King v. Burwell, 576 U.S. 473 (2015)

NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.

SUPREME COURT OF THE UNITED STATES

_________________

No. 14–114

_________________

DAVID KING, et al., PETITIONERS v. SYLVIA BURWELL, SECRETARY OF HEALTHAND HUMAN SERVICES, et al.

on writ of certiorari to the united states court of appeals for the fourth circuit

[June 25, 2015]

Chief Justice Roberts delivered the opinion of the Court.

The Patient Protection and Affordable Care Act adopts a series of interlocking reforms designed to expand coverage in the individual health insurance market. First, the Act bars insurers from taking a person’s health into account when deciding whether to sell health insurance or how much to charge. Second, the Act generally requires each person to maintain insurance coverage or make a payment to the Internal Revenue Service. And third, the Act gives tax credits to certain people to make insurance more affordable.

In addition to those reforms, the Act requires the creation of an “Exchange” in each State—basically, a marketplace that allows people to compare and purchase insurance plans. The Act gives each State the opportunity to establish its own Exchange, but provides that the Federal Government will establish the Exchange if the State does not.

This case is about whether the Act’s interlocking reforms apply equally in each State no matter who establishes the State’s Exchange. Specifically, the question pre-sented is whether the Act’s tax credits are available in States that have a Federal Exchange.

I

A

The Patient Protection and Affordable Care Act, 124Stat. 119, grew out of a long history of failed health insurance reform. In the 1990s, several States began experimenting with ways to expand people’s access to coverage. One common approach was to impose a pair of insurance market regulations—a “guaranteed issue” requirement, which barred insurers from denying coverage to any person because of his health, and a “community rating” requirement, which barred insurers from charging a person higher premiums for the same reason. Together, those requirements were designed to ensure that anyone who wanted to buy health insurance could do so.

The guaranteed issue and community rating requirements achieved that goal, but they had an unintended consequence: They encouraged people to wait until they got sick to buy insurance. Why buy insurance coverage when you are healthy, if you can buy the same coverage for the same price when you become ill? This consequence—known as “adverse selection”—led to a second: Insurers were forced to increase premiums to account for the fact that, more and more, it was the sick rather than the healthy who were buying insurance. And that consequence fed back into the first: As the cost of insurance rose, even more people waited until they became ill tobuy it.

This led to an economic “death spiral.” As premiums rose higher and higher, and the number of people buying insurance sank lower and lower, insurers began to leave the market entirely. As a result, the number of people without insurance increased dramatically.

This cycle happened repeatedly during the 1990s. For example, in 1993, the State of Washington reformed its individual insurance market by adopting the guaranteed issue and community rating requirements. Over the next three years, premiums rose by 78 percent and the number of people enrolled fell by 25 percent. By 1999, 17 of the State’s 19 private insurers had left the market, and the remaining two had announced their intention to do so. Brief for America’s Health Insurance Plans as Amicus Curiae 10–11.

For another example, also in 1993, New York adopted the guaranteed issue and community rating requirements. Over the next few years, some major insurers in the individual market raised premiums by roughly 40 percent. By 1996, these reforms had “effectively eliminated the commercial individual indemnity market in New York with the largest individual health insurer exiting the market.” L. Wachenheim & H. Leida, The Impact of Guaranteed Issue and Community Rating Reforms on States’ Individual Insurance Markets 38 (2012).

In 1996, Massachusetts adopted the guaranteed issue and community rating requirements and experienced similar results. But in 2006, Massachusetts added two more reforms: The Commonwealth required individuals to buy insurance or pay a penalty, and it gave tax credits to certain individuals to ensure that they could afford the insurance they were required to buy. Brief for Bipartisan Economic Scholars as Amici Curiae 24–25. The combination of these three reforms—insurance market regulations, a coverage mandate, and tax credits—reduced the uninsured rate in Massachusetts to 2.6 percent, by far the lowest in the Nation. Hearing on Examining Individual State Experiences with Health Care Reform Coverage Initiatives in the Context of National Reform before the Senate Committee on Health, Education, Labor, and Pensions, 111th Cong., 1st Sess., 9 (2009).

B

The Affordable Care Act adopts a version of the three key reforms that made the Massachusetts system successful. First, the Act adopts the guaranteed issue and community rating requirements. The Act provides that “each health insurance issuer that offers health insurance coverage in the individual . . . market in a State must accept every . . . individual in the State that applies for such coverage.” 42 U. S. C. §300gg–1(a). The Act also bars insurers from charging higher premiums on the basis of a person’s health. §300gg.

Second, the Act generally requires individuals to maintain health insurance coverage or make a payment to the IRS. 26 U. S. C. §5000A. Congress recognized that, without an incentive, “many individuals would wait to purchase health insurance until they needed care.” 42 U. S. C. §18091(2)(I). So Congress adopted a coverage requirement to “minimize this adverse selection and broaden the health insurance risk pool to include healthy individuals, which will lower health insurance premiums.” Ibid. In Congress’s view, that coverage requirement was “essential to creating effective health insurance markets.” Ibid. Congress also provided an exemption from the coverage requirement for anyone who has to spend more than eight percent of his income on health insurance. 26 U. S. C. §§5000A(e)(1)(A), (e)(1)(B)(ii).

Third, the Act seeks to make insurance more affordable by giving refundable tax credits to individuals with household incomes between 100 percent and 400 percent of the federal poverty line. §36B. Individuals who meet the Act’s requirements may purchase insurance with the tax credits, which are provided in advance directly to the individual’s insurer. 42 U. S. C. §§18081, 18082.

These three reforms are closely intertwined. As noted, Congress found that the guaranteed issue and community rating requirements would not work without the coverage requirement. §18091(2)(I). And the coverage requirement would not work without the tax credits. The reason is that, without the tax credits, the cost of buying insurance would exceed eight percent of income for a large number of individuals, which would exempt them from the coverage requirement. Given the relationship between these three reforms, the Act provided that they should take effect on the same day—January 1, 2014. See Affordable Care Act, §1253, redesignated §1255, 124Stat. 162, 895; §§1401(e), 1501(d), id., at 220, 249.

C

In addition to those three reforms, the Act requires the creation of an “Exchange” in each State where peoplecan shop for insurance, usually online. 42 U. S. C. §18031(b)(1). An Exchange may be created in one of two ways. First, the Act provides that “[e]ach State shall . . . establish an American Health Benefit Exchange . . . for the State.” Ibid. Second, if a State nonetheless chooses not to establish its own Exchange, the Act provides that the Secretary of Health and Human Services “shall . . . establish and operate such Exchange within the State.” §18041(c)(1).

The issue in this case is whether the Act’s tax credits are available in States that have a Federal Exchange rather than a State Exchange. The Act initially provides that tax credits “shall be allowed” for any “applicable taxpayer.” 26 U. S. C. §36B(a). The Act then provides that the amount of the tax credit depends in part on whether the taxpayer has enrolled in an insurance plan through “an Exchange established by the State under section 1311 of the Patient Protection and Affordable Care Act [hereinafter 42 U. S. C. §18031].” 26 U. S. C. §§36B(b)–(c) (emphasis added).

The IRS addressed the availability of tax credits by promulgating a rule that made them available on both State and Federal Exchanges. 77 Fed. Reg. 30378 (2012). As relevant here, the IRS Rule provides that a taxpayer is eligible for a tax credit if he enrolled in an insurance plan through “an Exchange,” 26 CFR §1.36B–2 (2013), which is defined as “an Exchange serving the individual market . . . regardless of whether the Exchange is established and operated by a State . . . or by HHS,” 45 CFR §155.20 (2014). At this point, 16 States and the District of Columbia have established their own Exchanges; the other 34 States have elected to have HHS do so.

