PRECEDENTIAL
UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT _______________
No. 24-1097 _______________
VETERANS GUARDIAN VA CLAIM CONSULTING LLC; JOHN F. RUDMAN; ANDRE JESUS SOTO, Appellants
v.
MATTHEW J. PLATKIN, in his official capacity as Attorney General of New Jersey _______________
On Appeal from the United States District Court for the District of New Jersey (D.C. No. 3:23-cv-20660) District Judge: Honorable Michael A. Shipp _______________
Argued: November 8, 2024
Before: KRAUSE, BIBAS, and SCIRICA, Circuit Judges
(Filed: April 1, 2025) Martine E. Cicconi [ARGUED] Kristen Loveland James E. Tysse Caroline L. Wolverton AKIN GUMP STRAUSS HAUER & FELD 2001 K Street NW Washington, DC 20006 Counsel for Appellants
Joshua P. Bohn Michael L. Zuckerman [ARGUED] NEW JERSEY ATTORNEY GENERAL’S OFFICE 25 Market Street Richard J. Hughes Justice Complex Trenton, NJ 08625
Meghan Musso Andrew H. Yang NEW JERSEY ATTORNEY GENERAL’S OFFICE 124 Halsey Street Newark, NJ 07101 Counsel for Appellee
Christian W. Lansinger Paul M. Sherman INSTITUTE FOR JUSTICE 901 N Glebe Road Suite 900 Arlington, VA 22203 Counsel for Amicus Institute for Justice in Support of Appellants
2 Caleb R. Trotter PACIFIC LEGAL FOUNDATION 555 Capitol Mall Suite 1290 Sacramento, CA 95814 Counsel for Amicus Pacific Legal Foundation in Support of Appellants
Glenn R. Bergmann Thomas Polseno Daniel D. Wedemeyer BERGMANN & MOORE 25 W Middle Lane Rockville, MD 20850 Counsel for Amici Veterans of Foreign Wars of the United States, American Legion, Paralyzed Veterans of America, & Vietnam Veterans of America in Support of Appellee
Brian Hunt Rowe WASHINGTON ATTORNEY GENERAL’S OFFICE 800 Fifth Avenue Suite 2000 Seattle, WA 98104
Cristina Sepe WASHINGTON ATTORNEY GENERAL’S OFFICE 1125 Washington Street SE P.O. Box 40100 Olympia, WA 98504
3 Counsel for Amici States of Washington, Colorado, Con- necticut, Delaware, Illinois, Maine, Maryland, Michigan, Minnesota, Nevada, New Mexico, Oregon, Rhode Island, Commonwealth of Pennsylvania, and District of Colum- bia in Support of Appellee _______________
OPINION OF THE COURT _______________
BIBAS, Circuit Judge. The marketplace of ideas is not just a metaphor. Many Americans, from journalists to playwrights to therapists, speak for a living. Laws that bar these professionals from earning money on that speech limit their ability to speak and so must survive First Amendment scrutiny. New Jersey recently passed one such law, banning charging for some advice on how to claim veterans benefits. Because this law likely burdens speech, yet the District Court thought otherwise and so denied a preliminary injunction, we will vacate and remand. I. NEW JERSEY DOUBLY RESTRICTS CHARGING FOR ADVICE ABOUT CLAIMING VETERANS BENEFITS Veterans who are left disabled by their service qualify for benefits. But first, they must claim them. To do that, they send a form to the Department of Veterans Affairs (VA) listing their disabilities and explaining how their service caused or aggra- vated them. The VA reviews the information and decides what benefits, if any, each veteran can get. For some, the story ends there: They get the benefits they earned and enjoy a grateful
4 nation’s compensation. Others get less happy news: The VA has awarded them less than they think they deserve. They can appeal by filing a notice of disagreement, triggering review by the Board of Veterans’ Appeals. 38 U.S.C. § 7105(a), (b)(3). Because this benefits bureaucracy can be frustrating, many veterans seek help submitting claims. And because the VA pays out billions per year, a booming industry of consultants has sprung up to help veterans claim benefits in exchange for a cut of the payout. As the benefits-consulting industry has grown, so have worries about veterans getting ripped off. To prevent that abuse, an elaborate federal code regulates the people whom veterans can hire to help them get benefits. Two rules are relevant here. First, before anyone “may act as an agent or attorney in the preparation, presentation, or prose- cution of any claim” for veterans benefits, he must be accred- ited by the VA. 38 U.S.C. § 5901(a); 38 C.F.R. § 14.629(b)(1); accord 38 C.F.R. § 14.636(b). Second, these agents and attor- neys may not charge for any services that they provide before the VA’s initial benefits decision. 