The Yale Law Journal


Reading the ACA’s Findings: Textualism, Severability and the ACA’s Return to the Court

02 Oct 2020

abstract. The Affordable Care Act (ACA) is back in the Court, but challengers’ “textualist” arguments are not textualist at all. They argue a findings section in the ACA is an “inseverability clause,” meaning that if the insurance mandate is eliminated, the whole ACA goes down with it. They argue this despite the fact that those findings are specific to one subsection, of one part, of one subtitle, of the ten-title law, and despite the fact that the language they seize on is boilerplate language that Congress often uses not for severability, but to justify its commerce power. Congress expressly tells us that too in the ACA’s text. Challengers argue that Congress has explicitly spoken to the issue, but Congress’s established drafting practices, and examples throughout the U.S. Code, make clear that when Congress actually writes an inseverability clause, it is unmistakably explicit about it and uses specific language absent from the ACA. Inseverability is a nuclear bomb. Congress doesn’t hide it in mouseholes. There is a burgeoning movement to better understand how Congress drafts laws, one as relevant for textualists as anyone else.


“(1) Read the statute; (2) read the statute; (3) read the statute!”

- Felix Frankfurter1

Textualism had a big term this year in the Supreme Court. From major decisions on sexual-orientation discrimination to Native American reservations,2 Justice Scalia’s imprint on the Court endures.

Enter, once more, the Affordable Care Act (ACA). As one of the most significant statutes of the modern era, it is no surprise that the ACA keeps pulling us into the statutory weeds. This fall, the ACA returns to the Supreme Court for the seventh time,3 in California v. Texas; it is the second time the Court has been asked to invalidate the entire 2,000-page law.4 This time, the law’s challengers are trying to use a false textualism to implode it. Specifically, they argue that Commerce Clause findings that appear in one subsection, of one part, of one title, of the ten-title ACA do double duty as an explicit “inseverability clause” that requires the entire ACA to be struck off the books if one provision is eliminated.

The challengers’ textualism is not real textualism: it ignores the ACA’s statutory organization, the words of the ACA itself, and Congress’s consistent drafting practices across the U.S. Code. The Court’s textualists can do better.

In California, the Court is being asked what it should do with the rest of the ACA if the Court finds the individual insurance-purchase mandate—the ACA’s requirement that most Americans obtain health insurance—unconstitutional. The question has arisen because Congress included a provision in the 2017 Tax Cuts and Jobs Act amending the ACA in one, single way: Congress reduced the tax penalty for noncompliance with the mandate to zero.5 Because Chief Justice Roberts’s opinion in the first existential challenge to the ACA, National Federation of Independent Business v. Sebelius (NFIB) v. Sebelius, held that the mandate could be construed as a legitimate exercise of Congress’s taxing power, but not its commerce power,6 Texas and seventeen other states now argue that without a tax penalty, the mandate can no longer be construed as a tax. They argue that if the mandate is not a tax, it therefore becomes an exercise of the commerce power, and so it is unconstitutional. But then they go further and argue that the mandate is so critical to key provisions of the law that if the mandate is unconstitutional, the entire 2,000-page law must fall with it. They rely on a “textualist” argument about the statutory findings that is not really textualist at all.

The case should be easy, although nothing ever is with the ACA. For decades, the Court has applied an established, apolitical, and uncontroversial interpretive doctrine: the “severability doctrine” presumes that unless Congress clearly indicates otherwise, Congress intends the remainder of a statute to stand if one provision is held invalid. Members of the Court reaffirmed these “ordinary severability principles” twice this past term, calling the doctrine “a strong presumption”7 and describing the test:

[W]e try to limit the solution to the problem, severing any problematic portions while leaving the remainder intact. . . . We will presume that Congress did not intend the validity of the statute in question to depend on the validity of the constitutionally offensive provision unless there is strong evidence that Congress intended otherwise.8

To use an analogy employed most recently by both Chief Justice Roberts and Justice Kavanaugh, the doctrine is a “scalpel rather than a bulldozer.”9 “The Court’s precedents reflect a decisive preference for surgical severance rather than wholesale destruction, even in the absence of a severability clause.”10

As I have detailed elsewhere, California is a simple and straightforward severability case: the 2017 Congress itself eliminated the penalty but left the rest of the ACA intact—leaving an explicitly enacted, text-based congressional indication for the rest of the statute to survive, expressed as clearly as it possibly could be by the continued existence of the ACA itself. That should be the end of it.

This Essay devotes little attention to the severability doctrine. My focus here instead is on the second-tier argument on which the challengers have now decided to hang their hats—an argument that the Court, knowing the 2017 Congress clearly left the rest of the ACA standing, should never have to reach. The challengers seize on a few words in one subparagraph of the ACA’s statutory findings taken entirely out of their location and context in the law. They now argue that those words are an explicit “inseverability clause” that applies to the statute as a whole11 and trumps the presumption of severability—meaning that if the mandate is eliminated, the whole ten-title ACA goes down with it.

The challengers argue this despite the fact that those findings are specific to one subsection, of one part, of one subtitle, of the ten-title law. They also argue this despite the fact that the language they seize on is boilerplate language that Congress has used in scores of other statutes, not for the purpose of severability, but to justify Congress’s commerce power. They argue this even though Congress expressly tells us, in the subsection itself and also in the subsections directly above and below it, that the findings are indeed directed at establishing congressional authority under the Commerce Clause. And, they argue this even though, reading the entire subsection literally as they ostensibly would have us do, it would mean that not only the ACA goes down but also that the nation’s entire pensions and employee benefits regulatory system—the Employee Retirement Income Security Act of 1974 (ERISA) statute—goes down too.