D

Petitioners are four individuals who live in Virginia, which has a Federal Exchange. They do not wish to purchase health insurance. In their view, Virginia’s Exchange does not qualify as “an Exchange established by the State under [ 42 U. S. C. §18031],” so they should not receive any tax credits. That would make the cost of buying insurance more than eight percent of their income, which would exempt them from the Act’s coverage requirement. 26 U. S. C. §5000A(e)(1).

Under the IRS Rule, however, Virginia’s Exchange would qualify as “an Exchange established by the State under [ 42 U. S. C. §18031],” so petitioners would receive tax credits. That would make the cost of buying insurance less than eight percent of petitioners’ income, which would subject them to the Act’s coverage requirement. The IRS Rule therefore requires petitioners to either buy health insurance they do not want, or make a payment to the IRS.

Petitioners challenged the IRS Rule in Federal District Court. The District Court dismissed the suit, holding that the Act unambiguously made tax credits available to individuals enrolled through a Federal Exchange. King v. Sebelius, 997 F. Supp. 2d 415 (ED Va. 2014). The Court of Appeals for the Fourth Circuit affirmed. 759 F. 3d 358 (2014). The Fourth Circuit viewed the Act as “ambiguous and subject to at least two different interpretations.” Id., at 372. The court therefore deferred to the IRS’s interpretation under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984) . 759 F. 3d, at 376.

The same day that the Fourth Circuit issued its decision, the Court of Appeals for the District of Columbia Circuit vacated the IRS Rule in a different case, holding that the Act “unambiguously restricts” the tax credits to State Exchanges. Halbig v. Burwell, 758 F. 3d 390, 394 (2014). We granted certiorari in the present case. 574 U. S. ___ (2014).

II

The Affordable Care Act addresses tax credits in what is now Section 36B of the Internal Revenue Code. That section provides: “In the case of an applicable taxpayer, there shall be allowed as a credit against the tax imposed by this subtitle . . . an amount equal to the premium assistance credit amount.” 26 U. S. C. §36B(a). Section 36B then defines the term “premium assistance credit amount” as “the sum of the premium assistance amounts determined under paragraph (2) with respect to all coverage months of the taxpayer occurring during the taxable year.” §36B(b)(1) (emphasis added). Section 36B goes on to define the two italicized terms—“premium assistance amount” and “coverage month”—in part by referring to an insurance plan that is enrolled in through “an Exchange established by the State under [ 42 U. S. C. §18031].” 26 U. S. C. §§36B(b)(2)(A), (c)(2)(A)(i).

The parties dispute whether Section 36B authorizes tax credits for individuals who enroll in an insurance plan through a Federal Exchange. Petitioners argue that a Federal Exchange is not “an Exchange established by the State under [ 42 U. S. C. §18031],” and that the IRS Rule therefore contradicts Section 36B. Brief for Petitioners 18–20. The Government responds that the IRS Rule is lawful because the phrase “an Exchange established by the State under [ 42 U. S. C. §18031]” should be read to include Federal Exchanges. Brief for Respondents 20–25.

When analyzing an agency’s interpretation of a statute, we often apply the two-step framework announced in Chevron, 467 U. S. 837 . Under that framework, we ask whether the statute is ambiguous and, if so, whether the agency’s interpretation is reasonable. Id., at 842–843. This approach “is premised on the theory that a statute’s ambiguity constitutes an implicit delegation from Congress to the agency to fill in the statutory gaps.” FDA v. Brown & Williamson Tobacco Corp., 529 U. S. 120, 159 (2000) . “In extraordinary cases, however, there may be reason to hesitate before concluding that Congress has intended such an implicit delegation.” Ibid.

This is one of those cases. The tax credits are among the Act’s key reforms, involving billions of dollars in spending each year and affecting the price of health insurance for millions of people. Whether those credits are available on Federal Exchanges is thus a question of deep “economic and political significance” that is central to this statutory scheme; had Congress wished to assign that question to an agency, it surely would have done so expressly. Utility Air Regulatory Group v. EPA, 573 U. S. ___, ___ (2014) (slip op., at 19) (quoting Brown & Williamson, 529 U. S., at 160). It is especially unlikely that Congress would have delegated this decision to the IRS, which has no expertise in crafting health insurance policy of this sort. See Gonzales v. Oregon, 546 U. S. 243 –267 (2006). This is not a case for the IRS.

It is instead our task to determine the correct reading of Section 36B. If the statutory language is plain, we must enforce it according to its terms. Hardt v. Reliance Standard Life Ins. Co., 560 U. S. 242, 251 (2010) . But oftentimes the “meaning—or ambiguity—of certain words or phrases may only become evident when placed in context.” Brown & Williamson, 529 U. S., at 132. So when deciding whether the language is plain, we must read the words “in their context and with a view to their place in the overall statutory scheme.” Id., at 133 (internal quotation marks omitted). Our duty, after all, is “to construe statutes, not isolated provisions.” Graham County Soil and Water Conservation Dist. v. United States ex rel. Wilson, 559 U. S. 280, 290 (2010) (internal quotation marks omitted).

A

We begin with the text of Section 36B. As relevant here, Section 36B allows an individual to receive tax credits only if the individual enrolls in an insurance plan through “an Exchange established by the State under [ 42 U. S. C. §18031].” In other words, three things must be true: First, the individual must enroll in an insurance plan through “an Exchange.” Second, that Exchange must be “established by the State.” And third, that Exchange must be established “under [ 42 U. S. C. §18031].” We address each requirement in turn.

First, all parties agree that a Federal Exchange qualifies as “an Exchange” for purposes of Section 36B. See Brief for Petitioners 22; Brief for Respondents 22. Section 18031 provides that “[e]ach State shall . . . establish an American Health Benefit Exchange . . . for the State.” §18031(b)(1). Although phrased as a requirement, the Act gives the States “flexibility” by allowing them to “elect” whether they want to establish an Exchange. §18041(b). If the State chooses not to do so, Section 18041 provides that the Secretary “shall . . . establish and operatesuch Exchange within the State.” §18041(c)(1) (emphasis added).

By using the phrase “such Exchange,” Section 18041 instructs the Secretary to establish and operate the same Exchange that the State was directed to establish under Section 18031. See Black’s Law Dictionary 1661 (10th ed. 2014) (defining “such” as “That or those; having just been mentioned”). In other words, State Exchanges and Fed-eral Exchanges are equivalent—they must meet the same requirements, perform the same functions, and serve the same purposes. Although State and Federal Exchanges are established by different sovereigns, Sections 18031 and 18041 do not suggest that they differ in any meaningful way. A Federal Exchange therefore counts as “an Exchange” under Section 36B.

Second, we must determine whether a Federal Exchange is “established by the State” for purposes of Section 36B. At the outset, it might seem that a Federal Exchange cannot fulfill this requirement. After all, the Act defines “State” to mean “each of the 50 States and the District of Columbia”—a definition that does not include the Federal Government. 42 U. S. C. §18024(d). But when read in context, “with a view to [its] place in the overall statutory scheme,” the meaning of the phrase “established by the State” is not so clear. Brown &Williamson, 529 U. S., at 133 (internal quotation marks omitted).

After telling each State to establish an Exchange, Section 18031 provides that all Exchanges “shall make available qualified health plans to qualified individuals.” 42 U. S. C. §18031(d)(2)(A). Section 18032 then defines the term “qualified individual” in part as an individual who “resides in the State that established the Exchange.” §18032(f)(1)(A). And that’s a problem: If we give the phrase “the State that established the Exchange” its most natural meaning, there would be no “qualified individuals” on Federal Exchanges. But the Act clearly contemplates that there will be qualified individuals on every Exchange. As we just mentioned, the Act requires all Exchanges to “make available qualified health plans to qualified individuals”—something an Exchange could not do if there were no such individuals. §18031(d)(2)(A). And the Act tells the Exchange, in deciding which health plans to offer, to consider “the interests of qualified individuals . . . in the State or States in which such Exchange operates”—again, something the Exchange could not do if qualified individ-uals did not exist. §18031(e)(1)(B). This problem arises repeatedly throughout the Act. See, e.g., §18031(b)(2) (allowing a State to create “one Exchange . . . for providing . . . services to both qualified individuals and qualified small employers,” rather than creating separate Exchanges for those two groups).[1]

These provisions suggest that the Act may not always use the phrase “established by the State” in its most natural sense. Thus, the meaning of that phrase may not be as clear as it appears when read out of context.

