Barr v. American Assn. of Political Consultants, Inc.

591 U.S. 610, 140 S. Ct. 2335, 207 L. Ed. 2d 784

• Court: Supreme Court of the United States
• Decided: July 6, 2020
• Docket: 19-631
• Status: Published

Opinion

B

Plaintiffs next focus on ordinary severability principles. Applying those principles, the question before the Court is whether (i) to invalidate the entire 1991 robocall restriction, as plaintiffs want, or (ii) to invalidate just the 2015 government-debt exception and sever it from the remainder of the statute, as the Government wants.

We agree with the Government that we must invalidate the 2015 government-debt exception and sever that exception from the remainder of the statute. To explain why, we begin with general severability principles and then apply those principles to this case.

When enacting a law, Congress sometimes expressly addresses severability. For example, Congress may include a severability clause in the law, making clear that the unconstitutionality of one provision does not affect the rest of the law. See, e.g. , 12 U.S.C. § 5302 ; 15 U.S.C. § 78gg ; 47 U.S.C. § 608 . Alternatively, Congress may include a nonseverability clause, making clear that the unconstitutionality of one provision means the invalidity of some or all of the remainder of the law, to the extent specified in the text of the nonseverability clause. See, e.g. , 4 U.S.C. § 125 ; note following 42 U.S.C. § 300aa-1 ; 94 Stat. 1797 .

When Congress includes an express severability or nonseverability clause in the relevant statute, the judicial inquiry is straightforward. At least absent extraordinary circumstances, the Court should adhere to the text of the severability or nonseverability clause. That is because a severability or nonseverability clause leaves no doubt about what the enacting Congress wanted if one provision of the law were later declared unconstitutional. A severability clause indicates "that Congress did not intend the validity of the statute in question to depend on the validity of the constitutionally offensive provision." Alaska Airlines, Inc. v. Brock , 480 U.S. 678 , 686, 107 S.Ct. 1476 , 94 L.Ed.2d 661 (1987). And a nonseverability clause does the opposite.

On occasion, a party will nonetheless ask the Court to override the text of a severability or nonseverability clause on the ground that the text does not reflect Congress's "actual intent" as to severability. That kind of argument may have carried some force back when courts paid less attention to statutory text as the definitive expression of Congress's will. But courts today zero in on the precise statutory text and, as a result, courts hew closely to the text of severability or nonseverability clauses. See Seila Law LLC v. Consumer Financial Protection Bureau , --- U.S. ----, ----, 140 S.Ct. 2183 , 2209, --- L.Ed.2d ---- (2020), 2020 WL 3492641 (plurality opinion); cf. Milner v. Department of Navy , 562 U.S. 562 , 569-573, 131 S.Ct. 1259 , 179 L.Ed.2d 268 (2011). 6

*2350 Of course, when enacting a law, Congress often does not include either a severability clause or a nonseverability clause.

In those cases, it is sometimes said that courts applying severability doctrine should search for other indicia of congressional intent. For example, some of the Court's cases declare that courts should sever the offending provision unless "the statute created in its absence is legislation that Congress would not have enacted." Alaska Airlines , 480 U.S., at 685 , 107 S.Ct. 1476 . But experience shows that this formulation often leads to an analytical dead end. That is because courts are not well equipped to imaginatively reconstruct a prior Congress's hypothetical intent. In other words, absent a severability or nonseverability clause, a court often cannot really know what the two Houses of Congress and the President from the time of original enactment of a law would have wanted if one provision of a law were later declared unconstitutional.

The Court's cases have instead developed a strong presumption of severability. The Court presumes that an unconstitutional provision in a law is severable from the remainder of the law or statute. For example, in Free Enterprise Fund v. Public Company Accounting Oversight Bd. , the Court set forth the "normal rule": "Generally speaking, when confronting a constitutional flaw in a statute, we try to limit the solution to the problem, severing any problematic portions while leaving the remainder intact." 561 U.S. 477 , 508, 130 S.Ct. 3138 , 177 L.Ed.2d 706 (2010) (internal quotation marks omitted); see also Seila Law , --- U.S. at ----, 140 S.Ct. at 2209, ante , at 32, 2020 WL 3492641 (same). In Regan v. Time, Inc. , the plurality opinion likewise described a "presumption" in "favor of severability" and stated that the Court should "refrain from invalidating more of the statute than is necessary." 468 U.S. 641 , 652-653, 104 S.Ct. 3262 , 82 L.Ed.2d 487 (1984).

The Court's power and preference to partially invalidate a statute in that fashion has been firmly established since Marbury v. Madison . There, the Court invalidated part of § 13 of the Judiciary Act of 1789. 1 Cranch 137 , 179-180, 2 L.Ed. 60 (1803). The Judiciary Act did not contain a severability clause. But the Court did not proceed to invalidate the entire Judiciary Act. As Chief Justice Marshall later explained, if any part of an Act is "unconstitutional, the provisions of that part may be disregarded while full effect will be given to such as are not repugnant to the constitution of the United States." Bank of Hamilton v. Lessee of Dudley , 2 Pet. 492 , 526, 7 L.Ed. 496 (1829) ; see also Dorchy v. Kansas , 264 U.S. 286 , 289-290, 44 S.Ct. 323 , 68 L.Ed. 686 (1924) ("A statute bad in part is not necessarily void in its entirety. Provisions within the legislative power may stand if separable from the bad"); Loeb v. Columbia Township Trustees , 179 U.S. 472 , 490, 21 S.Ct. 174 , 45 L.Ed. 280 (1900) ("one section of a statute may be repugnant to the Constitution without rendering the whole act void").

From Marbury v. Madison to the present, apart from some isolated detours mostly in the late 1800s and early 1900s, the Court's remedial preference after finding a provision of a federal law unconstitutional has been to salvage rather than destroy the rest of the law passed by Congress and signed by the President. The Court's precedents reflect a decisive preference for surgical severance rather than *2351 wholesale destruction, even in the absence of a severability clause.

The Court's presumption of severability supplies a workable solution-one that allows courts to avoid judicial policymaking or de facto judicial legislation in determining just how much of the remainder of a statute should be invalidated. 7 The presumption also reflects the confined role of the Judiciary in our system of separated powers-stated otherwise, the presumption manifests the Judiciary's respect for Congress's legislative role by keeping courts from unnecessarily disturbing a law apart from invalidating the provision that is unconstitutional. Furthermore, the presumption recognizes that plaintiffs who successfully challenge one provision of a law may lack standing to challenge other provisions of that law. See Murphy v. National Collegiate Athletic Assn. , 584 U. S. ----, ---- - ----, 138 S.Ct. 1461 , 1487, 200 L.Ed.2d 854 (2018) (THOMAS, J., concurring).

Those and other considerations, taken together, have steered the Court to a presumption of severability. Applying the presumption, the Court invalidates and severs unconstitutional provisions from the remainder of the law rather than razing whole statutes or Acts of Congress. Put in common parlance, the tail (one unconstitutional provision) does not wag the dog (the rest of the codified statute or the Act as passed by Congress). Constitutional 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. If the rule were otherwise, the entire Judiciary Act of 1789 would be invalid as a consequence of Marbury v. Madison . 8

*2352 Before severing a provision and leaving the remainder of a law intact, the Court must determine that the remainder of the statute is "capable of functioning independently" and thus would be "fully operative" as a law. Seila Law , --- U.S. at ----, 140 S.Ct. at 2209, ante , at 33, 2020 WL 3492641 ; see Murphy , 584 U. S., at ---- - ----, 138 S.Ct., at 1481-1484 . But it is fairly unusual for the remainder of a law not to be operative. 9

We next apply those general severability principles to this case.