38 U.S.C. § 5904(c)(1); 38 C.F.R. § 14.636(c). Though federal law sets these rules, it does not empower the VA or anyone else to enforce them. New Jersey took that job upon itself. It passed a law that it says mirrors federal law but adds civil-enforcement mecha- nisms. Under Section (a)(1) of New Jersey’s law, “[n]o person shall receive compensation for advising or assisting” anyone with “the preparation, presentation, or prosecution of any claim” for veterans benefits, except as allowed by federal law. N.J. Stat. Ann. § 56:8-228(a)(1), (d). And under Section (a)(4), “[n]o person shall receive any compensation for any services
5 rendered before” a veteran appeals the VA’s initial benefits decision to the Board of Veterans’ Appeals. § 56:8-228(a)(4). New Jersey’s law is a problem for Veterans Guardian, a nationwide consulting company that charges veterans for advice on how to claim benefits. Fearing that its business model vio- lated the law, it closed its doors in the state. The law is also a problem for John Rudman and Andre Soto, two New Jersey veterans who had planned to use Veterans Guardian’s services. Veterans Guardian, Rudman, and Soto sued New Jersey’s Attorney General, claiming that the law violates their First Amendment rights and seeking a preliminary injunction. The District Court declined to grant the injunction, and now they appeal. On appeal from the denial of a preliminary injunction, we normally “review the District Court’s factual findings for clear error, its legal rulings de novo, and its ultimate decision for abuse of discretion.” Del. State Sportsmen’s Ass’n v. Del. Dep’t of Safety & Homeland Sec., 108 F.4th 194, 198 (3d Cir. 2024). But in First Amendment cases, we must examine the whole record independently, drawing our own inferences and deferring to factual findings only if they are about witness cred- ibility. Child Evangelism Fellowship of N.J. v. Stafford Twp. Sch. Dist., 386 F.3d 514, 524 (3d Cir. 2004) (Alito, J.); Tenafly Eruv Ass’n v. Borough of Tenafly, 309 F.3d 144, 156–57 (3d Cir. 2002).
6 II. PRELIMINARY INJUNCTIONS ARE EXTRAORDINARY REMEDIES “A preliminary injunction is an extraordinary remedy granted in limited circumstances.” Issa v. Sch. Dist., 847 F.3d 121, 131 (3d Cir. 2017). Whether a plaintiff can get one depends on whether (1) he is likely to succeed on the merits, (2) he will suffer irreparable harm without preliminary relief, (3) the bal- ance of equities favors an injunction, and (4) an injunction is in the public interest. Winter v. NRDC, 555 U.S. 7, 20 (2008). The first two factors are “gateway factors”: A plaintiff must satisfy them both to be eligible for preliminary relief. Reilly v. City of Harrisburg, 858 F.3d 173, 179 (3d Cir. 2017). But they are not equally demanding. On the first factor, a likelihood of success on the merits means only a “reasonable probability” of success—odds that are “significantly better than negligible but not necessarily more likely than not.” Id. at 176, 179. And though plaintiffs usually bear the burden of showing that they are likely to suc- ceed, that burden flips in First Amendment cases. Greater Phila. Chamber of Com. v. City of Philadelphia, 949 F.3d 116, 133 (3d Cir. 2020). Because “[t]he government bears the bur- den of proving that” a law complies with the First Amendment, it also bears the burden of proving that a plaintiff is unlikely to succeed on the merits. Id. The standard for irreparable harm is more demanding: The plaintiff must show that it is more likely than not. Reilly, 858 F.3d at 179. That said, a violation of First Amendment rights presumptively inflicts irreparable harm. Del. State, 108 F.4th at 204.