Third, we must determine whether a Federal Exchange is established “under [ 42 U. S. C. §18031].” This too might seem a requirement that a Federal Exchange cannot fulfill, because it is Section 18041 that tells the Secretary when to “establish and operate such Exchange.” But here again, the way different provisions in the statute interact suggests otherwise.

The Act defines the term “Exchange” to mean “an American Health Benefit Exchange established under section 18031.” §300gg–91(d)(21). If we import that definition into Section 18041, the Act tells the Secretary to “establish and operate such ‘American Health Benefit Exchange established under section 18031.’ ” That suggests that Section 18041 authorizes the Secretary to establish an Exchange under Section 18031, not (or not only) under Section 18041. Otherwise, the Federal Exchange, by definition, would not be an “Exchange” at all. See Halbig, 758 F. 3d, at 399–400 (acknowledging that the Secretary establishes Federal Exchanges under Section 18031).

This interpretation of “under [ 42 U. S. C. §18031]” fits best with the statutory context. All of the requirements that an Exchange must meet are in Section 18031, so it is sensible to regard all Exchanges as established under that provision. In addition, every time the Act uses the word “Exchange,” the definitional provision requires that we substitute the phrase “Exchange established under section 18031.” If Federal Exchanges were not established under Section 18031, therefore, literally none of the Act’s requirements would apply to them. Finally, the Act repeatedly uses the phrase “established under [ 42 U. S. C. §18031]” in situations where it would make no sense to distinguish between State and Federal Exchanges. See, e.g., 26 U. S. C. §125(f)(3)(A) (2012 ed., Supp. I) (“The term ‘qualified benefit’ shall not include any qualified health plan . . . offered through an Exchange established under [ 42 U. S. C. §18031]”); 26 U. S. C. §6055(b)(1)(B)(iii)(I) (2012 ed.) (requiring insurers to report whether each insurance plan they provided “is a qualified health plan offered through an Exchange established under [ 42 U. S. C. §18031]”). A Federal Exchange may therefore be considered one established “under [ 42 U. S. C. §18031].”

The upshot of all this is that the phrase “an Exchange established by the State under [ 42 U. S. C. §18031]” is properly viewed as ambiguous. The phrase may be limited in its reach to State Exchanges. But it is also possible that the phrase refers to all Exchanges—both State and Federal—at least for purposes of the tax credits. If a State chooses not to follow the directive in Section 18031 that it establish an Exchange, the Act tells the Secretary to establish “such Exchange.” §18041. And by using the words “such Exchange,” the Act indicates that State and Federal Exchanges should be the same. But State and Federal Exchanges would differ in a fundamental way if tax credits were available only on State Exchanges—one type of Exchange would help make insurance more affordable by providing billions of dollars to the States’ citizens; the other type of Exchange would not.[2]

The conclusion that Section 36B is ambiguous is further supported by several provisions that assume tax credits will be available on both State and Federal Exchanges. For example, the Act requires all Exchanges to create outreach programs that must “distribute fair and impartial information concerning . . . the availability of premium tax credits under section 36B.” §18031(i)(3)(B). The Act also requires all Exchanges to “establish and make avail-able by electronic means a calculator to determine the actual cost of coverage after the application of any pre-mium tax credit under section 36B.” §18031(d)(4)(G). And the Act requires all Exchanges to report to the Treasury Secretary information about each health plan they sell, including the “aggregate amount of any advance payment of such credit,” “[a]ny information . . . necessary to determine eligibility for, and the amount of, such credit,” and any “[i]nformation necessary to determine whether a taxpayer has received excess advance payments.” 26 U. S. C. §36B(f)(3). If tax credits were not available on Federal Exchanges, these provisions would make little sense.

Petitioners and the dissent respond that the words “established by the State” would be unnecessary if Congress meant to extend tax credits to both State and Fed-eral Exchanges. Brief for Petitioners 20; post, at 4–5. But “our preference for avoiding surplusage constructions is not absolute.” Lamie v. United States Trustee, 540 U. S. 526, 536 (2004) ; see also Marx v. General Revenue Corp., 568 U. S. ___, ___ (2013) (slip op., at 13) (“The canon against surplusage is not an absolute rule”). And specifically with respect to this Act, rigorous application of the canon does not seem a particularly useful guide to a fair construction of the statute.

The Affordable Care Act contains more than a few examples of inartful drafting. (To cite just one, the Act creates three separate Section 1563s. See 124Stat. 270, 911, 912.) Several features of the Act’s passage contributed to that unfortunate reality. Congress wrote key partsof the Act behind closed doors, rather than through “the traditional legislative process.” Cannan, A Legislative History of the Affordable Care Act: How Legislative Procedure Shapes Legislative History, 105 L. Lib. J. 131, 163 (2013). And Congress passed much of the Act using a complicated budgetary procedure known as “reconciliation,” which limited opportunities for debate and amendment, and bypassed the Senate’s normal 60-vote filibuster requirement. Id., at 159–167. As a result, the Act does not reflect the type of care and deliberation that one might expect of such significant legislation. Cf. Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev. 527, 545 (1947) (describing a cartoon “in which a senator tells his colleagues ‘I admit this new bill is too complicated to understand. We’ll just have to pass it to find out what it means.’ ”).

Anyway, we “must do our best, bearing in mind the fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.” Util-ity Air Regulatory Group, 573 U. S., at ___ (slip op., at 15) (internal quotation marks omitted). After reading Section 36B along with other related provisions in the Act, we cannot conclude that the phrase “an Exchange established by the State under [Section 18031]” is unambiguous.

B

Given that the text is ambiguous, we must turn to the broader structure of the Act to determine the meaning of Section 36B. “A provision that may seem ambiguous in isolation is often clarified by the remainder of the statu-tory scheme . . . because only one of the permissible meanings produces a substantive effect that is compatible with the rest of the law.” United Sav. Assn. of Tex. v. Timbers of Inwood Forest Associates, Ltd., 484 U. S. 365, 371 (1988) . Here, the statutory scheme compels us to reject petitioners’ interpretation because it would destabilize the individual insurance market in any State with a Federal Exchange, and likely create the very “death spirals” that Congress designed the Act to avoid. See New York State Dept. of Social Servs. v. Dublino, 413 U. S. 405 –420 (1973) (“We cannot interpret federal statutes to negate their own stated purposes.”).[3]

As discussed above, Congress based the Affordable Care Act on three major reforms: first, the guaranteed issue and community rating requirements; second, a requirement that individuals maintain health insurance coverage or make a payment to the IRS; and third, the tax credits for individuals with household incomes between 100 percent and 400 percent of the federal poverty line. In a State that establishes its own Exchange, these three reforms work together to expand insurance coverage. The guaranteed issue and community rating requirements ensure that anyone can buy insurance; the coverage requirement creates an incentive for people to do so before they get sick; and the tax credits—it is hoped—make insurance more affordable. Together, those reforms “minimize . . . adverse selection and broaden the health in-surance risk pool to include healthy individuals, which will lower health insurance premiums.” 42 U. S. C. §18091(2)(I).