Recall how this statute came together. Passed by Congress and signed by President Franklin Roosevelt in 1934, the Communications Act is codified in Title 47 of the U. S. Code. The TCPA of 1991 amended the Communications Act by adding the robocall restriction, which is codified at § 227(b)(1)(A)(iii) of Title 47. The Bipartisan Budget Act of 2015 then amended the Communications Act by adding the government-debt exception, which is codified along with the robocall restriction at § 227(b)(1)(A)(iii) of Title 47.

Since 1934, the Communications Act has contained an express severability clause: "If any provision of this chapter or the application thereof to any person or circumstance is held invalid, the remainder of the chapter and the application of such provision to other persons or circumstances shall not be affected thereby." 47 U.S.C. § 608 (emphasis added). The "chapter" referred to in the severability clause is Chapter 5 of Title 47. And Chapter 5 in turn encompasses § 151 to § 700 of Title 47, and therefore covers § 227 of Title 47, the provision with the robocall restriction and the government-debt exception. 10

Enacted in 2015, the government-debt exception added an unconstitutional discriminatory exception to the robocall restriction. The text of the severability clause squarely covers the unconstitutional government-debt exception and requires that we sever it.

To get around the text of the severability clause, plaintiffs point out that the Communications Act's severability clause was enacted in 1934, long before the TCPA's 1991 robocall restriction and the 2015 government-debt exception. But a severability clause must be interpreted according to its terms, regardless of when Congress enacted it. See n. 6, supra .

Even if the severability clause did not apply to the government-debt provision at issue in this case (or even if there were no severability clause in the Communications Act), we would apply the presumption of severability as described and applied in cases such as Free Enterprise Fund . And *2353 under that presumption, we likewise would sever the 2015 government-debt exception, the constitutionally offending provision.

With the government-debt exception severed, the remainder of the law is capable of functioning independently and thus would be fully operative as a law. Indeed, the remainder of the robocall restriction did function independently and fully operate as a law for 20-plus years before the government-debt exception was added in 2015.

The Court's precedents further support severing the 2015 government-debt exception. The Court has long applied severability principles in cases like this one, where Congress added an unconstitutional amendment to a prior law. In those cases, the Court has treated the original, pre-amendment statute as the "valid expression of the legislative intent." Frost v. Corporation Comm'n of Okla ., 278 U.S. 515 , 526-527, 49 S.Ct. 235 , 73 L.Ed. 483 (1929). The Court has severed the "exception introduced by amendment," so that "the original law stands without the amendatory exception." Truax v. Corrigan , 257 U.S. 312 , 342, 42 S.Ct. 124 , 66 L.Ed. 254 (1921).

For example, in Eberle v. Michigan , the Court held that "discriminatory wine-and-cider amendments" added in 1899 and 1903 were severable from the underlying 1889 state law generally prohibiting the manufacture of alcohol. 232 U.S. 700 , 704-705, 34 S.Ct. 464 , 58 L.Ed. 803 (1914). In Truax , the Court ruled that a 1913 amendment prohibiting Arizona courts from issuing injunctions in labor disputes was invalid and severable from the underlying 1901 law authorizing Arizona courts to issue injunctions generally. 257 U.S., at 341-342 , 42 S.Ct. 124 . In Frost , the Court concluded that a 1925 amendment exempting certain corporations from making a showing of "public necessity" in order to obtain a cotton gin license was invalid and severable from the 1915 law that required that showing. 278 U.S., at 525-528 , 49 S.Ct. 235 . Echoing Marbury , the Court in Frost explained that an unconstitutional statutory amendment "is a nullity" and "void" when enacted, and for that reason has no effect on the original statute. 278 U.S., at 526-527 , 49 S.Ct. 235 (internal quotation marks omitted). 11

Similarly, in 1932, Congress enacted the Federal Kidnaping Act, and then in 1934, added a death penalty provision to the Act. The death penalty provision was later declared unconstitutional by this Court. In considering severability, the Court stated that the "law as originally enacted in 1932 contained no capital punishment provision." United States v. Jackson , 390 U.S. 570 , 586, 88 S.Ct. 1209 , 20 L.Ed.2d 138 (1968). And when Congress amended the Act in 1934 to add the death penalty, "the statute was left substantially unchanged in every other respect." Id. , at 587-588, 88 S.Ct. 1209 . The Court found it "difficult to imagine a more compelling case for severability." Id. , at 589, 88 S.Ct. 1209 . So too here.

In sum, the text of the Communications Act's severability clause requires that the *2354 Court sever the 2015 government-debt exception from the remainder of the statute. And even if the text of the severability clause did not apply here, the presumption of severability would require that the Court sever the 2015 government-debt exception from the remainder of the statute.

One final severability wrinkle remains. This is an equal-treatment case, and equal-treatment cases can sometimes pose complicated severability questions.

The "First Amendment is a kind of Equal Protection Clause for ideas." Williams-Yulee v. Florida Bar , 575 U.S. 433 , 470, 135 S.Ct. 1656 , 191 L.Ed.2d 570 (2015) (Scalia, J., dissenting). And Congress violated that First Amendment equal-treatment principle in this case by favoring debt-collection robocalls and discriminating against political and other robocalls.

When the constitutional violation is unequal treatment, as it is here, a court theoretically can cure that unequal treatment either by extending the benefits or burdens to the exempted class, or by nullifying the benefits or burdens for all. See, e.g. , Heckler v. Mathews , 465 U.S. 728 , 740, 104 S.Ct. 1387 , 79 L.Ed.2d 646 (1984). Here, for example, the Government would prefer to cure the unequal treatment by extending the robocall restriction and thereby proscribing nearly all robocalls to cell phones. By contrast, plaintiffs want to cure the unequal treatment by nullifying the robocall restriction and thereby allowing all robocalls to cell phones.

When, as here, the Court confronts an equal-treatment constitutional violation, the Court generally applies the same commonsense severability principles described above. If the statute contains a severability clause, the Court typically severs the discriminatory exception or classification, and thereby extends the relevant statutory benefits or burdens to those previously exempted, rather than nullifying the benefits or burdens for all. In light of the presumption of severability, the Court generally does the same even in the absence of a severability clause. The Court's precedents reflect that preference for extension rather than nullification. See, e.g. , Sessions v. Morales-Santana , 582 U. S. ----, ----, 137 S.Ct. 1678 , 1699, 198 L.Ed.2d 150 (2017) ; Califano v. Westcott , 443 U.S. 76 , 89-91, 99 S.Ct. 2655 , 61 L.Ed.2d 382 (1979) ; Califano v. Goldfarb , 430 U.S. 199 , 202-204, 213-217, 97 S.Ct. 1021 , 51 L.Ed.2d 270 (1977) (plurality opinion); Jimenez v. Weinberger , 417 U.S. 628 , 637-638, 94 S.Ct. 2496 , 41 L.Ed.2d 363 (1974) ; Department of Agriculture v. Moreno , 413 U.S. 528 , 529, 537-538, 93 S.Ct. 2821 , 37 L.Ed.2d 782 (1973) ; Frontiero v. Richardson , 411 U.S. 677 , 678-679, 690-691, 93 S.Ct. 1764 , 36 L.Ed.2d 583 (1973) (plurality opinion); Welsh v. United States , 398 U.S. 333 , 361-367, 90 S.Ct. 1792 , 26 L.Ed.2d 308 (1970) (Harlan, J., concurring in result).