7 If both gateway factors are satisfied, the court must still weigh all four factors before granting preliminary relief. Id. at 202–03; see also Winter, 555 U.S. at 26, 31–33 (denying pre- liminary injunction solely on the balance of equities and public interest). It should also consider whether the injunction is needed to keep the case alive until the court can award final relief. Del. State, 108 F.4th at 200–03. ***** Courts typically start with the merits. The District Court also stopped there, deciding that Veterans Guardian was unlikely to succeed and so the court did not need to reach the remaining preliminary-injunction factors. That holding rested on several mistakes, as we go on to explain. First, the District Court held that New Jersey’s law did not even trigger the First Amend- ment. We disagree: Veterans Guardian is likely engaged in speech, which New Jersey’s law burdens. Second, the District Court analyzed New Jersey’s law as a monolithic whole, when the law imposes two separate burdens on speech that must be analyzed separately. But though we correct these oversights, the record is not developed enough for us to decide the serious constitutional questions that the merits raise, let alone to weigh the other preliminary-injunction factors. So we will remand for the District Court to reconsider Veterans Guardian’s motion with the benefit of a fuller record and our guidance. III. NEW JERSEY’S LAW LIKELY BURDENS SPEECH, SO IT MUST SATISFY THE FIRST AMENDMENT
Veterans Guardian has a reasonable probability of show- ing that its services are speech and that New Jersey’s law
8 burdens that speech. So the state must show that its law satis- fies the First Amendment. A. Veterans Guardian is likely to succeed in showing that its services are speech At this early stage, Veterans Guardian need show only a reasonable probability of success on the merits, including show- ing that its services are speech. On the current record, they are. Professional services delivered by speaking or writing are speech. King v. Governor of N.J., 767 F.3d 216, 224–25 (3d Cir. 2014), abrogated on other grounds by Nat’l Inst. of Fam. & Life Advocs. v. Becerra (NIFLA), 585 U.S. 755 (2018). Vet- erans Guardian delivers professional services by speaking and writing. It advises clients about how to claim benefits: what disabilities to claim, what evidence to include, and how to fill out forms. That advice is likely speech. See Upsolve, Inc. v. James, 604 F. Supp. 3d 97, 112 (S.D.N.Y. 2022) (distinguishing conduct of filing motions from speech of offering legal advice). B. Laws that forbid charging for speech burden speech If indeed Veterans Guardian’s services are speech, then New Jersey’s law burdens it. The District Court thought other- wise because New Jersey does not bar giving advice but only charging for it. So, the court reasoned, the law does not burden the plaintiffs’ First Amendment rights because it regulates not speech but conduct. Yet laws that ban charging for speech burden the right to speak. Supreme Court cases establish this. See United States v. Nat’l Treasury Emps. Union, 513 U.S. 454, 468–70 (1995); Simon & Schuster, Inc. v. Members of N.Y. State Crime Victims
9 Bd., 502 U.S. 105, 115–16 (1991); see also Riley v. Nat’l Fed’n of the Blind of N.C., 487 U.S. 781, 801 (1988) (“[A] speaker is no less a speaker because he or she is paid to speak.”). Com- mon sense agrees. Someone who cannot earn money from speaking has less incentive to speak and so will speak less. Nat’l Treasury Emps. Union, 513 U.S. at 468–70. Indeed, many canonical examples of protected speech involve profes- sionals speaking for pay: think of novelists, speechwriters, and newspaper columnists. A law that allowed their speech but banned their livelihoods would reduce how much they could speak, which in turn would dam up the flow of ideas in our democracy. By barring payment for speech, New Jersey’s law burdens the right to speak. The District Court thought otherwise, misreading a passing comment in Expressions Hair Design v. Schneiderman, 581 U.S. 37 (2017). In that case, the Supreme Court observed that a hypothetical law setting sandwich prices at $10 would target conduct, even though it would also require delis to write “$10” on their menus and quote that price to inquiring customers. Id. at 47. Because “the law’s effect on speech would be only inci- dental to its primary effect on conduct,” the Court noted, it would not pose a First Amendment problem. Id. The District Court thought that Expressions Hair Design meant that price restrictions cannot implicate the First Amend- ment. But a law regulating the price of sandwiches burdens the conduct of selling sandwiches. By contrast, a law regulating the price of speech burdens speech.