Under petitioners’ reading, however, the Act would operate quite differently in a State with a Federal Exchange. As they see it, one of the Act’s three major reforms—the tax credits—would not apply. And a second major reform—the coverage requirement—would not apply in a meaningful way. As explained earlier, the coverage requirement applies only when the cost of buying health insurance (minus the amount of the tax credits) is less than eight percent of an individual’s income. 26 U. S. C. §§5000A(e)(1)(A), (e)(1)(B)(ii). So without the tax credits, the coverage requirement would apply to fewer individuals. And it would be a lot fewer. In 2014, approximately 87 percent of people who bought insurance on a Federal Exchange did so with tax credits, and virtually all of those people would become exempt. HHS, A. Burke, A. Misra, & S. Sheingold, Premium Affordability, Competition, and Choice in the Health Insurance Marketplace 5 (2014); Brief for Bipartisan Economic Scholars as Amici Curiae 19–20. If petitioners are right, therefore, only one of the Act’s three major reforms would apply in States with a Federal Exchange.

The combination of no tax credits and an ineffective coverage requirement could well push a State’s individual insurance market into a death spiral. One study predicts that premiums would increase by 47 percent and enrollment would decrease by 70 percent. E. Saltzman & C. Eibner, The Effect of Eliminating the Affordable Care Act’s Tax Credits in Federally Facilitated Marketplaces (2015). Another study predicts that premiums would increase by 35 percent and enrollment would decrease by 69 percent. L. Blumberg, M. Buettgens, & J. Holahan, The Implications of a Supreme Court Finding for the Plaintiff in King vs. Burwell: 8.2 Million More Uninsured and 35% Higher Premiums (2015). And those effects would not be limited to individuals who purchase insurance on the Exchanges. Because the Act requires insurers to treat the entire individual market as a single risk pool, 42 U. S. C. §18032(c)(1), premiums outside the Exchange would rise along with those inside the Exchange. Brief for Bipartisan Economic Scholars as Amici Curiae 11–12.

It is implausible that Congress meant the Act to operate in this manner. See National Federation of Independent Business v. Sebelius, 567 U. S. ___, ___ (2012) (Scalia, Kennedy, Thomas, and Alito, JJ., dissenting) (slip op., at 60) (“Without the federal subsidies . . . the exchanges would not operate as Congress intended and may not operate at all.”). Congress made the guaranteed issue and community rating requirements applicable in every State in the Nation. But those requirements only work when combined with the coverage requirement and the tax credits. So it stands to reason that Congress meant for those provisions to apply in every State as well.[4]

Petitioners respond that Congress was not worried about the effects of withholding tax credits from States with Federal Exchanges because “Congress evidently believed it was offering states a deal they would not refuse.” Brief for Petitioners 36. Congress may have been wrong about the States’ willingness to establish their own Exchanges, petitioners continue, but that does not allow this Court to rewrite the Act to fix that problem. That is particularly true, petitioners conclude, because the States likely would have created their own Exchanges in the absence of the IRS Rule, which eliminated any incentive that the States had to do so. Id., at 36–38.

Section 18041 refutes the argument that Congress believed it was offering the States a deal they would not refuse. That section provides that, if a State elects not to establish an Exchange, the Secretary “shall . . . establish and operate such Exchange within the State.” 42 U. S. C. §18041(c)(1)(A). The whole point of that provision is to create a federal fallback in case a State chooses not to establish its own Exchange. Contrary to petitioners’ argument, Congress did not believe it was offering States a deal they would not refuse—it expressly addressed what would happen if a State did refuse the deal.

C

Finally, the structure of Section 36B itself suggests that tax credits are not limited to State Exchanges. Section 36B(a) initially provides that tax credits “shall be allowed” for any “applicable taxpayer.” Section 36B(c)(1) then defines an “applicable taxpayer” as someone who (among other things) has a household income between 100 percent and 400 percent of the federal poverty line. Together, these two provisions appear to make anyone in the specified income range eligible to receive a tax credit.

According to petitioners, however, those provisions are an empty promise in States with a Federal Exchange. In their view, an applicable taxpayer in such a State would be eligible for a tax credit—but the amount of that tax credit would always be zero. And that is because—diving several layers down into the Tax Code—Section 36B says that the amount of the tax credits shall be “an amount equal to the premium assistance credit amount,” §36B(a); and then says that the term “premium assistance credit amount” means “the sum of the premium assistance amounts determined under paragraph (2) with respect to all coverage months of the taxpayer occurring during the taxable year,” §36B(b)(1); and then says that the term “premium assistance amount” is tied to the amount of the monthly premium for insurance purchased on “an Exchange established by the State under [42 U. S. C. §18031],” §36B(b)(2); and then says that the term “coverage month” means any month in which the taxpayer has insurance through “an Exchange established by the State under [ 42 U. S. C. §18031],” §36B(c)(2)(A)(i).

We have held that Congress “does not alter the fundamental details of a regulatory scheme in vague terms or ancillary provisions.” Whitman v. American Trucking Assns., Inc., 531 U. S. 457, 468 (2001) . But in petitioners’ view, Congress made the viability of the entire Affordable Care Act turn on the ultimate ancillary provision: a sub-sub-sub section of the Tax Code. We doubt that is what Congress meant to do. Had Congress meant to limit tax credits to State Exchanges, it likely would have done so in the definition of “applicable taxpayer” or in some other prominent manner. It would not have used such a winding path of connect-the-dots provisions about the amount of the credit.[5]

D

Petitioners’ arguments about the plain meaning of Section 36B are strong. But while the meaning of the phrase “an Exchange established by the State under [ 42 U. S. C. §18031]” may seem plain “when viewed in isolation,” such a reading turns out to be “untenable in light of [the statute] as a whole.” Department of Revenue of Ore. v. ACF Industries, Inc., 510 U. S. 332, 343 (1994) . In this instance, the context and structure of the Act compel us to depart from what would otherwise be the most natural reading of the pertinent statutory phrase.

Reliance on context and structure in statutory interpretation is a “subtle business, calling for great wariness lest what professes to be mere rendering becomes creation and attempted interpretation of legislation becomes legislation itself.” Palmer v. Massachusetts, 308 U. S. 79, 83 (1939) . For the reasons we have given, however, such reliance is appropriate in this case, and leads us to conclude that Section 36B allows tax credits for insurance purchased on any Exchange created under the Act. Those credits are necessary for the Federal Exchanges to function like their State Exchange counterparts, and to avoid the type of calamitous result that Congress plainly meant to avoid.

*  *  *

In a democracy, the power to make the law rests with those chosen by the people. Our role is more confined—“to say what the law is.” Marbury v. Madison, 1 Cranch 137, 177 (1803). That is easier in some cases than in others. But in every case we must respect the role of the Legislature, and take care not to undo what it has done. A fair reading of legislation demands a fair understanding of the legislative plan.

Congress passed the Affordable Care Act to improve health insurance markets, not to destroy them. If at all possible, we must interpret the Act in a way that is consistent with the former, and avoids the latter. Section 36B can fairly be read consistent with what we see as Congress’s plan, and that is the reading we adopt.

The judgment of the United States Court of Appeals for the Fourth Circuit is

Affirmed.

SUPREME COURT OF THE UNITED STATES

_________________

No. 14–114

_________________

DAVID KING, et al., PETITIONERS v. SYLVIA BURWELL, SECRETARY OF HEALTHAND HUMAN SERVICES, et al.

on writ of certiorari to the united states court of appeals for the fourth circuit

[June 25, 2015]

Justice Scalia, with whom Justice Thomas and Justice Alito join, dissenting.

The Court holds that when the Patient Protection and Affordable Care Act says “Exchange established by the State” it means “Exchange established by the State or the Federal Government.” That is of course quite absurd, and the Court’s 21 pages of explanation make it no less so.

I

The Patient Protection and Affordable Care Act makes major reforms to the American health-insurance market. It provides, among other things, that every State “shall . . . establish an American Health Benefit Exchange”—a marketplace where people can shop for health-insurance plans. 42 U. S. C. §18031(b)(1). And it provides that if a State does not comply with this instruction, the Secretary of Health and Human Services must “establish and operate such Exchange within the State.” §18041(c)(1).