Finally, and most importantly, they argue that Congress has actually spoken to the issue, even though Congress’s established drafting practices, substantiated by its drafting manuals and enacted examples throughout the U.S. Code, make clear that when Congress actually writes an inseverability clause, it is unmistakably explicit about it and writes with specific language. Congress used none of that language in the ACA.

The Court deploys a “strong presumption” of severability because striking down whole statutes is the most invasive and destructive of all statutory-case remedies. Congress mirrors that presumption by inserting explicit inseverability clauses in only a small number of select statutes, because of course Congress does not generally hope that all of its work product will be struck down. Inseverability is a nuclear bomb. Congress does not hide it in mouseholes.12

During the first existential challenge to the ACA, National Federation of Independent Business v. Sebelius (NFIB), Justice Scalia compared a cover-to-cover read of the ACA to cruel and unusual punishment, asking: “You really want us to go through these 2,700 pages? And do you really expect the Court to do that?”13

Well, yes. You are either a textualist or you are not. “Honest textualists,”14 to use another Scalia-ism, look at statutory organization, including provision placement. They look at all the words in context. They look at how Congress has drafted similar provisions in other statutes. They look for evidence that Congress knows how to speak clearly when it wants to make the point in question and that it has done so elsewhere. They make sure that Congress expressly said what the lawyers claim it expressly said. This Essay discusses each of those points in turn.

There is a burgeoning movement among legal scholars and jurists of all interpretive stripes to better understand how Congress drafts laws.15 Those developments, this analysis should make clear, are as relevant, if not more relevant, for textualists as for anyone else.

I. why are the findings even at issue?

Together with others, including some of the ACA’s most prominent former legal opponents,16 I have detailed the many shortcomings of the challengers’ application of severability doctrine. It is true that established severability doctrine, textualism notwithstanding, sometimes requires courts to hunt for crumbs of hypothetical congressional intent to determine if Congress would have wanted the statute to stand. That kind of inquiry can be especially uncomfortable for textualists and is one reason textualist members of the Court this term reaffirmed that the Court presumes Congress intended the statute to survive unless Congress explicitly indicates otherwise.17

But hypothesizing about congressional intent, as the challengers attempt to do with the findings, is supposed to happen only where Congress has not clearly expressed itself. Here, the Court should not ever need to reach any of the arguments about the findings.

There are serious democracy concerns here. There is no evidence that Congress’s repeal of the mandate’s penalty was a surreptitious effort to render the entire law unconstitutional and implode it. All evidence points to the contrary. And the Court never presumes Congress legislates to destroy; and it always presumes Congress legislates constitutionally.18 Moreover, to strike down the entire ACA after Congress itself tried and failed more than seventy times to repeal it and then concluded in 2017 that it could only eliminate the penalty—as Leader McConnell himself said, “we obviously were unable to completely repeal and replace”19—would not only violate severability doctrine; it would be a usurpation of the clearly expressed legislative prerogative. It would also retroactively superimpose an inseverability clause on the ACA that the 2017 Congress did not know existed, and clearly assumed did not exist, when it amended the law.

Another democracy point: the Court’s duty is to give proper effect to Congress’s actions in 2017; whatever any earlier Congress said or thought has no relevance now that a later Congress has acted. The challengers’ argument, because it focuses on the 2010 Congress, violates the basic principle that the Constitution prohibits entrenching the views of an earlier Congress over a later one, and always allows Congress to change its mind.20

While there is more to be said about severability, including some brewing opposition to aspects of the doctrine among two members of the Court21 (and also the question whether a penalty-less mandate is still a tax, a point on which this Essay takes no position22), the rest of this Essay focuses on the findings.

Given the clear evidence that the ACA can indeed function without an enforced mandate—evidence that the 2017 Congress relied on and that continues to grow since 2017—the challengers have mostly abandoned their earlier arguments that the ACA is unworkable without a mandate. All that is left is a thin statutory reed—a findings subparagraph in the ACA as originally enacted in 2010 that serves to justify the mandate under the Commerce Clause. And, unlike actual inseverability clauses in the U.S. Code, the findings subparagraph says nothing explicit about inseverability and uses none of the boilerplate language that Congress uses every single time it writes a real inseverability clause.

II. the findings: statutory location and organization, words in context, and established drafting practices throughout the u.s. code

The challengers’ main argument is that language from the mandate’s findings serves as an explicit “inseverability clause.” They argue this despite three central tenets of textualism that have applicability far beyond the ACA: textual location matters; words matter; and Congress’s consistent drafting practices across the U.S. Code matter. I introduce these arguments here and then detail them in turn.

Location Matters

Our textualist Court puts a heavy premium on statutory organization and provision placement. Sometimes Congress’s decisions on a provision’s location, as indicated in the public law, survive into the U.S. Code. Other times Congress’s nonpartisan codifiers rearrange statutory provisions after enactment, pulling public laws apart and reorganizing provisions into the fifty-four subject-matter titles of the U.S. Code. The Court has paid scant attention to this difference, but in this case, the placement of the ACA’s mandate findings in the statute was decided by Congress and is especially significant.23

The findings at issue are not the kind of general statutory findings that Congress commonly places at the beginning of a law. Rather, the ACA’s mandate findings are specific to one subsection. They are buried at the end of Title I of the ten-title ACA; they sit in the penultimate of seven separate subtitles of Title I and do not even apply to all of that subtitle, the second half of which, in a separate part, deals with the entirely distinct employer mandate.24 All of the other private insurance reforms sit in different subtitles before the one containing the findings, and the public insurance reforms, including reforms to Medicare and Medicaid, come after it, in different titles, subtitles, and parts of the law. The findings invoked by the ACA’s opponents are sandwiched between two paragraphs that explicitly state that the findings are there to “describe[]”25 how the mandate is “commercial and economic in nature, and substantially affects interstate commerce.”26 The ACA has two other, different subsection-specific findings elsewhere in the ten titles, further indicating that each findings subsection is local in its effect.27 Given this textual structure, no textualist should read the findings as having any operative effect on other sections of the law or beyond their limited descriptive use in the subsection.