To be sure, some equal-treatment cases can raise complex questions about whether it is appropriate to extend benefits or burdens, rather than nullifying the benefits or burdens. See, e.g. , Morales-Santana , 582 U. S. ----, 137 S.Ct. 1678 , 198 L.Ed.2d 150 . For example, there can be due process, fair notice, or other independent constitutional barriers to extension of benefits or burdens. Cf. Miller v. Albright , 523 U.S. 420 , 458-459, 118 S.Ct. 1428 , 140 L.Ed.2d 575 (1998) (Scalia, J., concurring in judgment); see generally Ginsburg, Some Thoughts on Judicial Authority to Repair Unconstitutional Legislation, 28 Clev. St. L. Rev. 301 (1979). There also can be knotty questions about what is the exception and what is the rule. But here, we need not tackle all of the possible *2355 hypothetical applications of severability doctrine in equal-treatment cases. The government-debt exception is a relatively narrow exception to the broad robocall restriction, and severing the government-debt exception does not raise any other constitutional problems.

Plaintiffs insist, however, that a First Amendment equal-treatment case is different. According to plaintiffs, a court should not cure "a First Amendment violation by outlawing more speech." Brief for Respondents 34. The implicit premise of that argument is that extending the robocall restriction to debt-collection robocalls would be unconstitutional. But that is wrong. A generally applicable robocall restriction would be permissible under the First Amendment. Extending the robocall restriction to those robocalls raises no First Amendment problem. So the First Amendment does not tell us which way to cure the unequal treatment in this case. Therefore, we apply traditional severability principles. And as we have explained, severing the 2015 government-debt exception cures the unequal treatment and constitutes the proper result under the Court's traditional severability principles. In short, the correct result in this case is to sever the 2015 government-debt exception and leave in place the longstanding robocall restriction. 12

Justice GORSUCH's well-stated separate opinion makes a number of important points that warrant this respectful response.

Justice GORSUCH suggests that our decision provides "no relief" to plaintiffs. Post , at 2366. We disagree. Plaintiffs want to be able to make political robocalls to cell phones, and they have not received that relief. But the First Amendment complaint at the heart of their suit was unequal treatment. Invalidating and severing the government-debt exception fully addresses that First Amendment injury. 13 Justice GORSUCH further suggests that plaintiffs may lack standing to challenge the government-debt exception, because that exception merely favors others. See ibid . But the Court has squarely held that a plaintiff who suffers unequal treatment has standing to challenge a discriminatory exception that favors others. See Heckler v. Mathews , 465 U.S., at 737-740 , 104 S.Ct. 1387 (a plaintiff who suffers unequal treatment has standing to seek "withdrawal of benefits from the favored class"); see also Northeastern Fla. Chapter, Associated Gen. Contractors of America v. Jacksonville , 508 U.S. 656 , 666, 113 S.Ct. 2297 , 124 L.Ed.2d 586 (1993) ("The 'injury in fact' in an equal protection case of this variety is the denial of equal treatment resulting from the imposition of the barrier, not the ultimate inability to obtain the benefit").

Justice GORSUCH also objects that our decision today "harms strangers to this suit" by eliminating favorable treatment *2356 for debt collectors. Post , at 2366 But that is necessarily true in many cases where a court cures unequal treatment by, for example, extending a burden or nullifying a benefit. See, e.g ., Morales - Santana , 582 U. S., at ----, 137 S.Ct., at 1701 (curing unequal treatment of children born to unwed U. S.-citizen fathers by extending a burden to children of unwed U. S.-citizen mothers); Orr v. Orr , 374 So.2d 895 , 896-897 (Ala. Civ. App. 1979) (extending alimony obligations to women after a male plaintiff successfully challenged Alabama's discriminatory alimony statute in this Court).

Moreover, Justice GORSUCH's approach to this case would not solve the problem of harming strangers to this suit; it would just create a different and much bigger problem. His proposed remedy of injunctive relief, plus stare decisis , would in effect allow all robocalls to cell phones-notwithstanding Congress's decisive choice to prohibit most robocalls to cell phones. That is not a judicially modest approach but is more of a wolf in sheep's clothing. That approach would disrespect the democratic process, through which the people's representatives have made crystal clear that robocalls must be restricted. Justice GORSUCH's remedy would end up harming a different and far larger set of strangers to this suit-the tens of millions of consumers who would be bombarded every day with nonstop robocalls notwithstanding Congress's clear prohibition of those robocalls.

Justice GORSUCH suggests more broadly that severability doctrine may need to be reconsidered. But when and how? As the saying goes, John Marshall is not walking through that door. And this Court, in this and other recent decisions, has clarified and refined severability doctrine by emphasizing firm adherence to the text of severability clauses, and underscoring the strong presumption of severability. The doctrine as so refined is constitutionally well-rooted, see, e.g. , Marbury v. Madison , 1 Cranch 137 (Marshall, C. J.), and can be predictably applied. True, there is no magic solution to severability that solves every conundrum, especially in equal-treatment cases, but the Court's current approach as reflected in recent cases such as Free Enterprise Fund and Seila Law is constitutional, stable, predictable, and commonsensical.

* * *

In 1991, Congress enacted a general restriction on robocalls to cell phones. In 2015, Congress carved out an exception that allowed robocalls made to collect government debt. In doing so, Congress favored debt-collection speech over plaintiffs' political speech. We hold that the 2015 government-debt exception added an unconstitutional exception to the law. We cure that constitutional violation by invalidating the 2015 government-debt exception and severing it from the remainder of the statute. The judgment of the U. S. Court of Appeals for the Fourth Circuit is affirmed.

It is so ordered.

Justice SOTOMAYOR, concurring in the judgment.

I agree with much of the partial dissent's explanation that strict scrutiny should not apply to all content-based distinctions. Cf. post , at 2359 - 2362 (BREYER, J., concurring in judgment with respect to severability and dissenting in part). In my view, however, the government-debt exception in 47 U.S.C. § 227 (b) still fails intermediate scrutiny because it is not "narrowly tailored to serve a significant governmental interest." Ward v. Rock Against Racism , 491 U.S. 781 , 791, 109 S.Ct. 2746 , 105 L.Ed.2d 661 (1989) (internal quotation marks omitted). Even under *2357 intermediate scrutiny, the Government has not explained how a debt-collection robocall about a government-backed debt is any less intrusive or could be any less harassing than a debt-collection robocall about a privately backed debt. As the Fourth Circuit noted, the government-debt exception is seriously underinclusive because it permits "many of the intrusive calls that the automated call ban was enacted to prohibit." American Assn. of Political Consultants, Inc. v. FCC , 923 F.3d 159 , 168 (2019) (case below). The Government could have employed far less restrictive means to further its interest in collecting debt, such as "secur[ing] consent from the debtors to make debt-collection calls" or "plac[ing] the calls itself." Id. , at 169, n. 10 ; see also § 227(b)(1)(A). Nor has the Government "sufficiently justified the differentiation between government-debt collection speech and other important categories of robocall speech, such as political speech, charitable fundraising, issue advocacy, commercial advertising, and the like." Ante , at 2347.