10 IV. THERE IS NO SEPARATE CATEGORY OF PROFESSIONAL SPEECH
Having decided that the First Amendment applies, we next ask how strong its protections are. Our circuit used to carve out a separate category of professional speech and give it less pro- tection. King, 767 F.3d at 232. But seven years ago, in NIFLA v. Becerra, the Supreme Court clarified that there is no separate category of professional speech. 585 U.S. at 767–68. With few exceptions, the same First Amendment principles apply when professionals speak to clients as when anyone else talks. Id. That said, NIFLA confirms that lesser scrutiny is warranted where there is “persuasive evidence of a long (if heretofore un- recognized) tradition to that effect.” Id. at 767 (cleaned up). And it preserved two exceptions when regulations of profes- sional speech get reviewed more deferentially: (1) “laws that require professionals to disclose factual, noncontroversial in- formation in their commercial speech” and (2) regulations of “professional conduct, even though that conduct incidentally involves speech.” Id. at 768 (internal quotation marks omitted). New Jersey tries to jam its law into NIFLA’s second excep- tion by arguing that the law regulates price instead of speech, that it regulates speech incidental to illegal conduct, and that Section (a)(1) is a neutral professional licensing scheme. Id. at 768–70; United States v. Hansen, 599 U.S. 762, 783 (2023); Appellee’s Br. 20–28. The first two arguments do not work, and the record is too sparse for us to confidently decide the third. Start with New Jersey’s contention that its law targets only the conduct of charging money. As we discussed above,
11 pricing regulations are not exempt from the First Amend- ment—restricting compensation to licensed counselors still imposes a financial burden on speech. Second, Veterans Guardian’s speech is not integral to ille- gal conduct. Because New Jersey reads its law as mirroring federal requirements, it argues that any speech it bans must be integral to breaking federal law. That argument is wrong twice over. For one, Veterans Guardian’s speech is not just one step in service of some separately illegal act, unlike the speech involved in soliciting a crime, demanding ransom, or posting a “White applicants only” sign as part of hiring discrimination. See Hansen, 599 U.S. at 783; Rumsfeld v. FAIR, 547 U.S. 47, 62 (2006). Veterans Guardian’s speech is the core of what it does. For another, though New Jersey says federal law outlaws Veteran Guardian’s activities, that federal law is equally sub- ject to the First Amendment. Veterans Guardian does not chal- lenge the federal scheme, and we take no position on whether it is valid. But states cannot immunize their laws from constitu- tional scrutiny by pointing to a federal scheme that may suffer the same constitutional defects. To hold otherwise would let states end-run around the First Amendment. Finally, New Jersey attempts to frame Section (a)(1), which incorporates federal accreditation requirements, as a neutral licensing scheme regulating professional conduct. Appellee’s Br. 20–21. Yet we have very little information on how the fed- eral accreditation scheme works or what it covers. The District Court ruled on a different basis and did not address whether the law should be viewed as a professional licensing scheme or whether, as a licensing scheme, it would fit within NIFLA’s
12 second exception. We leave it to the District Court to consider those questions in the first instance on remand. When it does, it should reconsider one more part of its rea- soning. It held that New Jersey’s law was content neutral in part because the state did not intend to suppress disfavored ideas. But courts judge laws based on their effects on speech, not just on legislatures’ purposes or motives. United States v. O’Brien, 391 U.S. 367, 383 (1968); Reed v. Town of Gilbert, 576 U.S. 155, 165–66 (2015). For instance, they usually decide whether a law is content based—and so presumptively uncon- stitutional—by judging whether it “single[s] out any topic or subject matter for differential treatment.” City of Austin v. Reagan Nat’l Advert. of Austin, LLC, 596 U.S. 61, 71 (2022); Reed, 576 U.S. at 163–64; Camp Hill Borough Republican Ass’n v. Borough of Camp Hill, 101 F.4th 266, 269–70 (3d Cir. 2024). We take no position on whether this law is content neu- tral, leaving that to the District Court on remand. But when it does reach that question, its answer should be based on whether the law applies to speech based on its content or topic, regard- less of the legislature’s good intentions. V. COURTS MUST ANALYZE PLAINTIFFS’ CHALLENGES TO SEPARATE PROVISIONS SEPARATELY
The District Court also overlooked that New Jersey’s law restricts speech in two different ways. It considered the first way: It thought that Section (a)(1) bars charging for claims advice without VA accreditation. That limits who can charge for advice— only people accredited by the VA. But it did not consider the second way: Section (a)(4) bars accepting pay for services done before a veteran files a notice