Words Matter

The words used in the findings have nothing to do with severability; they are specific to Congress’s Commerce Clause power. They not only say so explicitly but are patterned after numerous other findings in the U.S. Code that likewise use the terms “essential to” and “markets”—words the challengers now argue connote inseverability—to justify Congress’s authority under the commerce power.

The words in the findings precisely track the instructions in Congress’s drafting manuals about when findings should be used and written. The drafting manuals discourage findings in general but note they can be useful to clarify or head off legal disputes and single out Commerce Clause findings as the primary example of when they should be used and how they should be drafted.28 The other two findings sections in the ACA are likewise responsive to (different but specific) legal and statutory questions and likewise do not have broader applicability. Of course, any findings about the commerce power are moot after NFIB, which invalidated Congress’s Commerce Clause justification.29 And the findings never use the words “inseverability,” “invalid,” or any similar term.

Even more eye opening, the words in the findings themselves say the mandate is essential to Congress’s overall regulation of insurance, including statutes outside the ACA.30 Read as an actual inseverability directive, under the challengers’ argument, the findings therefore would mean that ERISA (the 1974 law that governs millions of employer-provided benefit and pension plans) and the Public Health Service Act (which, among many other things, gives Health and Human Services public-health emergency authority) must also fall with the mandate—an absurd result.

Established Congressional Practice, Evidenced by Other Enacted Statutes, Matters

Statutes throughout the U.S. Code make clear that when Congress does speak about inseverability, it is explicit. Congress uses specific words and breaks the point off into a separate section or subsection or paragraph, clearly marked. The House and Senate drafting manuals instruct against the inclusion of express severability clauses, citing the Court’s own stated presumption in favor of severability.31 But the manuals specify the need for explicit inseverability clauses for the same reason.32 They also instruct that inseverability clauses should be explicit about what is not to be severed if partial severability is desired.33 The inseverability clauses throughout the U.S. Code follow this direction. All of them, moreover, use the same language—including all using the word “invalid,” itself patterned in the manuals—language entirely absent from the ACA.34

The Court’s textualist statutory interpretation jurisprudence has held, over and over—including several times this past term—that when Congress shows it knows how to say something explicitly, the Court will not imply the same directive elsewhere if Congress has not likewise been explicit.35

* * *

It should be emphasized that the challengers are not arguing the ACA is not functional without a mandate. Sometimes, litigants make severability arguments using unworkability as a proxy for congressional intent. No one is doing that here because they cannot do so on the facts; the ACA is functioning (indeed, playing a critical role during the COVID-19 pandemic) and Congress had ample evidence before it in 2017 that the statute could function without a mandate. So instead, the ACA’s challengers must argue that Congress has spoken; that Congress has said the mandate cannot be severed and that those words now must be followed, come what may, even if the 2017 Congress did not realize it was destroying the law. So, the precise question is whether these findings are really how Congress speaks about inseverability.

Congress does not hide inseverability in haystacks and does not give courts the power to implode laws with implicit language. Justice Scalia himself once complained that the Court “changes the usual rules of statutory interpretation for the sake of the Affordable Care Act.”36 The point goes both ways.

A. The Subparagraph at Issue

The ACA has ten titles. Title I alone has seven subtitles, most with multiple parts and subparts. At the end of Title I, in the second to last subtitle, Subtitle F, in Part I of two parts of that subtitle, there is the findings subsection in question, 1501(a). It is important to read that subsection in full.37 Underlined are express references to the commerce power and effects on markets. Bolded are the provisions on which the challengers rely, with most of their argument and citations focused on Subparagraph (I). Bolded Subparagraphs (H), (I), and (J), which the challengers cite, also reference commerce effects on the national market.

(a) Findings.—Congress makes the following findings:

(1) In General.—The individual responsibility requirement provided for in this section (in this section referred to as the “requirement”) is commercial and economic in nature, and substantially affects interstate commerce, as a result of the effects described in paragraph (2).

(2) Effects on the National Economy and Interstate Commerce.—The effects described in this paragraph are the following:

(A) The requirement regulates activity that is commercial and economic in nature: economic and financial decisions about how and when health care is paid for, and when health insurance is purchased. In the absence of the requirement, some individuals would make an economic and financial decision to forego health insurance coverage and attempt to self-insure, which increases financial risks to households and medical providers.

(B) Health insurance and health care services are a significant part of the national economy. National health spending is projected to increase from $2,500,000,000,000, or 17.6 percent of the economy, in 2009 to $4,700,000,000,000 in 2019. Private health insurance spending is projected to be $854,000,000,000 in 2009, and pays for medical supplies, drugs, and equipment that are shipped in interstate commerce. Since most health insurance is sold by national or regional health insurance companies, health insurance is sold in interstate commerce and claims payments flow through interstate commerce.

(C) The requirement, together with the other provisions of this Act, will add millions of new consumers to the health insurance market, increasing the supply of, and demand for, health care services, and will increase the number and share of Americans who are insured.

(D) The requirement achieves near-universal coverage by building upon and strengthening the private employer-based health insurance system, which covers 176,000,000 Americans nationwide. In Massachusetts, a similar requirement has strengthened private employer-based coverage: despite the economic downturn, the number of workers offered employer-based coverage has actually increased.