Nevertheless, I agree that the offending provision is severable. See ante , at 2343; post , at 2362 - 2363 (opinion of BREYER, J.); see also City of Ladue v. Gilleo , 512 U.S. 43 , 51-53, 114 S.Ct. 2038 , 129 L.Ed.2d 36 (1994) (explaining that an appropriate "solution" to a law that covers "too little speech because its exemptions discriminate on the basis of [the speaker's] messages" could be to "remove" the discrimination).

With those understandings, I concur in the judgment.

Justice BREYER, with whom Justice GINSBURG and Justice KAGAN join, concurring in the judgment with respect to severability and dissenting in part.

A federal statute forbids, with some exceptions, making automatically dialed or prerecorded telephone calls (called robocalls) to cell phones. This case concerns one of these exceptions, which applies to calls "made solely to collect a debt owed to or guaranteed by the United States." 47 U.S.C. § 227 (b)(1)(A)(iii). A majority of the Court holds that the exception violates the Constitution's First Amendment. In my view, it does not.

I

This case concerns the Telephone Consumer Protection Act of 1991. That Act was designed to "protec[t] telephone consumers from th[e] nuisance and privacy invasion" caused by automated and prerecorded phone calls. § 2(12), 105 Stat. 2395 . The Act, among other things, bans almost all robocalls made to cell phones. In particular, it forbids "any call (other than a call made for emergency purposes or made with the prior express consent of the called party) using any automatic telephone dialing system or an artificial or prerecorded voice ... to any telephone number assigned to a ... cellular telephone service." § 3(a) (codified at 47 U.S.C. § 227 (b)(1)(A)(iii) ). The Act delegates authority to the Federal Communications Commission to make certain additional exceptions from that general cell phone robocall restriction. § 227(b)(2)(C).

More than 20 years later, Congress enacted another statute, which created the government-debt exception. The Office of Management and Budget had reported to Congress that in "this time of fiscal constraint ... the Federal Government should ensure that all debt owed to the United States is collected as quickly and efficiently as possible." Office of Management and Budget, Analytical Perspectives, Budget of the U. S. Government, Fiscal Year 2016, p. 128 (2015), https://www.govinfo.gov/content/pkg/BUDGET-2016-PER/pdf/BUDGET-2016-PER.pdf. It recommended that Congress permit "the use *2358 of automatic dialing systems and prerecorded voice messages" to contact "wireless phones in the collection of debt owed to or granted [ sic ] by the United States." Ibid.

Congress adopted that recommendation. It enacted a provision that excepts from the general cell phone robocall restriction any call "made solely to collect a debt owed to or guaranteed by the United States." 129 Stat. 588 ; see also ibid. (categorizing the exception as a "debt collection improvemen[t]" measure). The question here is whether the First Amendment prohibits the Federal Government from enacting that government-debt collection measure.

II

The plurality finds the government-debt exception unconstitutional primarily by applying a logical syllogism: (1) "Content-based laws are subject to strict scrutiny." Ante, at 2341 (citing Reed v. Town of Gilbert , 576 U.S. 155 , 163-164, 135 S.Ct. 2218 , 192 L.Ed.2d 236 (2015) ). (2) The exception is based on "content." Ante, at 2346 - 2347. (3) Hence, the exception is subject to "strict scrutiny." Ante , at 2347. (4) And the Government concedes that the exception cannot survive "strict scrutiny" examination. Ibid .

The problem with that approach, which reflexively applies strict scrutiny to all content-based speech distinctions, is that it is divorced from First Amendment values. This case primarily involves commercial regulation-namely, debt collection. And, in my view, there is no basis here to apply "strict scrutiny" based on "content-discrimination."

To appreciate why, it is important to understand at least one set of values that underlie the First Amendment and the related reasons why courts scrutinize some speech restrictions strictly. The concept is abstract but simple: "We the People of the United States" have created a government of laws enacted by elected representatives. For our government to remain a democratic republic, the people must be free to generate, debate, and discuss both general and specific ideas, hopes, and experiences. The people must then be able to transmit their resulting views and conclusions to their elected representatives, which they may do directly, or indirectly through the shaping of public opinion. The object of that transmission is to influence the public policy enacted by elected representatives. As this Court has explained, "[t]he First Amendment was fashioned to assure unfettered interchange of ideas for the bringing about of political and social changes desired by the people." Meyer v. Grant , 486 U.S. 414 , 421, 108 S.Ct. 1886 , 100 L.Ed.2d 425 (1988) (internal quotation marks omitted). See generally R. Post, Democracy, Expertise, and Academic Freedom: A First Amendment Jurisprudence for the Modern State 1-25 (2012).

In other words, the free marketplace of ideas is not simply a debating society for expressing thought in a vacuum. It is in significant part an instrument for "bringing about ... political and social chang[e]." Meyer , 486 U.S., at 421 , 108 S.Ct. 1886 . The representative democracy that "We the People" have created insists that this be so. See Sorrell v. IMS Health Inc. , 564 U.S. 552 , 583, 131 S.Ct. 2653 , 180 L.Ed.2d 544 (2011) (BREYER, J., dissenting). See generally, e.g., B. Neuborne, Madison's Music: On Reading the First Amendment (2015).

It is thus no surprise that our First Amendment jurisprudence has long reflected these core values. This Court's cases have provided heightened judicial protection for political speech, public forums, and the expression of all viewpoints on any given issue. See, e.g. , *2359 Buckley v. American Constitutional Law Foundation, Inc. , 525 U.S. 182 , 186-187, 119 S.Ct. 636 , 142 L.Ed.2d 599 (1999) (heightened protection for "core political speech"); Rosenberger v. Rector and Visitors of Univ. of Va. , 515 U.S. 819 , 829-830, 115 S.Ct. 2510 , 132 L.Ed.2d 700 (1995) (government discrimination on basis of "particular views taken by speakers on a subject" presumptively unconstitutional); Boos v. Barry , 485 U.S. 312 , 321, 108 S.Ct. 1157 , 99 L.Ed.2d 333 (1988) ("content-based restriction[s] on political speech in a public forum" subject to "most exacting scrutiny" (emphasis deleted)); Perry Ed. Assn. v. Perry Local Educators' Assn. , 460 U.S. 37 , 45-46, 103 S.Ct. 948 , 74 L.Ed.2d 794 (1983) (content-based exclusions in public forums subject to strict scrutiny). These cases reflect the straightforward principle that "governments must not be allowed to choose which issues are worth discussing or debating." Reed , 576 U.S., at 182 , 135 S.Ct. 2218 (KAGAN, J., concurring in judgment) (internal quotation marks omitted).