A separate part of the Act—housed in §36B of the Internal Revenue Code—grants “premium tax credits” to subsidize certain purchases of health insurance made on Exchanges. The tax credit consists of “premium assistance amounts” for “coverage months.” 26 U. S. C. §36B(b)(1). An individual has a coverage month only when he is covered by an insurance plan “that was enrolled in through an Exchange established by the State under [§18031].” §36B(c)(2)(A). And the law ties the size of the premium assistance amount to the premiums for health plans which cover the individual “and which were enrolled in through an Exchange established by the State under [§18031].” §36B(b)(2)(A). The premium assistance amount further depends on the cost of certain other insurance plans “offered through the same Exchange.” §36B(b)(3)(B)(i).

This case requires us to decide whether someone who buys insurance on an Exchange established by the Secretary gets tax credits. You would think the answer would be obvious—so obvious there would hardly be a need for the Supreme Court to hear a case about it. In order to receive any money under §36B, an individual must enroll in an insurance plan through an “Exchange established by the State.” The Secretary of Health and Human Services is not a State. So an Exchange established by the Secretary is not an Exchange established by the State—which means people who buy health insurance through such an Exchange get no money under §36B.

Words no longer have meaning if an Exchange that is not established by a State is “established by the State.” It is hard to come up with a clearer way to limit tax credits to state Exchanges than to use the words “established by the State.” And it is hard to come up with a reason to include the words “by the State” other than the purpose of limiting credits to state Exchanges. “[T]he plain, obvious, and rational meaning of a statute is always to be preferred to any curious, narrow, hidden sense that nothing but the exigency of a hard case and the ingenuity and study of an acute and powerful intellect would discover.” Lynch v. Alworth-Stephens Co., 267 U. S. 364, 370 (1925) (internal quotation marks omitted). Under all the usual rules of interpretation, in short, the Government should lose this case. But normal rules of interpretation seem always to yield to the overriding principle of the present Court: The Affordable Care Act must be saved.

II

The Court interprets §36B to award tax credits on both federal and state Exchanges. It accepts that the “most natural sense” of the phrase “Exchange established by the State” is an Exchange established by a State. Ante, at 11. (Understatement, thy name is an opinion on the Afford-able Care Act!) Yet the opinion continues, with no semblance of shame, that “it is also possible that the phrase refers to all Exchanges—both State and Federal.” Ante, at 13. (Impossible possibility, thy name is an opinion on the Affordable Care Act!) The Court claims that “the context and structure of the Act compel [it] to depart from what would otherwise be the most natural reading of the pertinent statutory phrase.” Ante, at 21.

I wholeheartedly agree with the Court that sound interpretation requires paying attention to the whole law, not homing in on isolated words or even isolated sections. Context always matters. Let us not forget, however, why context matters: It is a tool for understanding the terms of the law, not an excuse for rewriting them.

Any effort to understand rather than to rewrite a law must accept and apply the presumption that lawmakers use words in “their natural and ordinary signification.” Pensacola Telegraph Co. v. Western Union Telegraph Co., 96 U. S. 1, 12 (1878) . Ordinary connotation does not always prevail, but the more unnatural the proposed interpretation of a law, the more compelling the contex-tual evidence must be to show that it is correct. Today’s interpretation is not merely unnatural; it is unheard of. Who would ever have dreamt that “Exchange established by the State” means “Exchange established by the State or the Federal Government”? Little short of an express statutory definition could justify adopting this singular reading. Yet the only pertinent definition here provides that “State” means “each of the 50 States and the District of Columbia.” 42 U. S. C. §18024(d). Because the Secretary is neither one of the 50 States nor the District of Columbia, that definition positively contradicts the eccentric theory that an Exchange established by the Secretary has been established by the State.

Far from offering the overwhelming evidence of meaning needed to justify the Court’s interpretation, other contextual clues undermine it at every turn. To begin with, other parts of the Act sharply distinguish between the establishment of an Exchange by a State and the establishment of an Exchange by the Federal Government. The States’ authority to set up Exchanges comes from one provision, §18031(b); the Secretary’s authority comes from an entirely different provision, §18041(c). Funding for States to establish Exchanges comes from one part of the law, §18031(a); funding for the Secretary to establish Exchanges comes from an entirely different part of the law, §18121. States generally run state-created Ex-changes; the Secretary generally runs federally created Exchanges. §18041(b)–(c). And the Secretary’s authority to set up an Exchange in a State depends upon the State’s “[f]ailure to establish [an] Exchange.” §18041(c) (emphasis added). Provisions such as these destroy any pretense that a federal Exchange is in some sense also established by a State.

Reading the rest of the Act also confirms that, as relevant here, there are only two ways to set up an Exchange in a State: establishment by a State and establishment by the Secretary. §§18031(b), 18041(c). So saying that an Exchange established by the Federal Government is “established by the State” goes beyond giving words bizarre meanings; it leaves the limiting phrase “by the State” with no operative effect at all. That is a stark violation of the elementary principle that requires an interpreter “to give effect, if possible, to every clause and word of a statute.” Montclair v. Ramsdell, 107 U. S. 147, 152 (1883) . In weighing this argument, it is well to remember the difference between giving a term a meaning that duplicates another part of the law, and giving a term no meaning at all. Lawmakers sometimes repeat themselves—whether out of a desire to add emphasis, a sense of belt-and-suspenders caution, or a lawyerly penchant for doublets (aid and abet, cease and desist, null and void). Lawmakers do not, however, tend to use terms that “have no operation at all.” Marbury v. Madison, 1 Cranch 137, 174 (1803). So while the rule against treating a term as a redundancy is far from categorical, the rule against treating it as a nullity is as close to absolute as interpretive principles get. The Court’s reading does not merely give “by the State” a duplicative effect; it causes the phrase to have no effect whatever.

Making matters worse, the reader of the whole Act will come across a number of provisions beyond §36B that refer to the establishment of Exchanges by States. Adopting the Court’s interpretation means nullifying the term “by the State” not just once, but again and again throughout the Act. Consider for the moment only those parts of the Act that mention an “Exchange established by the State” in connection with tax credits:

The formula for calculating the amount of the tax credit, as already explained, twice mentions “an Exchange established by the State.” 26 U. S. C. §36B(b)(2)(A), (c)(2)(A)(i).

The Act directs States to screen children for eligibility for “[tax credits] under section 36B” and for “anyother assistance or subsidies available for coverage obtained through” an “Exchange established by the State.” 42 U. S. C. §1396w–3(b)(1)(B)–(C).

The Act requires “an Exchange established by the State” to use a “secure electronic interface” to determine eligibility for (among other things) tax credits. §1396w–3(b)(1)(D).

The Act authorizes “an Exchange established by the State” to make arrangements under which other state agencies “determine whether a State resident is eligible for [tax credits] under section 36B.” §1396w–3(b)(2).

The Act directs States to operate Web sites that allow anyone “who is eligible to receive [tax credits] under section 36B” to compare insurance plans offered through “an Exchange established by the State.” §1396w–3(b)(4).

One of the Act’s provisions addresses the enrollment of certain children in health plans “offered through an Exchange established by the State” and then dis-cusses the eligibility of these children for tax credits. §1397ee(d)(3)(B).

It is bad enough for a court to cross out “by the State” once. But seven times?

Congress did not, by the way, repeat “Exchange established by the State under [§18031]” by rote throughout the Act. Quite the contrary, clause after clause of the law uses a more general term such as “Exchange” or “Exchange established under [§18031].” See, e.g., 42 U. S. C. §§18031(k), 18033; 26 U. S. C. §6055. It is common sense that any speaker who says “Exchange” some of the time, but “Exchange established by the State” the rest of the time, probably means something by the contrast.