(E) The economy loses up to $207,000,000,000 a year because of the poorer health and shorter lifespan of the uninsured. By significantly reducing the number of the uninsured, the requirement, together with the other provisions of this Act, will significantly reduce this economic cost.

(F) The cost of providing uncompensated care to the uninsured was $43,000,000,000 in 2008. To pay for this cost, health care providers pass on the cost to private insurers, which pass on the cost to families. This cost-shifting increases family premiums by on average over $1,000 a year. By significantly reducing the number of the uninsured, the requirement, together with the other provisions of this Act, will lower health insurance premiums.

(G) 62 percent of all personal bankruptcies are caused in part by medical expenses. By significantly increasing health insurance coverage, the requirement, together with the other provisions of this Act, will improve financial security for families.

(H) Under the Employee Retirement Income Security Act of 1974 (29 U.S.C. 1001 et seq.), the Public Health Service Act (42 U.S.C. 201 et seq.), and this Act, the Federal Government has a significant role in regulating health insurance. The requirement is an essential part of this larger regulation of economic activity, and the absence of the requirement would undercut Federal regulation of the health insurance market.

(I) Under sections 2704 and 2705 of the Public Health Service Act [42 U.S.C. 300gg-3, 300gg-4] (as added by section 1201 of this Act), if there were no requirement, many individuals would wait to purchase health insurance until they needed care. By significantly increasing health insurance coverage, the requirement, together with the other provisions of this Act, will minimize this adverse selection and broaden the health insurance risk pool to include healthy individuals, which will lower health insurance premiums. The requirement is essential to creating effective health insurance markets in which improved health insurance products that are guaranteed issue and do not exclude coverage of pre-existing conditions can be sold.

(J) Administrative costs for private health insurance, which were $90,000,000,000 in 2006, are 26 to 30 percent of premiums in the current individual and small group markets. By significantly increasing health insurance coverage and the size of purchasing pools, which will increase economies of scale, the requirement, together with the other provisions of this Act, will significantly reduce administrative costs and lower health insurance premiums. The requirement is essential to creating effective health insurance markets that do not require underwriting and eliminate its associated administrative costs.

(3) Supreme Court Ruling.—In United States v. South-Eastern Underwriters Association (322 U.S. 533 (1944)), the Supreme Court of the United States ruled that insurance is interstate commerce subject to Federal Regulation.38

The crux of the challengers’ argument is that, in Subparagraph (I), Congress called the mandate “essential to creating effective health insurance markets in which improved health insurance products that are guaranteed issue and do not exclude coverage of pre-existing conditions can be sold.”39 The challengers rely in part on the argument of the Obama-era Department of Justice (DOJ) in NFIB, that that same phrase was evidence that Congress would not want the mandate invalidated without also invalidating two other provisions of the statute, namely, new insurance provisions prohibiting medical underwriting and discrimination based on preexisting conditions (which are laid out in Subtitle C, not Subtitle F, where these findings are).40 The challengers argue that because Congress did not repeal this findings subsection when it rendered the mandate unenforceable in 2017, the 2017 Congress reaffirmed its agreement with what the challengers claim is now an explicit “inseverability clause” in the form of this subparagraph enacted in 2010.

As I wrote before NFIB, the Obama DOJ was wrong to make its inseverability argument.41 But even the 2012 DOJ position did not say that Congress was trying to write inseverability into the law with the mandate’s findings. The DOJ position was that, even though the Commerce Clause findings were not an actual inseverability clause, the findings helped to answer a difficult evidentiary question back then as to what Congress would have wanted to do with the rest of the ACA if the mandate had been struck down.

The findings were to be used as some evidence of congressional intent in 2012 because settled severability doctrine asks courts to determine what Congress would have wanted if Congress’s views are not clear (one reason Justice Thomas frowns on the doctrine as “a nebulous inquiry into hypothetical congressional intent”).42 In 2012, there was some dispute as to what the 2010 Congress would have wanted to do with the ACA without a mandate.

That is not the case now. And that is not the way the challengers are now using the findings. That makes them doubly irrelevant. The 2017 Congress’s intent constitutionally supersedes the 2010 Congress’s intent. And the 2017 Congress’s intent as to the rest of the ACA is clear from Congress’s own actions in leaving the ACA standing after it defanged the mandate; there is no need for the atextual inquiry into “hypothetical congressional intent” that some members of the Court have bemoaned in severability cases and that Justice Kavanaugh recently deemed “imaginative reconstruction” and an “analytical dead end.”43 Although the Court sometimes turns to a functionality inquiry when considering severability—and the findings may have been illustrative in 2012 for such purposes—that kind of inquiry is only a proxy for congressional intent when Congress does not itself express its view, as the 2017 Congress did, about the continued existence of the law.

Of course, the 2017 Congress was allowed to change its mind from the 2010 Congress. The 2017 Congress amended the law to neuter the mandate with evidence from the Congressional Budget Office and others that the ACA could function without a mandate.44 The 2017 Congress did not need new findings to do so. The 2017 Congress amended the ACA through its taxing power, and the Court has not asked for findings in the context of taxation. Regardless there can be no doubt that the law continues to function.45

But the critical point is that the challengers are not arguing functionality; they are not arguing the ACA is unworkable without a mandate or that the findings are a proxy for unknown congressional intent. They are arguing that Congress explicitly directed, with this subparagraph, that the mandate is not severable. There is no plausible reading of the findings’ text, structure or context in the U.S. Code that supports this view.

