Ever since the Supreme Court granted review in Facebook, Inc. v. Duguid, businesses facing the risk of TCPA class actions have been waiting to see whether the Court would accept or reject a sweepingly broad interpretation—adopted by three circuits and rejected by three others—of what constitutes an autodialer under the statute.   

Today, the Supreme Court unanimously reversed (pdf), holding that equipment must be capable of random or sequential number generation in order to qualify as an “automatic telephone dialing system” under the TCPA.

The Court’s holding has enormous practical significance for defendants facing TCPA class actions.  The use of random or sequential number generators is not common (certainly far less so today than in 1991 when the TCPA was enacted and those devices were in use), while a wide range of equipment stores lists of numbers to be called.  Companies should remain mindful, however, that the Court’s decision affects only alleged violations of the TCPA’s restrictions on the use of autodialers, and not the TCPA’s separate restrictions on the use of artificial or prerecorded voices or on violations of do-not-call regulations.


The TCPA, among other things, restricts certain communications made with an “automatic telephone dialing system” (ATDS), colloquially referred to as an “autodialer.”  The statute defines an “automatic telephone dialing system” as “equipment which has the capacity—(A) to store or produce telephone numbers to be called, using a random or sequential number generator; and (B) to dial such numbers.”  47 U.S.C. § 227(a)(1).

The interpretive question before the Court was whether the phrase “using a random or sequential number generator” modifies both “store” and “produce,” or just “produce.”

As we have previously reported, the courts of appeals were divided on this question.  Adhering to its broad interpretation of an autodialer in Marks v. Crunch San Diego, LLC, the Ninth Circuit held in Duguid that Facebook’s equipment, which sent text messages to a set list of numbers stored in its database rather than to random or sequentially generated numbers, nonetheless qualified as an ATDS.  The Second and Sixth Circuits subsequently agreed with the Ninth Circuit’s approach.  In contrast, the Third, Seventh, and Eleventh Circuits had all held that an autodialer must be able to generate random or sequential phone numbers. (Our firm, including Archis, represented the defendant in the Seventh Circuit case, Gadelhak v. AT&T Services, Inc.)

The Court granted review to resolve this deep circuit split.

The Court’s Opinion

In an opinion by Justice Sotomayor, a unanimous Supreme Court reversed the Ninth Circuit.  The Court held that “Congress’ definition of an autodialer requires that in all cases, whether storing or producing numbers to be called, the equipment in question must use a random or sequential number generator.”

The Court began with the text, noting that “conventional rules of grammar” support reading the phrase “using a random or sequential number generator” to apply to both antecedent verbs, “store” and “produce.”  The Court first applied the “series-qualifier canon,” which says that a modifier at the end of a parallel list—here, “using a random or sequential number generator”—applies to all of the nouns and verbs in that list—here, both “store” and “produce.”  The Court also observed that the clause “store or produce telephone numbers to be called” is a single, cohesive clause, so “it would be odd” to apply the random or sequential number generation requirement to only a portion of that clause.

The Court then noted that its interpretation of the statutory definition “heeds the commands of its punctuation” as well.  The fact that Congress used a comma to set off the phrase “using a random or sequential number generator” demonstrated its intent for that phrase “to apply equally to both preceding elements.”

The Court did not rest on text alone but also relied on the “statutory context.”  The Court recounted that when Congress passed the TCPA in 1991, it was concerned with the “uniquely harmful” impact of then-emerging technology allowing companies to dial random or sequential blocks of telephone numbers.  For example, such technology could “seiz[e] the telephone lines of public emergency services,” or “simultaneously tie up all the lines of any business with sequentially numbered phone lines.”  And many of the TCPA’s restrictions on the use of autodialers enacted “[a]gainst this technological backdrop” reflect that Congress was “target[ing] a unique type of telemarketing equipment that risks dialing emergency lines randomly or tying up all the sequentially numbered lines at a single entity.”  For example, the TCPA “makes it unlawful to use an autodialer to call certain ‘emergency telephone line[s]’” (quoting 47 U.S.C. § 227(b)(1)(A)) or “‘in such a way that two or more telephone lines of a multi-line business are engaged simultaneously’” (quoting 47 U.S.C. § 227(b)(1)(D)).

The Court recognized that the autodialer provision of the TCPA was tailored narrowly: “Expanding the definition of an autodialer to encompass any equipment that merely stores and dials telephone numbers would take a chainsaw to these nuanced problems when Congress meant to use a scalpel.”  Indeed, the Court warned, the alternative reading of an ATDS “would capture virtually all modern cell phones” and commonplace activities such as speed dialing or sending automatic text message replies—“an outcome that makes even less sense.”