13 of disagreement (the equivalent of a notice of appeal). N.J. Stat. Ann. § 56:8-228(a)(4). That limits when people can charge for advice—only after a veteran appeals. And it imposes that total ban on everyone, even licensed counselors. It might also limit what advice people can charge for. If some choices in the initial claim cannot be changed on appeal (for instance, what disabil- ity the veteran claimed for), then by limiting paid advice to appeals, the law would effectively bar paid advice about these aspects of a claim. Either way, Sections (a)(1) and (a)(4) impose separate limits on speech and must be analyzed separately. On remand, the District Court should distinguish these sec- tions and test the constitutionality of each. VI. SECTION (a)(4)’S CONSTITUTIONALITY IS A SERIOUS QUESTION Distinguishing Section (a)(4) is especially important because we seriously doubt that it is constitutional. True, New Jersey has a strong interest in protecting veterans from fraud and pred- atory pricing. See Heffner v. Murphy, 745 F.3d 56, 92 (3d Cir. 2014). But any law pursuing this interest must be tailored to it. New Jersey has not pointed to any evidence that a ban on pre- appeal counseling fees furthers its interest. In fact, the state never even raised its concern that pre-appeal fees are exploita- tive until oral argument before us. Either way, New Jersey’s law must satisfy some form of heightened scrutiny. Though we need not decide which tier applies, we are skeptical that at least Section (a)(4) is tailored enough to satisfy even intermediate scrutiny. Under that stand- ard, New Jersey would need to show that “alternative measures that burden substantially less speech would fail to achieve the
14 government’s interests.” Bruni v. City of Pittsburgh, 824 F.3d 353, 367 (3d Cir. 2016) (quoting McCullen v. Coakley, 573 U.S. 464, 495 (2014)). At oral argument, New Jersey asserted that Section (a)(4)’s ban protects veterans because no one needs paid help to file a claim: The first stage is simple, and other services will help claimants for free. But New Jersey could use less-restrictive alternatives; for instance, it could require paid advisers to tell veterans that they can get free help elsewhere. It has never explained why a more targeted solution would not work. Still, the current record is too thin to resolve this question. In moving for a preliminary injunction, plaintiffs challenged both Sections (a)(1) and (a)(4) without analyzing them sepa- rately. Then the District Court ignored (a)(4). So that section has never gotten adequate briefing or produced a well-developed record. We will leave that task to the District Court on remand. VII. THE DISTRICT COURT SHOULD FIND MORE FACTS The District Court will need to gauge New Jersey’s interests and the law’s tailoring, and it may need to weigh preliminary- injunction factors other than the merits. Each inquiry is riddled with unknowns. Start with New Jersey’s interests. They depend on how often paid services covered by the law are predatory, how often they are merely useless, and how often they are valuable. The record contains no answers. Nor does it show how big a problem paid consultants are. Though the District Court noted that “benefits consultants and other businesses ha[ve] defrauded veterans of over $414 million,” the document it relies on lumps together all sources of fraud, from identity theft to “bogus investment
15 schemes” and “sweepstakes and lotteries.” App. 17, 217. In fact, this document lists the eleven most common sources of fraud, but predatory benefits-claiming services are not among them. The extent of tailoring is also foggy. Neither the District Court nor the parties have discussed whether less restrictive alternatives to Sections (a)(1) and (a)(4) would have achieved New Jersey’s interests. And each section raises its own ques- tions that the record does not answer. Section (a)(1) purports to fight exploitation by forcing providers to follow federal law. But how effective is federal law at stopping fraud and incom- petence? And at what cost to speech? Section (a)(4) bans paid advice before appeal. The weight of this burden depends in part on whether the appeal is too late to offer some advice. Is it? On remand, the District Court should fill these gaps. ***** Advice about claiming veterans benefits appears to be speech. But when weighing preliminary injunctions, courts must proceed cautiously and be wary of locking in their views on the merits. Del. State, 108 F.4th at 203. And though there are many non-merits factors, the District Court never analyzed them. We will remand to let it fill in the record and apply the law that we have laid out here.
16 KRAUSE, Circuit Judge, concurring.
I join the majority opinion in full but write separately with some observations about the review of reasonable professional licensing schemes in the wake of National Institute of Family and Life Advocates v. Becerra (NIFLA), 585 U.S. 755 (2018).
On the one hand, NIFLA established that professional speech, as a whole, is not a unique category subject to lesser protections than other protected speech and cautioned against giving states “unfettered power” to impose content-based restrictions on speech “by simply imposing a licensing requirement.” Id. at 773. Taken to the extreme, the Court observed, states could enact onerous licensing laws that, in effect, “impose invidious discrimination of disfavored subjects,” id. (internal quotation marks omitted), for example, by restricting publishers from printing books by certain authors or lawyers from advocating for clients outside the courtroom— all under the guise of regulating professional conduct, cf. Simon & Schuster, Inc. v. Members of N.Y. State Crime Victims Bd., 502 U.S. 105, 115 (1991) (striking down the “Son of Sam” law placing financial disincentives on convicts speaking about their crimes); NAACP v. Button, 371 U.S. 415, 433, 444 (1963) (holding a law “proscribing any arrangement by which prospective litigants are advised to seek the assistance of particular attorneys” unconstitutional).