Equating establishment “by the State” with establishment by the Federal Government makes nonsense of other parts of the Act. The Act requires States to ensure (on pain of losing Medicaid funding) that any “Exchange established by the State” uses a “secure electronic interface” to determine an individual’s eligibility for various benefits (including tax credits). 42 U. S. C. §1396w–3(b)(1)(D). How could a State control the type of electronic interface used by a federal Exchange? The Act allows a State to control contracting decisions made by “an Exchange established by the State.” §18031(f )(3). Why would a State get to control the contracting decisions of a federal Exchange? The Act also provides “Assistance to States to establish American Health Benefit Exchanges” and directs the Secretary to renew this funding “if the State . . . is making progress . . . toward . . . establishing an Exchange.” §18031(a). Does a State that refuses to set up an Exchange still receive this funding, on the premise that Exchanges established by the Federal Government are really established by States? It is presumably in order to avoid these questions that the Court concludes that federal Exchanges count as state Exchanges only “for purposes of the tax credits.” Ante, at 13. (Contrivance, thy name is an opinion on the Affordable Care Act!)

It is probably piling on to add that the Congress that wrote the Affordable Care Act knew how to equate two different types of Exchanges when it wanted to do so. The Act includes a clause providing that “[a] territory that . . . establishes . . . an Exchange . . . shall be treated as a State” for certain purposes. §18043(a) (emphasis added). Tellingly, it does not include a comparable clause providing that the Secretary shall be treated as a State for purposes of §36B when she establishes an Exchange.

Faced with overwhelming confirmation that “Exchange established by the State” means what it looks like it means, the Court comes up with argument after feeble argument to support its contrary interpretation. None of its tries comes close to establishing the implausible conclusion that Congress used “by the State” to mean “by the State or not by the State.”

The Court emphasizes that if a State does not set up an Exchange, the Secretary must establish “such Exchange.” §18041(c). It claims that the word “such” implies that federal and state Exchanges are “the same.” Ante, at 13. To see the error in this reasoning, one need only consider a parallel provision from our Constitution: “The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof; but the Congress may at any time by Law make or alter such Regulations.” Art. I, §4, cl. 1 (emphasis added). Just as the Affordable Care Act directs States to establish Exchanges while allowing the Secretary to establish “such Exchange” as a fallback, the Elections Clause directs state legislatures to prescribe election regulations while allowing Congress to make “such Regulations” as a fallback. Would anybody refer to an election regulation made by Congress as a “regulation prescribed by the state legislature”? Would anybody say that a federal election law and a state election law are in all respects equivalent? Of course not. The word “such” does not help the Court one whit. The Court’s argument also overlooks the rudimentary principle that a specific provision governs a general one. Even if it were true that the term “such Exchange” in §18041(c) implies that federal and state Exchanges are the same in general, the term “established by the State” in §36B makes plain that they differ when it comes to tax credits in particular.

The Court’s next bit of interpretive jiggery-pokery involves other parts of the Act that purportedly presuppose the availability of tax credits on both federal and state Exchanges. Ante, at 13–14. It is curious that the Court is willing to subordinate the express words of the section that grants tax credits to the mere implications of other provisions with only tangential connections to tax credits. One would think that interpretation would work the other way around. In any event, each of the provisions mentioned by the Court is perfectly consistent with limiting tax credits to state Exchanges. One of them says that the minimum functions of an Exchange include (alongside several tasks that have nothing to do with tax credits) setting up an electronic calculator that shows “the actual cost of coverage after the application of any premium tax credit.” 42 U. S. C. §18031(d)(4)(G). What stops a federal Exchange’s electronic calculator from telling a customer that his tax credit is zero? Another provision requires an Exchange’s outreach program to educate the public about health plans, to facilitate enrollment, and to “distribute fair and impartial information” about enrollment and “the availability of premium tax credits.” §18031(i)(3)(B). What stops a federal Exchange’s outreach program from fairly and impartially telling customers that no tax credits are available? A third provision requires an Exchange to report information about each insurance plan sold—including level of coverage, premium, name of the insured, and “amount of any advance payment” of the tax credit. 26 U. S. C. §36B(f)(3). What stops a federal Exchange’s report from confirming that no tax credits have been paid out?

The Court persists that these provisions “would make little sense” if no tax credits were available on federal Exchanges. Ante, at 14. Even if that observation were true, it would show only oddity, not ambiguity. Laws often include unusual or mismatched provisions. The Affordable Care Act spans 900 pages; it would be amazing if its provisions all lined up perfectly with each other. This Court “does not revise legislation . . . just because the text as written creates an apparent anomaly.” Michigan v. Bay Mills Indian Community, 572 U. S. ___, ___ (2014) (slip op., at 10). At any rate, the provisions cited by the Court are not particularly unusual. Each requires an Exchange to perform a standardized series of tasks, some aspects of which relate in some way to tax credits. It is entirely natural for slight mismatches to occur when, as here, lawmakers draft “a single statutory provision” to cover “different kinds” of situations. Robers v. United States, 572 U. S. ___, ___ (2014) (slip op., at 4). Lawmakers need not, and often do not, “write extra language specifically exempting, phrase by phrase, applications in respect to which a portion of a phrase is not needed.” Ibid.

Roaming even farther afield from §36B, the Court turns to the Act’s provisions about “qualified individuals.” Ante, at 10–11. Qualified individuals receive favored treatment on Exchanges, although customers who are not qualified individuals may also shop there. See Halbig v. Burwell, 758 F. 3d 390, 404–405 (CADC 2014). The Court claims that the Act must equate federal and state establishment of Exchanges when it defines a qualified individual as someone who (among other things) lives in the “State that established the Exchange,” 42 U. S. C. §18032(f )(1)(A). Otherwise, the Court says, there would be no qualified individuals on federal Exchanges, contradicting (for example) the provision requiring every Exchange to takethe “ ‘interests of qualified individuals’ ” into accountwhen selecting health plans. Ante, at 11 (quoting §18031(e)(1)(b)). Pure applesauce. Imagine that a university sends around a bulletin reminding every professor to take the “interests of graduate students” into account when setting office hours, but that some professors teach only undergraduates. Would anybody reason that the bulletin implicitly presupposes that every professor has “graduate students,” so that “graduate students” must really mean “graduate or undergraduate students”? Surely not. Just as one naturally reads instructions aboutgraduate students to be inapplicable to the extent a particular professor has no such students, so too would one naturally read instructions about qualified individuals to be inapplicable to the extent a particular Exchange has no such individuals. There is no need to rewrite the term “State that established the Exchange” in the definition of “qualified individual,” much less a need to rewrite the separate term “Exchange established by the State” in a separate part of the Act.

Least convincing of all, however, is the Court’s attempt to uncover support for its interpretation in “the structure of Section 36B itself.” Ante, at 19. The Court finds it strange that Congress limited the tax credit to state Exchanges in the formula for calculating the amount of the credit, rather than in the provision defining the range of taxpayers eligible for the credit. Had the Court bothered to look at the rest of the Tax Code, it would have seen that the structure it finds strange is in fact quite common. Consider, for example, the many provisions that initially make taxpayers of all incomes eligible for a tax credit, only to provide later that the amount of the credit is zero if the taxpayer’s income exceeds a specified threshold. See, e.g., 26 U. S. C. §24 (child tax credit); §32 (earned-income tax credit); §36 (first-time-homebuyer tax credit). Or consider, for an even closer parallel, a neighboring provision that initially makes taxpayers of all States eligible for a credit, only to provide later that the amount of the credit may be zero if the taxpayer’s State does not satisfy certain requirements. See §35 (health-insurance-costs tax credit). One begins to get the sense that the Court’s insistence on reading things in context applies to “established by the State,” but to nothing else.