From a democratic perspective, however, it is equally important that courts not use the First Amendment in a way that would threaten the workings of ordinary regulatory programs posing little threat to the free marketplace of ideas enacted as result of that public discourse. As a general matter, the strictest scrutiny should not apply indiscriminately to the very "political and social changes desired by the people"-that is, to those government programs which the "unfettered interchange of ideas" has sought to achieve. Meyer , 486 U.S., at 421 , 108 S.Ct. 1886 (internal quotation marks omitted). Otherwise, our democratic system would fail, not through the inability of the people to speak or to transmit their views to government, but because of an elected government's inability to translate those views into action.

Thus, once again, it is not surprising that this Court has applied less strict standards when reviewing speech restrictions embodied in government regulatory programs. This Court, for example, has applied a "rational basis" standard for reviewing those restrictions when they have only indirect impacts on speech. See Glickman v. Wileman Brothers & Elliott, Inc. , 521 U.S. 457 , 469-470, 477, 117 S.Ct. 2130 , 138 L.Ed.2d 585 (1997). And it has applied a mid-level standard of review-often termed "intermediate scrutiny"-when the government directly restricts protected commercial speech. See Central Hudson Gas & Elec. Corp. v. Public Serv. Comm'n of N. Y. , 447 U.S. 557 , 561-564, 100 S.Ct. 2343 , 65 L.Ed.2d 341 (1980).

This account of well-established principles at the core of the First Amendment demonstrates the problem with the plurality's approach. To reflexively treat all content-based distinctions as subject to strict scrutiny regardless of context or practical effect is to engage in an analysis untethered from the First Amendment's objectives. And in this case, strict scrutiny is inappropriate. Recall that the exception at issue here concerns debt collection-specifically a method for collecting government-owned or -backed debt. Regulation of debt collection does not fall on the first side of the democratic equation. It has next to nothing to do with the free marketplace of ideas or the transmission of the people's thoughts and will to the government. It has everything to do with the second side of the equation, that is, with government response to the public will through ordinary commercial regulation. To apply the strictest level of scrutiny to the economically based exemption here is thus remarkable.

I recognize that the underlying cell phone robocall restriction primarily concerns a means of communication. And that *2360 fact, as I discuss below, triggers some heightened scrutiny, reflected in an intermediate scrutiny standard. Strict scrutiny and its strong presumption of unconstitutionality, however, have no place here.

The plurality claims that its approach, which categorically applies strict scrutiny to content-based distinctions, will not "affect traditional or ordinary economic regulation of commercial activity." Ante, at 2347. But how is that so? Much of human life involves activity that takes place through speech. And much regulatory activity turns upon speech content. See, e.g. , Reed , 576 U.S., at 177-178 , 135 S.Ct. 2218 (BREYER, J., concurring in judgment) (giving examples). Consider, for example, the regulation of securities sales, drug labeling, food labeling, false advertising, workplace safety warnings, automobile airbag instructions, consumer electronic labels, tax forms, debt collection, and so on. All of those regulations necessarily involve content-based speech distinctions. What are the differences between regulatory programs themselves other than differences based on content? After all, the regulatory spheres in which the Securities and Exchange Commission or the Federal Trade Commission operate are defined by content. Put simply, treating all content-based distinctions on speech as presumptively unconstitutional is unworkable and would obstruct the ordinary workings of democratic governance.

That conclusion is true here notwithstanding the plurality's effort to bring political speech into the First Amendment analysis. See ante, at 2346 - 2347, 2356 (characterizing Congress as having "favored debt-collection speech over plaintiffs' political speech"). It is true that the underlying cell phone robocall restriction generally prohibits political speakers from making robocalls. But that has little to do with the government-debt exception or its practical effect. Nor does it justify the application of strict scrutiny.

Consider prescription drug labels, securities forms, and tax statements. A government agency might reasonably specify just what information the form or label must contain and further provide that the form or label may not contain other information (thereby excluding political statements). No one would think that the exclusion of political speech, say, from a drug label, means that courts must examine all other regulatory exceptions with strict scrutiny. Put differently, it is hard to imagine that such exceptions threaten political speech in the marketplace of ideas, or have any significant impact on the free exchange of ideas. To treat those exceptions as presumptively unconstitutional would work a significant transfer of authority from legislatures and agencies to courts, potentially inhibiting the creation of the very government programs for which the people (after debate) have voiced their support, despite those programs' minimal speech-related harms. See Sorrell , 564 U.S., at 584-585 , 131 S.Ct. 2653 (BREYER, J., dissenting). Given the values at the heart of the First Amendment, see supra, at 2358 - 2360, that interpretation threatens to stand that Amendment on its head. It could also lead the Court to water down the strict scrutiny standard, which would limit speech protections in situations where strict scrutiny's strong protections should properly apply. Reed , 576 U.S., at 178 , 135 S.Ct. 2218 (BREYER, J., concurring in judgment).

If, as I have argued, the First Amendment does not support the mechanical conclusion that content discrimination automatically triggers strict scrutiny, what role might content discrimination play? The plurality is correct when it quotes this Court as having said that the government may not discriminate " 'in the regulation of expression on the basis of the content of *2361 that expression.' " Ante, at 2346 (quoting Hudgens v. NLRB , 424 U.S. 507 , 520, 96 S.Ct. 1029 , 47 L.Ed.2d 196 (1976) ). If, however, this Court is to apply the First Amendment consistently with the democratic values embodied within that Amendment, that kind of statement must reflect a rule of thumb applicable only in certain circumstances. See Reed , 576 U.S., at 176 , 135 S.Ct. 2218 (BREYER, J., concurring in judgment); id., at 183, 135 S.Ct. 2218 (KAGAN, J., concurring in judgment) ("We can administer our content-regulation doctrine with a dose of common sense, so as to leave standing laws that in no way implicate its intended function").

Indeed, that must be so given that this Court's First Amendment jurisprudence itself ties the constitutional protection speech receives to the content or purpose of that speech. The Court has held that entire categories of speech-for example, obscenity, fraud, and speech integral to criminal conduct-are generally unprotected by the First Amendment entirely because of their content. See Miller v. California , 413 U.S. 15 , 23, 93 S.Ct. 2607 , 37 L.Ed.2d 419 (1973) (obscenity); Virginia Bd. of Pharmacy v. Virginia Citizens Consumer Council, Inc. , 425 U.S. 748 , 771, 96 S.Ct. 1817 , 48 L.Ed.2d 346 (1976) (fraud); Giboney v. Empire Storage & Ice Co. , 336 U.S. 490 , 498, 69 S.Ct. 684 , 93 L.Ed. 834 (1949) (speech integral to criminal conduct). As Justice Stevens pointed out, "our entire First Amendment jurisprudence creates a regime based on the content of speech." R. A. V. v. St. Paul , 505 U.S. 377 , 420, 112 S.Ct. 2538 , 120 L.Ed.2d 305 (1992) (opinion concurring in judgment); see id., at 420-422 , 112 S.Ct. 2538 (providing examples). Given that this Court looks to the nature and content of speech to determine whether, or to what extent, the First Amendment protects it, it makes little sense to treat every content-based distinction Congress has made as presumptively unconstitutional.