The Court concluded by rejecting the respondent’s counterarguments, noting that they could not overcome the “clear commands” of the statutory text and context.  The Court remarked that respondent’s resort to his view of the broad “sense” of the text “would have some appeal” only if traditional tools of interpretation led to an impossible or implausible outcome, which they did not.  The Court was unpersuaded by the respondent’s resort to broad statements of Congress’s purpose, noting that they “do not mean it adopted a broad autodialer definition.”  The Court observed, for instance, that the statute separately prohibits calls using artificial or prerecorded voices regardless of the technology used—prohibitions that are unaffected by Congress’ narrow definition of an autodialer.  And, more fundamentally, the respondent’s “quarrel is with Congress, which did not define an autodialer as malleably as he would have liked.”

Finally, Justice Alito’s short concurrence reflects a thought-provoking dialogue between him and the majority about the proper method of statutory interpretation.  He expresses concerns about the majority’s “heavy reliance” on canons of interpretation, which he warns are not “inflexible rules” capable of mechanical application and depend heavily on context.  The debate is largely academic in the context of this case, given Justice Alito’s express agreement with the majority’s reading of the statute, but it will be interesting to see how that debate plays out in future cases.




Yesterday, the Supreme Court heard oral argument (pdf) (audio) in TransUnion, LLC v. Ramirez, a Fair Credit Reporting Act case in which a federal court entered a class-wide judgment awarding statutory damages for two practices that TransUnion ended years ago.

The case boils down to two issues:

  1. Can “risk” of harm confer Article III standing on all members of a class when the challenged policy has ended and the risk never materialized for the overwhelming majority of the class?  And, if so, how much of a “risk” is needed?
  2. Can a class representative satisfy Rule 23(a)’s typicality requirement when he experienced a distinct and exceptionally severe injury as compared to other class members?

The Justices asked difficult questions of both the parties and the Office of the Solicitor General, which participated in the argument as an amicus curiae supporting the plaintiff on the question of standing but suggesting that the case be remanded to the Ninth Circuit to reconsider the typicality question.  It is difficult to predict how the Court will rule.  Although some observers believed that the Court would focus its attention chiefly on typicality—which is the position the Solicitor General’s office urged—the argument suggested that the Justices are just as focused on the question of Article III standing.

At a minimum, the Court seems poised to confirm that all members of a class—not just the named representative—must have Article III standing to obtain a judgment in their favor.  In fact, the respondent conceded as much.

Members of the Court will likely also use this case as an opportunity to clarify its prior holding in Spokeo Inc. v. Robins, in which it ruled that a bare violation of a statute, without other accompanying harm, is not an injury in fact. (Our firm, including Archis, represented Spokeo; and the two of us, along with our colleagues, filed an amicus brief (pdf) in support of TransUnion.)


Under federal law, U.S. companies are forbidden to do business with certain persons who are believed to threaten the security of the United States.  The Treasury Department’s Office of Foreign Assets Control (OFAC) maintains a list of these individuals, which includes suspected terrorists and narcotics traffickers.

TransUnion offers, as part of its credit-reporting services, to identify individuals whose first and last names match those of someone on the OFAC list to assist companies in complying with the law. TransUnion does not guarantee that an OFAC alert on a credit report means that the person is actually on the OFAC list, just that the individual shares a name with someone who is.

Respondent Sergio Ramirez has the same first and last name as an individual on the OFAC list.  When he attempted to purchase a car, the dealership pulled his credit report from TransUnion, saw the OFAC notation, and declined to sell him the car.  Ramirez was humiliated in front of his wife and father-in-law and cancelled an upcoming vacation because he feared being on a “terrorist” list would make traveling difficult.

Procedural History

Ramirez filed a class action against TransUnion, alleging two sets of FCRA violations.  First, he asserted that TransUnion violated FCRA’s disclosure requirements because, when individuals asked for their credit file, TransUnion notified them that they were a potential OFAC match in a separate letter.  In other words, while TransUnion sent all of the information that FCRA requires it to disclose, it did so in two envelopes instead of one, allegedly in violation of 15 U.S.C. § 1681g(a)(1) and § 1681g(c)(2).  Second, Ramirez asserted that TransUnion failed to maintain “reasonable procedures to assure maximum possible accuracy” of its credit reports, in violation of 15 U.S.C. § 1681e(b).