Indeed, the Court has long cautioned that states may neither “foreclose the exercise of constitutional rights by mere labels” nor ignore those rights “under the guise of prohibiting professional misconduct.” Button, 371 U.S. at 429, 439. Thus, in NIFLA, the Court determined California’s law requiring licensed clinics to disclose the availability of services that “in
1 no way relate[d] to the services that licensed clinics provide,” 585 U.S. at 769, was a content-based regulation of speech and therefore identified no reason to shield it from strict scrutiny1—notwithstanding its commercial context, id. at 766, 773.
On the other hand, NIFLA confirmed that more deferential scrutiny continues to apply in the commercial context where there is “‘persuasive evidence . . . of a long (if heretofore unrecognized) tradition’ to that effect,” and it identified two such situations. Id. at 767–68 (quoting United States v. Alvarez, 567 U.S. 709, 722 (2012)). The first was “laws that require professionals to disclose factual, noncontroversial information in their ‘commercial speech,’” like the fee disclaimer regulation for lawyers in Zauderer v. Office of Disciplinary Counsel of Supreme Court of Ohio, 471 U.S. 626, 650–53 (1985). Id. at 768. And the second was for “regulations of professional conduct that incidentally burden speech.” Id. at 769. In support of both exceptions, the Court cited Ohralik v. Ohio State Bar Association, in which the Court held that a state can discipline a lawyer for in-person solicitation in certain circumstances, 436 U.S. 447, 449 (1978),
1 The Court did not foreclose that some reason may exist to “treat[] professional speech as a unique category that is exempt from ordinary First Amendment principles,” but it did not decide that issue because the challenged regulation failed under intermediate scrutiny. Nat’l Inst. of Fam. & Life Advocs. v. Becerra (NIFLA), 585 U.S. 755, 773 (2018). Of course, under ordinary First Amendment principles, “content-based regulations of speech are subject to strict scrutiny.” Id. at 767; see also Sorrell v. IMS Health Inc., 564 U.S. 552, 565–66 (2011).
2 reasoning the solicitation ban was more like a conduct regulation in which speech “is an essential but subordinate component,” id. at 457, and observing “that the State does not lose its power to regulate commercial activity deemed harmful to the public whenever speech is a component of that activity,” id. at 456 (listing examples of “communications that are regulated without offending the First Amendment”).
As another example of NIFLA’s second category, the Court cited Planned Parenthood of Southeastern Pennsylvania v. Casey, which “rejected a free-speech challenge to [an] informed consent requirement” for an abortion procedure, explaining that the law “regulated speech only ‘as part of the practice of medicine, subject to reasonable licensing and regulation by the State.’” NIFLA, 585 U.S. at 770 (quoting Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 884 (1992)). Even though the regulations in Ohralik and Casey were content based, in that they targeted certain subject matter in the context of regulating a particular profession,2 the NIFLA Court cited them as part of our Nation’s “long . . . tradition” of professional-conduct regulations that incidentally burden speech and nevertheless are excepted from the demands of strict scrutiny. Id. at 767–68 (quoting Alvarez, 567 U.S. at 722).
That long tradition of professional licensing schemes in our law dates back well before the Founding, with deep roots in English law. In the Middle Ages, craft guilds, chartered by the monarch, functioned as early licensing authorities, with
2 See City of Austin v. Reagan Nat’l Advert. of Austin, 596 U.S. 61, 69 (2022); Reed v. Town of Gilbert, 576 U.S. 155, 163 (2015).
3 membership a prerequisite to entering trades like weaving and goldsmithing.3 As early as 1421, doctors in England petitioned Parliament to exclude unqualified practitioners,4 and the Royal College of Physicians was established in 1518 to grant licenses and regulate the practice of medicine.5 Membership in the English Inns of Court, founded in the mid-fourteenth century, ensured the qualifications of those practicing law.6 And by the seventeenth century, licensing schemes governed the
3 See Stella Kramer, The English Craft Gilds and the Government 17 (1905); Paul J. Larkin, Jr., Public Choice Theory and Occupational Licensing, 39 Harv. J.L. & Pub. Pol’y 209, 212–13 (2016) [hereinafter Larkin]. 4 John H. Raach, English Medical Licensing in the Early Seventeenth Century, 16 Yale J. Biology & Med. 267, 268 (1944). 5 B. Abbott Goldberg, Horseshoers, Doctors and Judges and the Law on Medical Competence, 9 Pac. L.J. 107, 122 (1978). The College was established by a charter issued by King Henry VIII in 1518 and confirmed by Parliament in 1522. Parliament initially authorized the licensing of physicians by bishops in 1511, but the effect of that law was undercut by the establishment of the College. Id.; see 3 Henry VIII ch. 11, reprinted in Charles Goodall, Royal College of Physicians of London Founded and Established by Law 1 (M. Flesher ed., 1684) [hereinafter Goodall]; 14, 15 Henry VIII ch.5, reprinted in Goodall at 5. 6 Jonathan Rose, The Legal Profession in Medieval England: A History of Regulation, 48 Syracuse L. Rev. 1, 24 (1998); see also id. at 5 (“[M]edieval regulation reveals the foundation for the modern control of the admission and the conduct of practicing lawyers.”).