B. Location, Location, Location

Textualists from Justice Scalia to Dean John Manning have emphasized that honest textualists do not take words out of context.46 Justice Scalia wrote there is “no greater interpretative fault than the failure to . . . consider the entire text, in view of its structure and of the physical and logical relation of its many parts.”47 When it comes to legal interpretation, as Manning has written, statutory context importantly includes “specialized conventions and linguistic practices peculiar to the law.”48

Every Justice on the Court routinely relies on location—where a provision is placed relative to other provisions in the statute—and statutory organization as aids to interpretation. There are myriad examples. To offer just a few from recent cases, in Florida Department of Revenue v. Piccadilly Cafeterias, Inc.,49 a bankruptcy case, Justice Thomas, writing for the Court, relied in part on the location of a stamp tax exemption in a statutory chapter50: “We find it informative that Congress placed § 1146(a) in a subchapter entitled, ‘POSTCONFIRMATION MATTERS.’ . . . The placement of § 1146(a) . . . undermines Piccadilly’s view that § 1146(a) covers preconfirmation transfers.”51 In Yates v. United States,52 in which the question was whether a fish was a “tangible object” for purposes of the evidence-destruction provisions of the Sarbanes-Oxley Act,53 a textualist plurality opinion by Justice Ginsburg relied on U.S. Code placement, emphasizing that the provision in question followed provisions specific to the corporate context, not generally-applicable prohibitions.54In another recent case, Justice Alito, in holding that the Wartime Suspension of Limitations Act (WSLA) applies only to criminal, not civil, charges, relied in part on the placement of WSLA in Title 18 of the U.S. Code,55 which concerns “Crimes and Criminal Procedure.”56

Cases from just last term offer more examples. In Maine Community Health Options v. United States, another ACA case, Justice Sotomayor wrote for eight Justices that “Section 1342’s adjacent provisions also underscore” the Court’s interpretation of the text as requiring insurers to be paid.57 In Seila Law, the Chief Justice wrote (indeed, of a severability provision there): the language unmistakably references ‘any provision of this Act.’ And it appears in a logical and prominent place, immediately following the Act’s title and definitions sections, reinforcing the conclusion that it applies to the entirety of the Act.”58

My own work on the legislative process with Jesse Cross has detailed various considerations about utilizing statutory placement in this manner.59 Sometimes Congress decides where provisions are placed before they are enacted and they remain there. Other times, Congress’s nonpartisan codifying office, the Office of Law Revision Counsel, rearranges provisions within the fifty-four titles of the U.S. Code after they are enacted. There is a plausible argument (the nuances of which and counterarguments to Cross and I discuss) that such post-hoc arranging is inferior textualist evidence of congressional meaning compared to the arranging work done prior to enactment by the original drafters of the statute.

The findings subsection of the mandate is one of those instances in which the provision’s location was dictated by Congress before enactment and so provides particularly compelling evidence of the confined reach of its effect.

1. The ACA’s Structure and Organization

Let’s look at the ACA’s ten titles. Title I concerns private insurance reforms, including the establishment of the new state insurance marketplaces. Title II is called “the role of public programs,” and largely deals with Medicaid and the Children’s Health Insurance Program. It does not concern private insurance. Title III concerns delivery system reforms, including some new Medicare initiatives. Title IV covers public health. Title V covers health care workforce. Title VI concerns fraud, transparency, and enforcement. Title VII covers access to medicines including biologics. Title VIII is the now defunct “Class Act” for long-term care. Title IX is fees and revenue provisions. Title X contains a variety of miscellaneous amendments made to sections throughout the text.60

The statutory structure provides important information. First, the ACA does not include a general statement of findings at the top of the entire Act, as many statutes do.61 Instead, the ACA uses section-specific findings embedded within separate subtitles, parts, and subsections of the law.

Title I alone is divided into seven subtitles. The first five detail the private insurance reforms, including increased benefits, subsidies, and the establishment of the new insurance exchanges. Subtitle A is entitled “Immediate Improvements in Health Care Coverage for All Americans” and concerns provisions that took effect quickly, like the prohibition on coverage rescissions and extending dependent coverage to adult children up to age twenty-six. Subtitle B is entitled “Immediate Actions to Preserve and Expand Coverage” and includes reinsurance provisions and establishes a temporary, transitional high-risk pool until the law’s broader protections take effect. Subtitle C is entitled “Quality Health Insurance Coverage for All Americans PART I” and enacts market reforms, including the ban on discrimination based on preexisting conditions and the requirement that each plan provide essential health benefits. Subtitle D is entitled “Available Coverage Choices for All Americans” and launches the new health insurance marketplaces. Subtitle E is entitled “Affordable Coverage Choices for All Americans” and concerns tax credits and cost-sharing reductions, payments the Act offers to make health insurance more affordable.

The findings in question do not appear until Subtitle F, entitled “Shared Responsibility for Health Care.” Part I of that subtitle concerns the individual mandate; and Part II concerns employer responsibilities, including the wholly separate employer insurance mandate. The findings do not even apply to the entire Subtitle F, much less the entire Title I. To argue that they somehow apply to all of the sections that come before and also to all sections that come after it—which include nearly five hundred sections of the law includingMedicare, Medicaid, health workforce, biologic drugs and more—discards Congress’s textual, organizational choices. As members of the Court have repeatedly written, including during this term, the “Court generally presumes that when Congress includes particular language in one section of a statute but omits it in another, Congress intended a difference in meaning.”62 Subtitle G contains miscellaneous provisions.63

2. How the ACA Uses Other Subsection-Specific Findings

That the mandate findings do not apply beyond their limited territory is buttressed by the fact that there are three separate “findings” subsections in the ACA. Each is clearly location specific.