Moreover, it is no answer to claim that this Court's precedents categorically require such an analysis. See ante, at 2347 - 2348, n. 5 (plurality opinion). Our First Amendment jurisprudence has always been contextual and has defied straightforward reduction to unyielding categorical rules. The idea that broad language in any one case (even Reed ) has categorically determined how content discrimination should be applied in every single context is both wrong and reflects an oversimplification and over-reading of our precedent. The diversity of approaches in this very case underscores the point that the law here is far from settled. Indeed, the plurality itself disclaims the idea that its rule would apply to unsettle "traditional or ordinary economic regulation of commercial activity," indicating that the plurality presumably thinks there are some outer bounds to its broad language. Ante, at 2347. The question here is whether the Court's general statements about content discrimination triggering strict scrutiny, including in Reed , make sense as applied in this context. As I have explained, they do not.

That said, I am not arguing for the abolition of the concept of "content discrimination." There are times when using content discrimination to trigger scrutiny is eminently reasonable. Specifically, when content-based distinctions are used as a method for suppressing particular viewpoints or threatening the neutrality of a traditional public forum, content discrimination triggering strict scrutiny is generally appropriate. See Reed , 576 U.S., at 176 , 135 S.Ct. 2218 (BREYER, J., concurring in judgment); id., at 182-183, 135 S.Ct. 2218 (KAGAN, J., concurring in judgment).

Neither of those situations is present here. Outside of these circumstances, content *2362 discrimination can at times help determine the strength of a government justification or identify a potential interference with the free marketplace of ideas. See id., at 176-177, 135 S.Ct. 2218 (BREYER, J., concurring in judgment). But, as I have explained, this case is not about protecting the marketplace of ideas. It is not about the formation of public opinion or the transmission of the people's will to elected representatives. It is fundamentally about a method of regulating debt collection.

III

I would examine the validity of the regulation at issue here using a First Amendment standard that (unlike strict scrutiny) does not strongly presume that a regulation that affects speech is unconstitutional. However, given that the government-debt exception does directly impact a means of communication, the appropriate standard requires a closer look at the restriction than does a traditional "rational basis" test. A proper inquiry should examine the seriousness of the speech-related harm, the importance of countervailing objectives, the likelihood that the restriction will achieve those objectives, and whether there are other, less restrictive ways of doing so. Narrow tailoring in this context, however, does not necessarily require the use of the least-restrictive means of furthering those objectives. Cf. Ward v. Rock Against Racism , 491 U.S. 781 , 797-799, and n. 6, 109 S.Ct. 2746 , 105 L.Ed.2d 661 (1989) (explaining that outside of strict scrutiny review, narrow tailoring does not require the use of least-restrictive-means analysis). That inquiry ultimately evaluates a restriction's speech-related harms in light of its justifications. We have typically called this approach "intermediate scrutiny," though we have sometimes referred to it as an assessment of "fit," sometimes called it "proportionality," and sometimes just applied it without using a label. See United States v. Alvarez , 567 U.S. 709 , 730-731, 132 S.Ct. 2537 , 183 L.Ed.2d 574 (2012) (BREYER, J., concurring in judgment); Reed , 576 U.S., at 179 , 135 S.Ct. 2218 (BREYER, J., concurring in judgment).

Applying this Court's intermediate scrutiny analysis, I would begin by asking just what the First Amendment harm is here. As Justice KAVANAUGH notes, the government-debt exception provides no basis for undermining the general cell phone robocall restriction. Ante, at 2348 - 2349. Indeed, looking at the government-debt exception in context, we can see that the practical effect of the exception, taken together with the rest of the statute, is to put non -government debt collectors at a disadvantage. Their speech operates in the same sphere as government-debt collection speech, communicates comparable messages, and yet does not have the benefit of a particular instrument of communication (robocalls). While this is a speech-related harm, debt-collection speech is both commercial and highly regulated. See Brief for Petitioners 20-21 (describing multiple restrictions imposed by the Fair Debt Collection Practices Act on communications by debt collectors in the course of debt collection). The speech-related harm at issue here-and any related effect on the marketplace of ideas-is modest.

What, then, is the justification for this harm? The purpose of the exception is to further the protection of the public fisc. See supra, at 2357 - 2358. That protection is an important governmental interest. Private debt typically involves private funds; public debt typically involves funds that, in principle, belong to all of us, and help to implement numerous governmental policies that the people support.

Finally, is the exception narrowly tailored? Its limited scope shows that it is.

*2363 Congress has minimized any speech-related harm by tying the exception directly to the Government's interest in preserving the public fisc. The statutory text makes clear that calls will only fall within the bounds of that exception if they are "made solely to collect" Government debt. 47 U.S.C. § 227 (b)(1)(A)(iii) (emphasis added). Thus, the exception cannot be used to permit communications unrelated or less directly related to that public fiscal interest.

The upshot is that the government-debt exception, taken in context, inflicts some speech-related harm. But the harm, as I have explained, is related not to public efforts to develop ideas or transmit them to the Government, but to the Government's response to those efforts, which here takes the form of highly regulated commercial communications. Moreover, there is an important justification for that harm, and the exception is narrowly tailored to further that goal. Given those facts, the government-debt exception should survive intermediate First Amendment scrutiny.

IV

For the reasons described above, I would find that the government-debt exception does not violate the First Amendment. A majority of the Court, however, has concluded the contrary. It must thus decide whether that provision is severable from the rest of the statute. As to that question, I agree with Justice KAVANAUGH's conclusion that the provision is severable. Accordingly, I respectfully concur in the judgment with respect to severability and dissent in part.

Justice GORSUCH, with whom Justice THOMAS joins as to Part II, concurring in the judgment in part and dissenting in part.

I agree with Justice KAVANAUGH that the provision of the Telephone Consumer Protection Act before us violates the First Amendment. Respectfully, however, I disagree about why that is so and what remedial consequences should follow.

The TCPA is full of regulations on robocalls. The statute limits robocalls to residential landlines, hospitals, emergency numbers, and business lines. The only provision before us today, however, concerns robocalls to cell phones, mobile devices, or "any service for which the called party is charged for the call." 47 U.S.C. § 227 (b)(1)(A)(iii). Before the law's enactment, many cell phone users had to pay for each call, so they suffered not only the pleasure of robocalls, but also the privilege of paying for them. In 1991, Congress sought to address the problem by banning nearly all unsolicited robocalls to cell phones.

But much has changed since then. Now, cell phone users often pay a flat monthly fee for unlimited minutes, reducing the cost (if not the annoyance) of hearing from robocallers. New weapons in the fight against robocallers have emerged, too-including tools that allow consumers to more easily screen and block unwanted calls. Perhaps in recognition of these changes, Congress relaxed the ban on cellphone robocallers in 2015. Today, unsolicited calls are permitted if they are "made solely to collect a debt owed to or guaranteed by the United States."

That leaves robocallers no shortage of material. The government backs millions upon millions of loans-student loans, home mortgages, veterans' loans, farm loans, business loans. When it comes to student loans alone, the government guarantees more than $150 billion in private loans involving over 7 million individuals.

*2364 And, to be clear, it's not just the government that's allowed to call about these loans. Private lenders and debt collectors are free to send in the robots too, so long as the debt at issue is ultimately guaranteed by the government.