For what it is worth, lawmakers usually draft tax-credit provisions the way they do—i.e., the way they drafted §36B—because the mechanics of the credit require it. Many Americans move to new States in the middle of the year. Mentioning state Exchanges in the definition of “coverage month”—rather than (as the Court proposes) in the provisions concerning taxpayers’ eligibility for the credit—accounts for taxpayers who live in a State with a state Exchange for a part of the year, but a State with a federal Exchange for the rest of the year. In addition, §36B awards a credit with respect to insurance plans “which cover the taxpayer, the taxpayer’s spouse, or any dependent . . . of the taxpayer and which were enrolled in through an Exchange established by the State.” §36B(b)(2)(A) (emphasis added). If Congress had mentioned state Exchanges in the provisions discussing taxpayers’ eligibility for the credit, a taxpayer who buys insurance from a federal Exchange would get no money, even if he has a spouse or dependent who buys insurance from a state Exchange—say a child attending college in a different State. It thus makes perfect sense for “Exchange established by the State” to appear where it does, rather than where the Court suggests. Even if that were not so, of course, its location would not make it any less clear.

The Court has not come close to presenting the compelling contextual case necessary to justify departing from the ordinary meaning of the terms of the law. Quite the contrary, context only underscores the outlandishness of the Court’s interpretation. Reading the Act as a whole leaves no doubt about the matter: “Exchange established by the State” means what it looks like it means.

III

For its next defense of the indefensible, the Court turns to the Affordable Care Act’s design and purposes. As relevant here, the Act makes three major reforms. The guaranteed-issue and community-rating requirements prohibit insurers from considering a customer’s health when deciding whether to sell insurance and how much to charge, 42 U. S. C. §§300gg, 300gg–1; its famous individ-ual mandate requires everyone to maintain insurance coverage or to pay what the Act calls a “penalty,” 26 U. S. C. §5000A(b)(1), and what we have nonetheless called a tax, see National Federation of Independent Business v. Sebelius, 567 U. S. ___, ___ (2012) (slip op., at 39); and its tax credits help make insurance more affordable. The Court reasons that Congress intended these three reforms to “work together to expand insurance coverage”; and because the first two apply in every State, so must the third. Ante, at 16.

This reasoning suffers from no shortage of flaws. To begin with, “even the most formidable argument concerning the statute’s purposes could not overcome the clarity [of ] the statute’s text.” Kloeckner v. Solis, 568 U. S. ___, ___, n. 4 (2012) (slip op., at 14, n. 4). Statutory design and purpose matter only to the extent they help clarify an otherwise ambiguous provision. Could anyone maintain with a straight face that §36B is unclear? To mention just the highlights, the Court’s interpretation clashes with a statutory definition, renders words inoperative in at least seven separate provisions of the Act, overlooks the contrast between provisions that say “Exchange” and those that say “Exchange established by the State,” gives the same phrase one meaning for purposes of tax credits but an entirely different meaning for other purposes, and (let us not forget) contradicts the ordinary meaning of the words Congress used. On the other side of the ledger, the Court has come up with nothing more than a general provision that turns out to be controlled by a specific one, a handful of clauses that are consistent with either understanding of establishment by the State, and a resemblance between the tax-credit provision and the rest of the Tax Code. If that is all it takes to make something ambiguous, everything is ambiguous.

Having gone wrong in consulting statutory purpose at all, the Court goes wrong again in analyzing it. The purposes of a law must be “collected chiefly from its words,” not “from extrinsic circumstances.” Sturges v. Crowninshield, 4 Wheat. 122, 202 (1819) (Marshall, C. J.). Only by concentrating on the law’s terms can a judge hope to uncover the scheme of the statute, rather than some other scheme that the judge thinks desirable. Like it or not, the express terms of the Affordable Care Act make only two of the three reforms mentioned by the Court applicable in States that do not establish Exchanges. It is perfectly possible for them to operate independently of tax credits. The guaranteed-issue and community-rating requirements continue to ensure that insurance companies treat all customers the same no matter their health, and the individual mandate continues to encourage people to maintain coverage, lest they be “taxed.”

The Court protests that without the tax credits, the number of people covered by the individual mandate shrinks, and without a broadly applicable individual mandate the guaranteed-issue and community-rating requirements “would destabilize the individual insurance market.” Ante, at 15. If true, these projections would show only that the statutory scheme contains a flaw; they would not show that the statute means the opposite of what it says. Moreover, it is a flaw that appeared as well in other parts of the Act. A different title established a long-term-care insurance program with guaranteed-issue and community-rating requirements, but without an individual mandate or subsidies. §§8001–8002, 124Stat. 828–847 (2010). This program never came into effect “only because Congress, in response to actuarial analyses predicting that the [program] would be fiscally unsustainable, repealed the provision in 2013.” Halbig, 758 F. 3d, at 410. How could the Court say that Congress would never dream of combining guaranteed-issue and community-rating requirements with a narrow individual mandate, when it combined those requirements with no individual mandate in the context of long-term-care insurance?

Similarly, the Department of Health and Human Services originally interpreted the Act to impose guaranteed-issue and community-rating requirements in the Federal Territories, even though the Act plainly does not make the individual mandate applicable there. Ibid.; see 26 U. S. C. §5000A(f)(4); 42 U. S. C. §201(f). “This combination, predictably, [threw] individual insurance markets in the territories into turmoil.” Halbig, supra, at 410. Responding to complaints from the Territories, the Department at first insisted that it had “no statutory authority” to address the problem and suggested that the Territories “seek legislative relief from Congress” instead. Letter from G. Cohen, Director of the Center for Consumer Information and Insurance Oversight, to S. Igisomar, Secretary of Commerce of the Commonwealth of Northern Mariana Islands (July 12, 2013). The Department changed its mind a year later, after what it described as “a careful review of [the] situation and the relevant statutory language.” Letter from M. Tavenner, Administrator of the Centers for Medicare and Medicaid Services, to G. Francis, Insurance Commissioner of the Virgin Islands (July 16, 2014). How could the Court pronounce it “implausible” for Congress to have tolerated instability in insurance markets in States with federal Exchanges, ante, at 17, when even the Government maintained until recently that Congress did exactly that in American Samoa, Guam, the Northern Mariana Islands, Puerto Rico, and the Virgin Islands?

Compounding its errors, the Court forgets that it is no more appropriate to consider one of a statute’s purposes in isolation than it is to consider one of its words that way. No law pursues just one purpose at all costs, and no statutory scheme encompasses just one element. Most relevant here, the Affordable Care Act displays a congressional preference for state participation in the establishment of Exchanges: Each State gets the first opportunity to set up its Exchange, 42 U. S. C. §18031(b); States that take up the opportunity receive federal funding for “activities . . . related to establishing” an Exchange, §18031(a)(3); and the Secretary may establish an Exchange in a State only as a fallback, §18041(c). But setting up and running an Exchange involve significant burdens—meeting strict deadlines, §18041(b), implementing requirements related to the offering of insurance plans, §18031(d)(4), setting up outreach programs, §18031(i), and ensuring that the Exchange is self-sustaining by 2015, §18031(d)(5)(A). A State would have much less reason to take on these burdens if its citizens could receive tax credits no matter who establishes its Exchange. (Now that the Internal Revenue Service has interpreted §36B to authorize tax credits everywhere, by the way, 34 States have failed to set up their own Exchanges. Ante, at 6.) So even if making credits available on all Exchanges advances the goal of improving healthcare markets, it frustrates the goal of encouraging state involvement in the implementation of the Act. This is what justifies going out of our way to read “established by the State” to mean “established by the State or not established by the State”?