4 professions of tavern owner, peddler, coach driver, and many more on that side of the Atlantic.7
Credentialing practices followed the colonists to early America, where the medical and legal professions were among the first to be licensed and regulated,8 along with traders, tanners, printers, peddlers, boat pilots, tavern and innkeepers, distillers, and purveyors of liquor.9 By the early-nineteenth century, licensing schemes expanded to include barbers, boarding house operators, insurance agents, midwives, real estate brokers, steamboat operators, embalmers, horseshoers,
7 Thomas K. Urdhal, The Fee System in the United States, 77– 80 (1898) [hereinafter Urdhal]. 8 Id. at 102–03, 128–29; Barlow F. Christensen, The Unauthorized Practice of Law: Do Good Fences Really Make Good Neighbors-or Even Good Sense?, 5 Am. Bar Found. Rsch. J. 159, 163 (1980) (“The earliest legislation ‘for the better regulation of attorneys and the great fees exacted by them’ was enacted in 1642-43. It severely limited fees [and] prohibited pleading without a license from the court.”); Lawrence M. Friedman, Freedom of Contract and Occupational Licensing 1890-1910: A Legal and Social Study, 53 Cal. L. Rev. 487, 494 (1965) (“The licensing of inn-keepers and of lawyers went back to colonial times and indeed to English practice.”). By the late-nineteenth century, “more than half of the states required licenses to practice as a physician, dentist, pharmacist, or lawyer.” Larkin, supra note 3, at 213. 9 Urdhal, supra note 7, at 97 n.1, 100 & n.5, 102–05; Larkin, supra note 3, at 212–13; see also, e.g., Act of Oct. 25, 1710, reprinted in 1 Acts of Assembly, Passed in the Colony of Virginia, from 1662 to 1715, at 325 (1727) (requiring license for sale of alcohol).
5 undertakers, veterinarians, auctioneers, and pawnbrokers, among many others.10
The inevitable challenges to these regulatory regimes gave the Supreme Court opportunity to explain their place in our legal system and to acknowledge their importance. In rejecting a challenge to a medical licensing regime in 1889, for example, the Supreme Court recognized the inherent authority of the states to prescribe regulations to combat “consequences of ignorance and incapacity, as well as of deception and fraud,”
10 Larkin, supra note 3, at 213; Urdhal, supra note 7, at 129; see also, e.g., Laws of the State of New-Jersey 19–20, 235–40, 246, 355, 416 (William Patterson ed., 1800) (licensing hunters, tavern owners, innkeepers, hawkers, peddlers, petty chapmen, and fishermen); Act of Sept. 17, 1807, reprinted in Laws of the Indiana Territory 340–44 (1807) (requiring a license to practice as an “Attorney or Counsellor at Law”); 1 The General Laws of Massachusetts, From the Adoption of the Constitution to February, 1822, at 92, 100–01, 199, 297–304, 348, 417, 473–74, 535 (Theron Metcalf ed., 1823) (licensing tree cutters, estate executors and administrators, attorneys, innholders, tavern keepers, victuallers, liquor retailers, vintners, auctioneers, public good sellers, and carriage drivers); Act of Mar. 10, 1832, reprinted in 2 The General Public Statutory Law and Public Local Law of the State of Maryland 1032–34 (Clement Dorsey ed., 1840) (requiring two years of legal study and evidence of good character to be admitted as an attorney). By 1935, “architects, barbers, beauticians, chiropractors, civil engineers, embalmers, registered nurses, optometrists, osteopaths, real estate brokers and sale personnel, surveyors, and veterinarians” were licensed in more than half the states. Larkin, supra note 3, at 213 n.16.