Today's plaintiffs wish to use robocalls for something different: to campaign and solicit donations for political causes. The plaintiffs allege that the law's continuing ban on calls like theirs violates the First Amendment, and on the main points of their argument the parties agree. First, no one doubts the TCPA regulates speech. Second, everyone accepts that restrictions on speech-no matter how evenhanded-must be justified by at least a " 'significant governmental interest.' " Ward v. Rock Against Racism , 491 U.S. 781 , 791, 109 S.Ct. 2746 , 105 L.Ed.2d 661 (1989). And, third, the parties agree that laws that go further by regulating speech on the basis of content invite still greater scrutiny. When the government seeks to censor speech based on its content, favoring certain voices and punishing others, its restrictions must satisfy "strict scrutiny"-meaning they must be justified by interests that are "compelling," not just significant. After all, a constitutional right would hardly be needed to protect popular speakers; the First Amendment does its real work in giving voice to those a majority would silence. See McCullen v. Coakley , 573 U.S. 464 , 477-478, 134 S.Ct. 2518 , 189 L.Ed.2d 502 (2014) ; but see ante, at 2359 (BREYER, J., concurring in judgment with respect to severability and dissenting in part) (seeking to overturn precedent and allow the government sometimes to impose content-based restrictions to "respon[d] to the public will").

In my view, the TCPA's rule against cellphone robocalls is a content-based restriction that fails strict scrutiny. The statute is content-based because it allows speech on a subject the government favors (collecting its debts) while banning speech on other disfavored subjects (including political matters). Cf. ante, at 2361 - 2363 (opinion of BREYER, J.) (mistakenly characterizing the content discrimination as "not about" political activities). The statute fails strict scrutiny because the government offers no compelling justification for its prohibition against the plaintiffs' political speech. In fact, the government does not dispute that, if strict scrutiny applies, its law must fall.

It's easy enough to see why the government makes no effort to satisfy strict scrutiny. Now that most cell phone plans do not charge by the call, the only justification the government cites for its robocall ban is its interest in protecting consumer privacy. No one questions that protecting consumer privacy qualifies as a legitimate and "genuine" interest for the government to pursue. Ante, at 2357 - 2358, 2362. But before the government may censor the plaintiffs' speech based on its content, it must point to a compelling interest. And if the government thinks consumer privacy interests are insufficient to overcome its interest in collecting debts, it's hard to see how the government might invoke consumer privacy interests to justify banning private political speech. Especially when consumers seem to find debt collection efforts particularly intrusive: Year after year, the Federal Trade Commission receives more complaints about the debt collection industry than any other. The nature and breadth of the law's exception calls into question the necessity of its rule.

Much precedent supports this course. As this Court has long explained, a law's failure to address a wide swath of conduct implicating its supposed concern "diminish[es] the credibility of the government's [stated] rationale for [its] restrict[ion]."

*2365 City of Ladue v. Gilleo , 512 U.S. 43 , 52, 114 S.Ct. 2038 , 129 L.Ed.2d 36 (1994). Or, as the Court has elsewhere put it, the compellingness of the government's putative interest is undermined when its law "leaves appreciable damage to [the] supposedly vital interest unprohibited." Church of Lukumi Babalu Aye, Inc. v. Hialeah , 508 U.S. 520 , 547, 113 S.Ct. 2217 , 124 L.Ed.2d 472 (1993) (internal quotation marks omitted); see also Gonzales v. O Centro Espírita Beneficente União do Vegetal , 546 U.S. 418 , 433, 126 S.Ct. 1211 , 163 L.Ed.2d 1017 (2006). The insight is simple: A law's failure to cover "significant tracts of conduct implicating [its] putatively compelling interes[t] can raise ... the inference that the ... claimed interest isn't ... so compelling after all." Yellowbear v. Lampert , 741 F.3d 48 , 60 (C.A.10 2014).

That's not to say the inference is irrebuttable. The government might, for example, show that the apparent inconsistency in its law is justified by some qualitative or quantitative difference between the speech it favors and the speech it disfavors. See id., at 61 . So if debt collection robocalls were less invasive of consumer privacy than other kinds of robocalls, or if they were inherently rare, an exception permitting debt collection calls might not undermine the government's claimed interest in banning other calls. But the government, a party with every incentive and ample resources, has not even tried to suggest conditions like those are present here, and understandably so: The government-debt exception allows a seemingly infinite number of robocalls of the type consumers appear to find most invasive.

With a First Amendment violation proven, the question turns to remedy. Because the challenged robocall ban unconstitutionally infringes on their speech, I would hold that the plaintiffs are entitled to an injunction preventing its enforcement against them. This is the traditional remedy for proven violations of legal rights likely to work irreparable injury in the future. Preventing the law's enforcement against the plaintiffs would fully address their injury. And going this far, but no further, would avoid "short circuit[ing] the democratic process" by interfering with the work of Congress any more than necessary. Washington State Grange v. Washington State Republican Party , 552 U.S. 442 , 451, 128 S.Ct. 1184 , 170 L.Ed.2d 151 (2008).

Justice KAVANAUGH's opinion pursues a different course. Invoking "severability doctrine," it declares the government-debt exception void and severs it from the statute. As revised by today's decision, the law prohibits nearly all robocalls to cell phones, just as it did back in 1991. In support of this remedy, we are asked to consider cases involving equal protection violations, where courts have sometimes solved the problem of unequal treatment by leveling others "down" to the plaintiff 's status rather than by leveling the plaintiff "up" to the status others enjoy.

I am doubtful of our authority to rewrite the law in this way. Many have questioned the propriety of modern severability doctrine, * and today's case illustrates some of the reasons why. To start, it's hard to see how today's use of severability doctrine qualifies as a remedy at all: The plaintiffs *2366 have not challenged the government-debt exception, they have not sought to have it severed and stricken, and far from placing "unequal treatment" at the "heart of their suit," they have never complained of unequal treatment as such. Ante, at 2355 - 2356. The plaintiffs point to the government-debt exception only to show that the government lacks a compelling interest in restricting their speech. It isn't even clear the plaintiffs would have standing to challenge the government-debt exception. They came to court asserting a right to speak, not a right to be free from other speakers. Severing and voiding the government-debt exception does nothing to address the injury they claim; after today's ruling, federal law bars the plaintiffs from using robocalls to promote political causes just as stoutly as it did before. What is the point of fighting this long battle, through many years and all the way to the Supreme Court, if the prize for winning is no relief at all?

A severance remedy not only fails to help the plaintiffs, it harms strangers to this suit. Just five years ago, Congress expressly authorized robocalls to cell phones to collect government-backed debts. Yet, today, the Court reverses that decision and outlaws the entire industry. It is highly unusual for judges to render unlawful conduct that Congress has explicitly made lawful-let alone to take such an extraordinary step without warning to those who have ordered their lives and livelihoods in reliance on the law, and without affording those individuals any opportunity to be heard. This assertion of power strikes me as raising serious separation of powers questions, and it marks no small departure from our usual reliance on the adversarial process.

Nor does the analogy to equal protection doctrine solve the problem. That doctrine promises equality of treatment, whatever that treatment may be. The First Amendment isn't so neutral. It pushes, always, in one direction: against governmental restrictions on speech. Yet, somehow, in the name of vindicating the First Amendment, our remedial course today leads to the unlikely result that not a single person will be allowed to speak more freely and, instead, more speech will be banned.

