why this argument does not work: even a mandate, with a penalty of $0, provides some compulsion. Once again, according to the Congressional Budget Office and the Joint Committee on Taxation, “with no penalty at all, only a small number of people who enroll in insurance because of the mandate under current law would continue to do so solely because of a willingness to comply with the law.” The number is no doubt “small,” but it is not zero. Jonathan Adler describes the CBO report in this fashion: “the practical consequence of eliminating the tax penalty is substantially the same as erasing the mandate altogether.” The key word is “substantially.” It’s not exactly the same. This is a weak mandate, for sure, but not a rump mandate. (See generally my analysis in Part I.) For example, in December 2008, CBO explained how social dynamics, beyond monetary fines, support compliance with the law:

Compliance, then, is probably affected by an individual’s personal values and by social norms. Many individuals and employers would comply with a mandate, even in the absence of penalties, because they believe in abiding by the nation’s laws. However, such compliance may also be moderated by perceptions of fairness; individuals may comply more readily if they believe that a mandate is fair and is consistently enforced. If enforcement efforts appear to be unevenly applied, compliance may diminish. Social psychologists find that compliance could be affected not only by personal values but also by individuals’ perceptions of how others will act. Such studies find that many people want to take the popular—as well as the moral—course of action. (pp. 53-54)

Indeed, one of the plaintiffs in the Texas litigation expressed a similar sentiment in his affidavit:

I value compliance with my legal obligations, and believe that following the law is the right thing to do. The repeal of the associated health insurance tax penalty did not relieve me of the requirement to purchase health insurance. I continue to maintain minimum essential health insurance coverage because I am obligated to comply with the Affordable Care Act’s individual mandate, even though doing so is a burden to me.”

This argument may not be a valid injury for purposes of standing–Texas has other grounds to challenge the law–but it does comport with how the mandate was described in NFIB. The mandate still has some force, even if no penalty accompanies it.

Recall that Section 5000A(a) provides that “[a]n applicable individual shall for each month beginning after 2013 ensure that the individual, and any dependent of the individual who is an applicable individual, is covered under minimum essential coverage for such month.” The Court recognized that the most important word in that provision is shall. Chief Justice Roberts noted that “[t]he plaintiffs contend that Congress’s choice of language—stating that individuals ‘shall’ obtain insurance or pay a ‘penalty’—requires reading § 5000A as punishing unlawful conduct, even if that interpretation would render the law unconstitutional.” Under his saving construction, Roberts rejects this argument, because like in New York v. United States, the Court could “sustain[] the charge paid to the Federal Government as an exercise of the taxing power.” Yet, in Part III.B of the opinion, which focuses on the Commerce and Necessary and Proper Clauses, the Chief Justice reached the exactly opposite result.  He stated:

The most straightforward reading of the mandate is that it commands individuals to purchase insurance. After all, it states that individuals “shall” maintain health insurance. 26 U.S.C. § 5000A(a). Congress thought it could enact such a command under the Commerce Clause, and the Government primarily defended the law on that basis. But, for the reasons explained above, the Commerce Clause does not give Congress that power.

Now that the penalty has been zeroed out, and the saving construction cannot hold, we are left with “[t]he most straightforward reading of the mandate.” What is that reading? “[I]t commands individuals to purchase insurance.” And that Congress cannot do pursuant to its powers under the Commerce and Necessary and Proper Clauses. Even if the penalty is $0. Under the most straightforward reading of NFIB, the individual mandate cannot survive constitutional review.

The Attorney General’s Decision Not to Defend the Individual Mandate

On Thursday, Attorney General Sessions informed Congress that DOJ made two decisions. First, the Department “will not defend the constitutionality of 26 U.S.C. 5000A(a).” Second, the Department “will argue that certain provisions of the Affordable Care Act (ACA) are inseverable from that provision.”  For reasons I will discuss in a subsequent post, I do not agree with the Attorney General’s decision concerning severability, but his decision concerning the mandate is defensible.

This is not the first time that Attorney General Sessions has publicly made a decision regarding the constitutionality of a government policy. First, in September 2017, he determined that DACA was implemented “without proper statutory authority” and that this “open-ended circumvention of immigration laws” was “an unconstitutional exercise of authority by the Executive Branch.” He reaffirmed his “duty to defend the Constitution and to faithfully execute the laws passed by Congress.” Critically, he made this decision without the benefit of a decision from the Supreme Court, or any court from that matter, that DACA was unlawful. His analysis only reasoned by analogy from the 5th Circuit’s decision concerning the related DAPA policy. He also disregarded—but did not withdraw—a 2014 opinion from the Office of Legal Counsel, which concluded that DACA was consistent with Article II’s duty to faithfully execute the laws. (Though the Take Care Clause has seen a resurgence of interest after January 20, 2017 as a general restraint on all of President Trump’s actions, this provision concerns the suspension of the law.)

Second, in October 2017, Attorney General Sessions determined that the Affordable Care Act “does not appropriate funds” for certain subsidies. He acknowledged that his predecessor had defended the payments, but “concluded that the best interpretation of the law is that the permanent appropriation” cannot be used to fund the subsidies. This decision that the payments were unconstitutional was supported by the District Court’s decision in House of Representatives v. Burwell, though the D.C. Circuit had not yet opined.

Third, in January 2018, Attorney General Sessions rescinded the so-called Cole Memorandum concerning marijuana enforcement in states that decriminalized the controlled substance. The memorandum itself makes no mention of the Constitution. Though, as my colleague Randy Barnett observed in the Wall Street Journal, “by establishing what amounted to sanctuary status for legal-marijuana states, the Cole memorandum seemed to flout the president’s constitutional obligation to ‘take care that the laws be faithfully executed.’”