Worst of all for the repute of today’s decision, the Court’s reasoning is largely self-defeating. The Court predicts that making tax credits unavailable in States that do not set up their own Exchanges would cause disastrous economic consequences there. If that is so, however, wouldn’t one expect States to react by setting up their own Exchanges? And wouldn’t that outcome satisfy two of the Act’s goals rather than just one: enabling the Act’s reforms to work and promoting state involvement in the Act’s implementation? The Court protests that the very existence of a federal fallback shows that Congress expected that some States might fail to set up their own Exchanges. Ante, at 19. So it does. It does not show, however, that Congress expected the number of recalcitrant States to be particularly large. The more accurate the Court’s dire economic predictions, the smaller that number is likely to be. That reality destroys the Court’s pretense that applying the law as written would imperil “the viability of the entire Affordable Care Act.” Ante, at 20. All in all, the Court’s arguments about the law’s purpose and design are no more convincing than its arguments about context.

IV

Perhaps sensing the dismal failure of its efforts to show that “established by the State” means “established by the State or the Federal Government,” the Court tries to palm off the pertinent statutory phrase as “inartful drafting.” Ante, at 14. This Court, however, has no free-floating power “to rescue Congress from its drafting errors.” Lamie v. United States Trustee, 540 U. S. 526, 542 (2004) (internal quotation marks omitted). Only when it is patently obvious to a reasonable reader that a drafting mistake has occurred may a court correct the mistake. The occurrence of a misprint may be apparent from the face of the law, as it is where the Affordable Care Act “creates three separate Section 1563s.” Ante, at 14. But the Court does not pretend that there is any such indication of a drafting error on the face of §36B. The occurrence of a misprint may also be apparent because a provision decrees an absurd result—a consequence “so monstrous, that all mankind would, without hesitation, unite in rejecting the application.” Sturges, 4 Wheat., at 203. But §36B does not come remotely close to satisfying that demanding standard. It is entirely plausible that tax credits were restricted to state Exchanges deliberately—for example,in order to encourage States to establish their own Exchanges. We therefore have no authority to dismiss the terms of the law as a drafting fumble.

Let us not forget that the term “Exchange established by the State” appears twice in §36B and five more times in other parts of the Act that mention tax credits. What are the odds, do you think, that the same slip of the pen occurred in seven separate places? No provision of the Act—none at all—contradicts the limitation of tax credits to state Exchanges. And as I have already explained, uses of the term “Exchange established by the State” beyond the context of tax credits look anything but accidental. Supra, at 6. If there was a mistake here, context suggests it was a substantive mistake in designing this part of the law, not a technical mistake in transcribing it.

V

The Court’s decision reflects the philosophy that judges should endure whatever interpretive distortions it takes in order to correct a supposed flaw in the statutory machinery. That philosophy ignores the American people’s decision to give Congress “[a]ll legislative Powers” enumerated in the Constitution. Art. I, §1. They made Congress, not this Court, responsible for both making laws and mending them. This Court holds only the judicial power—the power to pronounce the law as Congress has enacted it. We lack the prerogative to repair laws that do not work out in practice, just as the people lack the ability to throw us out of office if they dislike the solutions we concoct. We must always remember, therefore, that “[o]ur task is to apply the text, not to improve upon it.” Pavelic & LeFlore v. Marvel Entertainment Group, Div. of Cadence Industries Corp., 493 U. S. 120, 126 (1989) .

Trying to make its judge-empowering approach seem respectful of congressional authority, the Court asserts that its decision merely ensures that the Affordable Care Act operates the way Congress “meant [it] to operate.” Ante, at 17. First of all, what makes the Court so sure that Congress “meant” tax credits to be available everywhere? Our only evidence of what Congress meant comes from the terms of the law, and those terms show beyond all question that tax credits are available only on state Exchanges. More importantly, the Court forgets that ours is a government of laws and not of men. That means we are governed by the terms of our laws, not by the unen-acted will of our lawmakers. “If Congress enacted into law something different from what it intended, then it should amend the statute to conform to its intent.” Lamie, supra, at 542. In the meantime, this Court “has no roving license . . . to disregard clear language simply on the view that . . . Congress ‘must have intended’ something broader.” Bay Mills, 572 U. S., at ___ (slip op., at 11).

Even less defensible, if possible, is the Court’s claim that its interpretive approach is justified because this Act “does not reflect the type of care and deliberation that one might expect of such significant legislation.” Ante, at 14–15. It is not our place to judge the quality of the care and deliberation that went into this or any other law. A law enacted by voice vote with no deliberation whatever is fully as binding upon us as one enacted after years of study, months of committee hearings, and weeks of debate. Much less is it our place to make everything come out right when Congress does not do its job properly. It is up to Congress to design its laws with care, and it is up to the people to hold them to account if they fail to carry out that responsibility.

Rather than rewriting the law under the pretense of interpreting it, the Court should have left it to Congress to decide what to do about the Act’s limitation of tax credits to state Exchanges. If Congress values above everything else the Act’s applicability across the country, it could make tax credits available in every Exchange. If it prizes state involvement in the Act’s implementation, it could continue to limit tax credits to state Exchanges while taking other steps to mitigate the economic consequences predicted by the Court. If Congress wants to accommodate both goals, it could make tax credits available everywhere while offering new incentives for States to set up their own Exchanges. And if Congress thinks that the present design of the Act works well enough, it could do nothing. Congress could also do something else alto-gether, entirely abandoning the structure of the Affordable Care Act. The Court’s insistence on making a choice that should be made by Congress both aggrandizes judicial power and encourages congressional lassitude.

Just ponder the significance of the Court’s decision to take matters into its own hands. The Court’s revision of the law authorizes the Internal Revenue Service to spend tens of billions of dollars every year in tax credits on federal Exchanges. It affects the price of insurance for millions of Americans. It diminishes the participation of the States in the implementation of the Act. It vastly expands the reach of the Act’s individual mandate, whose scope depends in part on the availability of credits. What a parody today’s decision makes of Hamilton’s assurances to the people of New York: “The legislature not only commands the purse but prescribes the rules by which the duties and rights of every citizen are to be regulated. The judiciary, on the contrary, has no influence over . . . the purse; no direction . . . of the wealth of society, and can take no active resolution whatever. It may truly be said to have neither force nor will but merely judgment.” The Federalist No. 78, p. 465 (C. Rossiter ed. 1961).

*  *  *

Today’s opinion changes the usual rules of statutory interpretation for the sake of the Affordable Care Act. That, alas, is not a novelty. In National Federation of Independent Business v. Sebelius, 567 U. S. ___, this Court revised major components of the statute in order to save them from unconstitutionality. The Act that Congress passed provides that every individual “shall” maintain insurance or else pay a “penalty.” 26 U. S. C. §5000A. This Court, however, saw that the Commerce Clause does not authorize a federal mandate to buy health insurance. So it rewrote the mandate-cum-penalty as a tax. 567 U. S., at ___–___ (principal opinion) (slip op., at 15–45). The Act that Congress passed also requires every State to accept an expansion of its Medicaid program, or else risk losing all Medicaid funding. 42 U. S. C. §1396c. This Court, however, saw that the Spending Clause does not authorize this coercive condition. So it rewrote the law to withhold only the incremental funds associated with the Medicaid expansion. 567 U. S., at ___–___ (principal opinion) (slip op., at 45–58). Having transformed two major parts of the law, the Court today has turned its attention to a third. The Act that Congress passed makes tax credits available only on an “Exchange established by the State.” This Court, however, concludes that this limitation would prevent the rest of the Act from working as well as hoped. So it rewrites the law to make tax credits available everywhere. We should start calling this law SCOTUScare.

Perhaps the Patient Protection and Affordable Care Act will attain the enduring status of the Social Security Act or the Taft-Hartley Act; perhaps not. But this Court’s two decisions on the Act will surely be remembered through the years. The somersaults of statutory interpretation they have performed (“penalty” means tax, “further [Medicaid] payments to the State” means only incremental Medicaid payments to the State, “established by the State” means not established by the State) will be cited by litigants endlessly, to the confusion of honest jurisprudence. And the cases will publish forever the discouraging truth that the Supreme Court of the United States favors some laws over others, and is prepared to do whatever it takes to uphold and assist its favorites.

I dissent.