In an effort to mitigate at least some of these problems, Justice KAVANAUGH suggests that the ban on government-debt collection calls announced today might be applied only prospectively. See ante , at 2355, n. 13. But prospective decisionmaking has never been easy to square with the judicial power. See, e.g. , James B. Beam Distilling Co. v. Georgia , 501 U.S. 529 , 548-549, 111 S.Ct. 2439 , 115 L.Ed.2d 481 (1991) (Scalia, J., concurring in judgment) (judicial power is limited to "discerning what the law is , rather than decreeing ... what it will tomorrow be"). And a holding that shields only government-debt collection callers from past liability under an admittedly unconstitutional law would wind up endorsing the very same kind of content discrimination we say we are seeking to eliminate.

Unable to solve the problems associated with its preferred severance remedy, today's decision seeks at least to identify "harm[s]" associated with mine. Cf. ante , at 2356 (opinion of KAVANAUGH, J.). In particular, we are reminded that granting an injunction in this case would allow the plaintiffs' (unpopular) speech, and that could induce others to seek injunctions of their own, resulting in still more (unpopular) speech. But this "harm" is hardly comparable to the problems associated with using severability doctrine: Having to tolerate unwanted speech imposes no cognizable constitutional injury on anyone; it is life under the First Amendment, which is *2367 almost always invoked to protect speech some would rather not hear.

*

In the end, I agree that 47 U.S.C. § 227 (b)(1)(A)(iii) violates the First Amendment, though not for the reasons Justice KAVANAUGH offers. Nor am I able to support the remedy the Court endorses today. Respectfully, if this is what modern "severability doctrine" has become, it seems to me all the more reason to reconsider our course.

See, e.g ., Seila Law LLC v. Consumer Financial Protection Bureau , --- U.S. at ---- - ----, 140 S.Ct. at 2218 - 2224, ante , at 14-24, 2020 WL 3492641 (THOMAS, J., concurring in part and dissenting in part); Harrison, Severability, Remedies, and Constitutional Adjudication, 83 Geo. Wash. L. Rev. 56 (2014) ; see also Movsesian, Severability in Statutes and Contracts, 30 Ga. L. Rev. 41 , 41-42 (1995) (collecting academic criticism of severability doctrine).

* Counsel of Record

6 When Congress enacts a law with a severability clause and later adds new provisions to that statute, the severability clause applies to those new provisions to the extent dictated by the text of the severability clause. Likewise, when Congress has not included a severability clause in initial legislation, Congress can subsequently enact a severability clause that applies to the existing statute to the extent dictated by the text of the later-added severability clause. In both scenarios, the text of the severability clause remains central to the severability inquiry.

7 If courts had broad license to invalidate more than just the offending provision, a reviewing court would have to consider what other provisions to invalidate: the whole section, the chapter, the statute, the public law, or something else altogether. Courts would be largely at sea in making that determination, and usually could not do it in a principled way. Here, for example, would a court invalidate all or part of the Bipartisan Budget Act of 2015 rather than all or part of the 1991 TCPA? After all, that 2015 Bipartisan Budget Act, not the 1991 TCPA, added the constitutionally problematic government-debt exception. That is the kind of free-wheeling policy question that the Court's presumption of severability avoids.

8 The term "invalidate" is a common judicial shorthand when the Court holds that a particular provision is unlawful and therefore may not be enforced against a plaintiff. To be clear, however, when it "invalidates" a law as unconstitutional, the Court of course does not formally repeal the law from the U. S. Code or the Statutes at Large. Instead, in Chief Justice Marshall's words, the Court recognizes that the Constitution is a "superior, paramount law," and that "a legislative act contrary to the constitution is not law" at all. Marbury v. Madison , 1 Cranch 137 , 177, 2 L.Ed. 60 (1803). The Court's authority on this front "amounts to little more than the negative power to disregard an unconstitutional enactment." Massachusetts v. Mellon , 262 U.S. 447 , 488, 43 S.Ct. 597 , 67 L.Ed. 1078 (1923).

Justice THOMAS's thoughtful approach to severability as outlined in Murphy v. National Collegiate Athletic Assn. , 584 U. S. ----, ---- - ----, 138 S.Ct. 1461 , 1485-1487, 200 L.Ed.2d 854 (2018), and Seila Law LLC v. Consumer Financial Protection Bureau , --- U.S. at ---- - ----, 140 S.Ct. at 2218 - 2224, ante , at 14-24, 2020 WL 3492641 , (joined by Justice GORSUCH in the latter) would simply enjoin enforcement of a law as applied to the particular plaintiffs in a case. Under either the Court's approach or Justice THOMAS's approach, an offending provision formally remains on the statute books (at least unless Congress also formally repeals it). Under either approach, the formal remedy afforded to the plaintiff is an injunction, declaration, or damages. One difference between the two approaches is this: Under the Court's approach, a provision is declared invalid and cannot be lawfully enforced against others. Under Justice THOMAS's approach, the Court's ruling that a provision cannot be enforced against the plaintiff, plus executive respect in its enforcement policies for controlling decisional law, plus vertical and horizontal stare decisis in the courts, will mean that the provision will not and cannot be lawfully enforced against others. The Court and Justice THOMAS take different analytical paths, but in many cases, the different paths lead to the same place.

9 On occasion, of course, it may be that a particular surrounding or connected provision is not operative in the absence of the unconstitutional provision, even though the rest of the law would be operative. That scenario may require severance of somewhat more than just the offending provision, albeit not of the entire law. Courts address that scenario as it arises.

10 A codifier's note explains a change in wording from the original Public Law: "This chapter, referred to in text, was in the original 'this Act', meaning act June 19, 1934, ch. 652, 48 Stat. 1064 , known as the Communications Act of 1934, which is classified principally to this chapter." Note following 47 U.S.C. § 608 .

11 The cases cited in the text above are pre- Erie decisions involving the constitutionality of state laws. See Erie R. Co. v. Tompkins , 304 U.S. 64 , 58 S.Ct. 817 , 82 L.Ed. 1188 (1938). In that era, the Court often treated severability of state laws and federal laws in the same general way. In the post- Erie era, severability of state laws can potentially pose different questions than severability of federal laws. We need not address post- Erie severability of state laws. See, e.g. , Ayotte v. Planned Parenthood of Northern New Eng. , 546 U.S. 320 , 328-331, 126 S.Ct. 961 , 163 L.Ed.2d 812 (2006) ; Leavitt v. Jane L. , 518 U.S. 137 , 139, 116 S.Ct. 2068 , 135 L.Ed.2d 443 (1996) ( per curiam ) ("Severability is of course a matter of state law").

12 As the Government acknowledges, although our decision means the end of the government-debt exception, no one should be penalized or held liable for making robocalls to collect government debt after the effective date of the 2015 government-debt exception and before the entry of final judgment by the District Court on remand in this case, or such date that the lower courts determine is appropriate. See Reply Brief 24. On the other side of the ledger, our decision today does not negate the liability of parties who made robocalls covered by the robocall restriction.

13 Plaintiffs suggest that parties will not have incentive to sue if the cure for challenging an unconstitutional exception to a speech restriction is to eliminate the exception and extend the restriction. But many individuals and organizations often have incentive to challenge unequal treatment of speech, especially when a competitor is regulated less heavily.