6 and to fulfill the government’s responsibility “from time immemorial, to exact in many pursuits a certain degree of skill and learning upon which the community may confidently rely.”11 Dent v. West Virginia, 129 U.S. 114, 122 (1889). While the Court has also policed the constitutional bounds of such regulations, particularly after the incorporation of the First Amendment, see Gitlow v. New York, 268 U.S. 652 (1925); Stromberg v. California, 283 U.S. 359 (1931), and as legislatures in the twentieth century increasingly targeted the speech of professionals,12 it has continued to recognize that
11 See also Reetz v. Michigan, 188 U.S. 505, 506 (1903) (observing that “[t]he power of a state to make reasonable provisions for determining the qualifications of those engaging in the practice of medicine . . . is not open to question”). 12 See, e.g., Riley v. Nat’l Fed’n of the Blind of N.C., Inc., 487 U.S. 781, 795–802 (1988) (applying traditional First Amendment principles when reviewing disclosure and licensing requirements of professional fundraisers); In re Primus, 436 U.S. 412, 432 (1978) (subjecting South Carolina’s criminal punishment of a lawyer “soliciting a prospective litigant by mail, on behalf of the ACLU” to “exacting scrutiny”); Bhd. of R.R. Trainmen v. Virginia ex rel. Va. State Bar, 377 U.S. 1, 6–8 (1964) (holding that the Virginia Bar’s ban on labor union’s solicitation of personal injury cases amongst its members violated the First Amendment, as the union was not engaging in the practice of law); see also Reed, 576 U.S. at 167 (observing that NAACP v. Button properly recognized that a state’s interest in “regulation of professional conduct,” namely “prohibiting ‘improper solicitation’ by attorneys,” did not exempt the law from normal First Amendment principles (quoting NAACP v. Button, 371 U.S.
7 states “have broad power to establish standards for licensing practitioners and regulating the practice of professions,” Goldfarb v. Va. State Bar, 421 U.S. 773, 792 (1975); Fla. Bar v. Went For It, Inc., 515 U.S. 618, 625 (1995) (same); see also Gunn v. Minton, 568 U.S. 251, 264 (2013) (recognizing that states “have ‘a special responsibility for maintaining standards among members of the licensed professions’” (quoting Ohralik, 436 U.S. at 460)).
Against the backdrop of this long tradition, it would be anomalous indeed to read NIFLA as an endorsement of heightened scrutiny for all professional licensing schemes. After all, the Supreme Court “does not normally overturn, or [] dramatically limit, earlier authority sub silentio,” Shalala v. Ill. Council on Long Term Care, Inc., 529 U.S. 1, 18 (2000), and, as a general rule, we “leav[e] to th[e] [Supreme] Court the prerogative of overruling its own decisions,” Agostini v. Felton, 521 U.S. 203, 237 (1997) (quoting Rodriguez de Quijas v. Shearson/Am. Exp., Inc., 490 U.S. 477, 484 (1989)). If anything, NIFLA confirms that lesser scrutiny applies to licensing regimes that “regulate[] speech only ‘as part of the practice of [a profession],’” 585 U.S. at 770 (quoting Casey, 505 U.S. at 884), and by highlighting that strict scrutiny did not apply to the regulations in Ohralik and Casey, the Court made clear that some restrictions of speech, though content based, remain subject to more deferential review as burdens incidental to the regulation of professional conduct.
415, 438 (1963)); Holder v. Humanitarian L. Project, 561 U.S. 1, 27–28 (2010) (applying strict scrutiny to a law barring organizations from giving material support to terrorist organizations in the form of speech).
8 With its recent grant of certiorari in Chiles v. Salazar, No. 24-539, 2025 WL 746313, at *1 (U.S. Mar. 10, 2025), the Court may bring greater clarity, but for now, in the wake of NIFLA, whether a particular component of a licensing scheme imposes a content-based regulation on professional speech subject to strict scrutiny, id. at 767, or “regulate[s] speech only ‘as part of the practice of [a profession], subject to reasonable licensing and regulation by the State,” id. at 770 (quoting Casey, 505 U.S. at 884), must be decided on a case-by-case basis.13
Here, as the majority observes, we have little information on the workings of New Jersey’s accreditation scheme for veterans benefits counselors, and the District Court ruled on a different basis. Maj. Op. 12. We, thus, leave the application of NIFLA in this case to the District Court in the first instance.
13 Compare Billups v. City of Charleston, 961 F.3d 673, 683 (4th Cir. 2020) (holding the burden on speech was not merely incidental because the ordinance “completely prohibit[ed] unlicensed tour guides from leading visitors on paid tours—an activity which, by its very nature, depends upon speech or expressive conduct”), with Cap. Associated Indus., Inc. v. Stein, 922 F.3d 198, 207–08 (4th Cir. 2019) (upholding “UPL statutes [that] don’t target the communicative aspects of practicing law, such as the advice lawyers may give to clients” but, instead, “focus more broadly on the question of who may conduct themselves as a lawyer”).