In addition to the findings subsetion concerning the mandate,64 there is another one in Title I, Subtitle G, Section 1563,65 which concerns solvency of the Medicare and the Social Security trust funds. And there is a third in Title II, Subtitle E, Section 2406,66 acknowledging that “Congress has never acted” on its 1990 bipartisan study of long-term care needs and has not yet addressed with Medicaid reimbursement reform the Court’s decision in Olmstead v. L.C.67 that “individuals with disabilities have the right to choose to receive their long-term services and supports in the community.”68 The words in each set of respective findings, as discussed next, further confirm that each set of findings is placed in a specific location to make a limited point.

C. Congress’s Words Matter

All three of the ACA’s separate subsection-specific findings—the subsection 1501(a) mandate findings and the other two subsection-specific findings as well—also follow the precise instructions concerning the use and writing of findings contained in the drafting manuals issued by Congress’s nonpartisan drafting offices, the Offices of House and Senate Legislative Counsel. I have previously detailed the centrality of the work of those drafters when it comes to putting statutory words on the page.69 Members of the Court—in opinions by Justices Kennedy, Ginsburg, and Alito—have cited these drafting manuals at least three times in recent years.70

The House drafting manual “discourage[s]” findings sections entirely.71 A footnote notes, however:

[T]here are certain circumstances in which congressional findings may be imperative to establish the constitutional basis for congressional action. This is particularly the case in legislation in which congressional action is based on the effect of an activity in interstate commerce. As the Supreme Court recently reaffirmed in its opinion in the case of United States v. Lopez, 63 U.S.L.W. 4343, 4347 (U.S. Apr. 26, 1995), “as part of our independent evaluation of constitutionality under the Commerce Clause we of course consider legislative findings, and indeed even congressional committee findings, regarding effect on interstate commerce.”72

The Senate drafting manual likewise states:

A findings or purposes section maybe included in a draft if, for purposes of clarity, constitutionality concerns, or other reasons, such a section would aid the draft. See, e.g., United States v. Lopez, 115 S.Ct. 1624, 1631 (1995) (“[A]s part of our independent evaluation of constitutionality under the Commerce Clause we of course consider legislative findings, and indeed even congressional committee findings, regarding effect on interstate commerce . . .”).73

Later, the Senate manual refers to findings as provisions that are different from “operative” provisions.74 This is not to say that findings are not “law”—the challengers here argue that findings are indeed enacted law, and my own work makes that point.75 But there is a difference between enacted law that is precatory or descriptive and enacted law directing specific action, such as the intensely powerful remedial action of complete inseverability. That is why the drafting manuals view findings as “nonoperative,” even if they are sometimes illustrative of Congress’s views about policy or its own power. The Court likewise has described findings as nonoperative provisions.76

All three of the ACA’s subsection-specific findings provisions follow these directives from the drafting manuals.77 The mandate findings exactly track the House and Senate Offices of Legislative Counsel’s advice to include findings to justify the commerce power. Subsection 1501(a) mentions the commerce-power justification explicitly in each of its three paragraphs. (The challengers cite only one of the three, (a)(2), but there are three.) As to the other two findings subsections, as the drafting manuals direct, the fiscal findings are inserted to clarify a point of debate about surpluses from the long-term care and social security funds,78 and the long-term care findings respond directly to a Supreme Court case.79

1. Congress Could Not Have Been Clearer that the Findings Are About the Commerce Power

As noted, the first paragraph of Subsection 1501(a) makes explicit what the findings are for: the Commerce Clause, not severability. It begins: “(1) In General.—The individual responsibility requirement provided for in this section (in this section referred to as the ‘requirement’) is commercial and economic in nature, and substantially affects interstate commerce, as a result of the effects described in paragraph (2).80 The text of Paragraph 1 makes clear what is the purpose of the “effects described in paragraph (2)”—namely to document that the mandate is “commercial and economic in nature, and substantially affects interstate commerce.”

We will come back to Paragraph (2) in a moment, but if this were not clear enough, Paragraph 3 makes the point again: “(3) Supreme court ruling.—In United States v. South-Eastern Underwriters Association (322 U.S. 533 (1944)), the Supreme Court of the United States ruled that insurance is interstate commerce subject to Federal regulation.”81

Sandwiched between those paragraphs is the paragraph about the mandate’s significance to the insurance markets, with the subparagraphs therein that the challengers cite. But Paragraph (2) makes explicit beyond doubt that the ensuing words about insurance markets are words about the economic impact of the law. The paragraph is titled “effects on the national economy and interstate commerce.”82 And the first words of text, which precede the subparagraphs, reinforce that the findings relate to the “effects” referenced in the title, stating “[t]he effects described in this paragraph are the following.”83 The bolded and underlined sections set forth earlier in this Essay illustrate the numerous explicit references to commerce in that subsection.

Congress could not have made clearer that the findings are about the commerce power. No words that Congress uses for inseverability—whether “severability,” “separability”, or “invalid”—appear in the findings. As detailed below, “essential to” is not how Congress indicates inseverability; it is how Congress speaks about commerce.

Of course, even though irrelevant to severability, any findings inserted into the ACA to justify Congress’s Commerce Clause authority to enact the mandate are null and moot after NFIB, where the Court rejected that basis for Congress’s authority.

2. The Findings State the Mandate Is “Essential to” Statutes Outside the ACA and Would Pull Those Statutes Down Too on the Challengers’ Argument—An Absurd Result

The challengers argue that because the findings say the mandate is “essential to” the healthy functioning of insurance markets and also references other parts of the ACA, we should read those words as an explicit inseverability directive. The complete words of the mandate’s findings make clear, however, that this cannot possibly be the case. This is evident because the findings’ words of essentiality reference laws outside the ACA too.