Each of these three “departmentalist” decisions were internal, and in no way affected laws enacted by Congress. Though, these decisions affected Congress, indirectly. By rescinding executive actions concerning immigration, the ACA, and controlled substances, the Trump Administration put pressure on Congress to legislate in those areas. Randy made a similar observation in the Journal with respect to the Cole Memorandum: “by eliminating the safe havens mandated by the executive branch, he created a powerful incentive for Congress finally to legislate on the matter.” Likewise, in light of another suit filed by Texas, DACA’s legality may be in jeopardy. If the policy is halted, Congress will be forced to address the issue. Suspensions of the law obviate the need for Congress to legislate. I continue to hope that Congress reaches a comprehensive legislative solution for the Dreamers.

The decision concerning the individual mandate, however, is different, because it directly implicated the failure to defend a law enacted by Congress, rather than executive action. Pursuant to 28 U.S.C. § 530D(a)(B)(ii), the Attorney General must submit a report to Congress if he “determines to refrain (on the grounds that the provision is unconstitutional) from defending or asserting, in any judicial, administrative, or other proceeding, the constitutionality of any provision of any Federal statute, rule, regulation, program, policy, or other law.” Without question, his decision to refrain from defending the constitutionality of Section 5000A(a) warranted a report to Congress.

New Jersey’s arguments in defense of the mandate, though unlikely to prevail, are far from frivolous. Without question, the Department of Justice could have made similar arguments in its defense against Texas’s law suit. But it did not. Sessions explained in his letter that “not every professionally responsible argument is necessarily reasonable in this context, as ‘different cases can raise very different issues with respect to statutes of doubtful constitutional validity,’ and thus there are ‘a variety of factors that bear on whether the Department will defend the constitutionality of a statute.’” Those quotes come from a letter Assistant Attorney General Andrew Fois wrote to Sen. Orrin Hatch in 1996.

Sessions identifies (as far as I can tell) three factors why he concluded that “Section 5000A(a) will be unconstitutional when the Jobs Act’s amendment becomes effective in 2019.” First, he writes that because “Section 5000A(a) will produce no revenue for the Government,” then “the NFIB Court’s saving construction will no longer be available.” His position, and DOJ’s brief, tracks the analysis of the saving construction described above.) Second, he states that his decision “adheres to the Department’s longstanding respect for comity between the three branches of government.” Why? He writes that Congress repealed the mandate with full awareness of the saving construction’s limits. In other words, Congress knew that by repealing the penalty, they rendered the mandate unconstitutional. I doubt that many Republicans cast their vote for the tax bill with Chief Justice Roberts’s opinion in mind, but this is a reasonable presumption. (This presumption, as I will discuss in Part III, does not extend to the argument that members of Congress intended to render other provisions of the ACA unconstitutional.) Third, the Attorney General writes “the Department’s decision not to defend Section 5000A(a)’s constitutionality will not prevent the court in Texas v. United States from resolving the question, given the posture of the case.” That is, because of New Jersey’s successful intervention, the mandate will receive a vigorous defense. Unlike in United States v. Windsor, where it was questionable whether the Bipartisan Legal Advisory Group (BLAG) had standing, here, New Jersey has shown concrete injuries.

These reasons are all valid, but they do not directly address why New Jersey’s arguments may be “professionally responsible,” but are not “necessarily reasonable in this context.” I find the Attorney General’s explication of his own constitutional authority lacking, much like I did with his letter concerning the rescission of DACA.

Allow me to bridge the gap. On June 28, 2012, then-Senator Sessions issued a press release criticizing the Court’s just-released decision in NFIB v. Sebelius:

For the Court to affirm the mandate portion of the law, it was forced to reject the President’s and the then-majority Democratic Congress’ contention that the mandate was not a tax. Under this ruling, the big spenders have once again succeeded in surreptitiously imposing a tax on the American people while pretending they are not. The problem is that the mandate remains a mandate. It remains a demand that Americans purchase a product they do not wish to purchase. I do not believe the central government possesses such a broad power. . . . The majority redefines a provision of the law that declares a mandate to be a tax. I am very troubled by this. Scholars will give great thought to what the Court has done and I am afraid it will be concluded that this is a legal sleight of hand rather than a principled decision. The question has to be asked to what extent all government mandates and demands can just be referred to as a tax, thus unleashing the power of the central government to dictate individual Americans’ private, everyday decisions.

This release expressed what I suspect quickly became an article of faith for every right-of-center attorney: the Chief’s saving construction was “a legal sleight of hand rather than a principled decision.” I suspect when Senator Sessions became Attorney General Sessions, he did not abandon this position. Straining to construe the facts such that the mandate still falls within the saving construction—as did New Jersey—is no doubt “professionally responsible,” but is not “necessarily reasonable in this context” because of the Attorney General’s opinion about the mandate’s underlying constitutionality. The Attorney General’s independent judgment about the Constitution, absent a definitive answer from the courts, undergirded his position on the mandate, much like it did his positions on DACA, the ACA subsidies, and the Cole Memorandum. I don’t doubt that Sessions’s policy views on immigration, Obamacare, and marijuana informed these decisions. But as Attorney General, he has the authority to issue such constitutional judgments.

This analysis, however, does not apply to the Attorney General’s decision concerning severability, which I will address in my next post.