Consider for example, Subparagraph (2)(H), the second sentence of which the challengers cite84:

Under the Employee Retirement Income Security Act of 1974 (29 U.S.C. 1001 et seq.), the Public Health Service Act (42 U.S.C. 201 et seq.), and this Act, the Federal Government has a significant role in regulating health insurance. The requirement is an essential part of this larger regulation of economic activity, and the absence of the requirement would undercut Federal regulation of the health insurance market.85

Read as the challengers would read it, striking down the mandate would render ERISA and the Public Health Service Act unconstitutional too. Textualists read statutory language, especially when the same language is used within a single statutory section, coherently—like subsections and phrases are read alike.86 This subparagraph makes clear that the findings—and the word “essential”—are about the economic bases of the ACA’s reforms and their broader connection to the health markets, including other statutes—and not about their indelible link to any particular provision of the ACA itself or to any other statute.

3. This Is Boilerplate Language for Commerce Clause Findings

The language used in the ACA—that the mandate is “essential to” the “markets”—is common congressional parlance for justifying Congress’s Commerce Clause power.87 Reviewing Westlaw’s Public Laws Database of all statutes enacted since 1973, “essential to” is used at least 138 times in findings sections, and in virtually all instances to justify Congress’s constitutional lawmaking power. The majority of those relate to commerce; otherwise, such findings describe a current policy problem otherwise justifying congressional intervention.88 Language about conditions for healthy markets,89 the importance of the regulated activity to the economy,90 and references to commerce,91 all of which appear in the ACA’s findings, are also common fare for Commerce Clause findings.

There are countless examples. For use of “essential” in this context, for instance, the Adam Walsh Child Protection and Safety Act of 2006 provides:

(D) Intrastate incidents of production, transportation, distribution, receipt, advertising, and possession of child pornography, as well as the transfer of custody of children for the production of child pornography, have a substantial and direct effect upon interstate commerce. . . .

. . . .

(F) Federal control of the intrastate incidents of the production . . . and possession of child pornography, as well as the intrastate transfer of children for the production of child pornography, is essential to the effective control of the interstate market in child pornography.92

The ANCSA Land Bank Protection Act of 1998 states that “the implementation of the exchange referenced in this subsection is essential to helping Calista utilize its assets to carry out those responsibilities and to realize the benefits of the Alaska Native Claims Settlement Act.”93

The Great Lakes Fish and Wildlife Restoration Act of 1990 states:

(2) The fishery resources of the Great Lakes support recreational fisheries enjoyed by more than 5,000,000 people annually and commercial fisheries providing approximately 9,000 jobs. Together, these fisheries generate economic activity worth more than $4,400,000,000 annually to the United States.

(3) The availability of a suitable forage base is essential to lake trout, walleye, yellow perch, and other recreational and commercially valuable fishery resources of the Great Lakes Basin.94

Members of the Court have referenced findings in at least 100 opinions since 1980—the dawn of textualism. 95 The majority are findings, like these, related to commerce. It does not appear that the Court has ever held over this forty year period that such Commerce Clause findings do double duty as an explicit inseverability clause.

D. Congress’s Consistent Drafting Practices: Other Sections Across the U.S. Code Make Clear the Findings Section Is Not Boilerplate for Inseverability

Just this term, Justice Kavanaugh wrote:

When interpreting statutes, as the Court has often said, we “usually presume differences in language” convey “differences in meaning.” When Congress chooses distinct phrases to accomplish distinct purposes, and does so over and over again for decades, we may not lightly toss aside all of Congress’s careful handiwork. As Justice Scalia explained for the Court, “it is not our function” to “treat alike subjects that different Congresses have chosen to treat differently.” And the Court has likewise stressed that we may not read “a specific concept into general words when precise language in other statutes reveals that Congress knew how to identify that concept.”96

When Congress does include an explicit inseverability clause, it does so with different language and is unmistakably direct. A search of both the Westlaw Historical Public Laws Database and the Office of Law Revision Counsel’s compilation of the U.S. Code for the root terms “inseverab!,” inseparab!,” “notseverab!,” “nonseparab!,” “not severab!,” “not separab!” and “invalid!” produced nine statutes.97 Each sets out the inseverability clause, broken out by a separate section, subsection or paragraph.

Each statute uses the same boilerplate language. As noted, while the House and Senate drafting manuals state that severability clauses are not necessary, they both do suggest explicit nonseverability clauses or clauses providing that a statute is severable and inseverable in part and specifying which parts fall into each category.98

Every single one of the enacted inseverability clauses in the U.S. Code uses the word “invalid.” That word appears nowhere in the ACA findings. And when only partial inseverability is desired, Congress is very specific (as the drafting manuals themselves direct). Here are examples of the consistent ways Congress speaks about inseverability:

  • Nonseverability. If any provision of this title or the application of any provision of this title to any person or circumstance is held invalid by reason of a violation of the Constitution, the entire title shall be considered invalid.99
  • Severability. If any provision of this Act or of any amendment made by this Act, or the application of any such provision to any person or circumstance, is held to be invalid for any reason, the remainder of this Act and of amendments made by this Act, and the application of the provisions and of the amendments made by this Act to any other person or circumstance shall not be affected by such holding, except that each of subclauses (II), (III), and (IV) of section 205(d)(2)(I)(i) is deemed to be inseverable from the other 2, such that if any 1 of those 3 subclauses is held to be invalid for any reason, neither of the other 2 such subclauses shall be given effect.100
  • Inseparability. In the event that any provision of section 4 of this Act is held invalid, it is the intent of Congress that the entire Act be invalidated. In the event that any other section or provision of this Act is held invalid, it is the intent of Congress that the remaining sections of this Act shall continue in full force and effect.101

Congress has shown it knows how to be, and in fact is, explicit, about inseverability. This is not a surprising result given the extraordinary remedial power an inseverability clause gives to a court. The Court (and especially its textualists) does not imply what Congress has shown elsewhere it knows how to say expressly.

Justice Thomas wrote just this term, citing the Scalia and Garner treatise on statutory interpretation: “It is a fundamental principle of statutory interpretation that ‘absent provision[s] cannot be supplied by the courts.’”102 As then-Judge Kavanaugh wrote, in another healthcare case in 2017: “We must respect Congress’s use of different language and its establishment of different . . . requirements in [two different statutes] . . . . [A] material variation in terms suggests a variation in meaning.”103 Justice Gorsuch’s affirming opinion for the Court in that case in 2019 noted that Congress had shown its ability to speak directly elsewhere and so rejected “the doubtful proposition that Congress sought to accomplish in a ‘surpassingly strange manner’ what it could have accomplished in a much more straightforward way.. . . When legislators did not adopt ‘obvious alternative’ language, ‘the natural implication is that they did not intend’ the alternative.”104

III. statutory history

Finally, textualists sometimes look at previous versions of a law—so-called “statutory history”—to divine its meaning as it changed through Congress, even though many will not look to floor statements and committee reports (so-called “legislative history”). The statutory history of the ACA adds further support to the conclusion that the findings subsection was added to address only Commerce Clause concerns and not severability.

The bill that became the ACA, as has been chronicled elsewhere, was an amalgam of the Senate Finance Committee105 and Senate Committee on Health, Education, Labor, and Pensions (HELP)106 drafts. The House draft was ultimately not utilized at all.107 Both Senate committee bills’ individual-responsibility sections established the requirement by amending the Internal Revenue Code. The HELP Committee bill referred to the penalty for not complying with the mandate as a “payment” while the Finance draft explicitly called it a “tax.” The findings subsection did not appear in either bill.

After both bills passed out of committee, Senate Majority Leader Harry Reid produced a combined version.108 The text of the amalgam bill’s insurance-purchase requirement closely tracked that of the Finance Committee bill in language and organization109 but made two noticeable changes. First, it deleted the Finance Committee’s language that had explicitly referred to the penalty as a “tax.”110 Second, it added for the first time a findings subsection to the mandate’s section, specifying the Commerce Power as the basis for Congress’s authority.111 This statutory history makes clear that the purpose of the amendment was to address the Commerce Clause. Severability had no relevance or role.

Legislative history confirms further that the findings were added, as the manuals direct, to anticipate a constitutional claim. Some senators were concerned that a tax would be politically unpopular;112 other senators were concerned that the mandate might be unconstitutional if not a tax. Debate on the Senate floor on December 22 and 23, 2009 indeed raised a constitutional point of order specifically about the mandate.113 Several Republican senators argued that the individual mandate was not permissible under the Commerce Clause.114 Among exhibits put in the record was a memo from the Conservative Action Project that directly referenced the findings: “The Commerce Clause requires an actual economic effect, not merely a congressional finding of an economic effect. . . . Therefore the various interstate-commerce findings in the Senate version of the ‘Obamacare’ legislation do not make the bill constitutional.”115

Senator Patrick Leahy, then-Chairman of the Senate Judiciary Committee, explicitly countered these Commerce Clause objections and used the word “essential” to argue, among other things: “A requirement that all Americans have health insurance—like requirements to be vaccinated or to have car insurance or to register for the draft or to pay taxes—is within congressional power if Congress determines it to be essential to controlling spiraling health care costs.”116

Statutory and legislative history confirms what text, location and context already make clear.


Recall Justice Scalia’s question about the ACA: “You really want us to go through these 2,700 pages”?117

Textualists cannot have their proverbial cake and feast on it too. Congress’s intent to maintain the ACA is crystal clear from its own 2017 actions. That should be enough for severability. But if the Court has any doubt, arguments that ignore the location, words, and context of the findings as well as ignore how Congress usually speaks when it does talk about inseverability have no place in textualist analysis.

In NFIB, ACA opponents told the Court to give no weight to Congress’s Commerce Clause findings. Now they are saying Congress’s findings deserve some kind of hyper-deference for what they do not even say—including a secret-yet-somehow-also-explicit order to destroy the law. In NFIB, it is also worth noting, the Court invalidated the requirement that states expand Medicaid, a far bigger change to the ACA than invalidating a mandate that has never been fully enforced.118 The majority there wrote: “The question here is whether Congress would have wanted the rest of the Act to stand . . . . Unless it is evident that the answer is no, we must leave the rest of the Act intact.”119

False textualism, as Thomas Merrill warned in 1994, risks turning the serious legal work of statutory interpretation into nothing more than a clever, empty game,120 casting a shadow of illegitimacy on the Court itself. Justice Kavanaugh, himself a noted textualist, warned just this term that “[c]onstitutional litigation is not a game of gotcha against Congress, where litigants can ride a discrete constitutional flaw in a statute to take down the whole, otherwise constitutional statute.”121

Respecting Congress means understanding Congress, and reading what Congress actually wrote and how it wrote it.

Professor of Law and Faculty Director, Solomon Center for Health Law and Policy, Yale Law School. Thanks to Jonathan Adler, Tim Jost, Mark Regan, the YLJ editors and, for research assistance, Sherry Tanious, Erica Turret, Sahrula Kubie, Jishian Ravinthiran, Jade Ford, Andrew DeGuglielmo, and Timur Akman-Duffy.