The district court certified a class of all individuals who had received a letter from TransUnion informing them that they were potential OFAC matches.  It is undisputed that credit reports were not disseminated for 6,332 of the 8,185 class members. Therefore, the fact that TransUnion had marked them as an OFAC match was not provided to a third party.

The case proceeded to a full jury trial, where Ramirez’s testimony about the embarrassment he experienced at the car dealership played a starring role.  The jury awarded each class member $984.22 in statutory damages and $6,353.09 in punitive damages

Facing a $60 million damages award, TransUnion appealed, challenging both Ramirez’s suitability as a class representative, given his exceptionally bad experience (one not shared by other class members), and the standing of the absent class members.

The Ninth Circuit affirmed.  Although it reduced the punitive damages award, it otherwise upheld the class-wide judgment.  Ramirez, it said, was sufficiently typical of the class because his legal claim was the same as and arose out of the same circumstances as the other class members.  According to the court, the fact that his injury was far more severe than other class members played no part in Rule 23’s typicality analysis.  The court of appeals also held that the 6,332 class members whose credit reports were not provided to others had standing because, in the court’s view, they faced a material risk that their credit reports with the OFAC alert would be disseminated to others.

The oral argument

At yesterday’s oral argument, the Justices focused most of their questions on when and how a risk of harm—as opposed to actual harm—can amount to an injury in fact. As the Court had explained in Spokeo, a plaintiff cannot “allege a bare procedural violation, divorced from any concrete harm,” but statutory violations that result in a “risk of real harm” can be concrete.  The Spokeo Court, which also had been considering “FCRA’s procedural requirements,” explained that not all violations of those requirements “cause harm or present any material risk of harm.

In Transunion, two discrete aspects of what counts as a “material risk of harm” took center stage.  First, because this case is a class action, it presents a unique dilemma:  Some members of the class who were not actually harmed may not even have been aware of the potential for any risk of harm at the time because they did not realize that an OFAC alert was on their credit reports.  Second, because TransUnion has since changed its practices and because 6,332 of the class members never had their credit report disseminated, any risk of harm that might have previously existed had, by the time of final judgment, dissipated entirely.

Justice Gorsuch tackled the first issue in an exchange with TransUnion’s counsel, remarking that “in order to have emotional distress” from a risk of harm, “you have to have knowledge of the thing that would cause the emotional distress.”  Counsel agreed and argued that risk unaccompanied by emotional distress could rarely serve as an injury in fact; the risk would have to be quite high, approaching near certainty, and the potential harm far more serious than an incorrect credit report.  Justice Alito, meanwhile, recalled the Court’s admonition in Spokeo that injuries in fact should have a common-law analogue, and wondered whether there was any common-law analogue where an unknown risk of harm was actionable.

Justice Barrett addressed the second aspect, remarking that if an injury—here, the risk of harm—existed at one point, but later evaporates, the case likely cannot move forward, but courts would call that a mootness problem, as opposed to standing.  Chief Justice Roberts pushed Ramirez’s counsel on this point as well, asking him whether, rather than bringing a lawsuit, litigants shouldn’t feel grateful to have avoided harm when they learn they were exposed to a risk that never materialized.

The Chief Justice also attempted to find an outer limit to Ramirez’s standing theory.  Suppose, he asked Ramirez’s counsel, “Congress creates a cause of action for statutory damages for anyone driving within a quarter mile of a drunk driver.”  Would Article III allow someone to bring a claim under this provision if, several days later and long after the risk had passed, she finds out a drunk driver had been nearby? Ramirez’s counsel replied that it would—essentially conceding that his client’s theory of standing would authorize lawsuits over statutory violations that could never result in actual harm nor a known risk of harm.  As counsel for TransUnion pointed out in rebuttal, if Ramirez’s theory were correct, “everybody [could] bring actions for traffic violations that didn’t actually [result in] any harm.”  That result would open Article III courts “to all sorts of trivial injuries,” when people should actually be “toasting their good luck, not suing the person who posed a risk to them, but didn’t actually injure them.”

The question of typicality received far less attention. Justices Breyer and Sotomayor in particular expressed doubt that Ramirez was “atypical” within the meaning of Rule 23.  They expressed the belief that any unfairness caused by his testimony at trial should have been remedied by objecting to his testimony, countering it with testimony from absent class members—something that defendants will fasten on in future class action trials—or using a verdict form that would allow for different statutory-damages awards.  The other Justices paid comparatively less attention to the typicality issue.  Despite that, the lawyer arguing for the Solicitor General’s office made a powerful case for why Ramirez was an atypical class representative, explaining that whether a representative’s “claim” was typical of the class includes evaluating the representative’s injury.  Further, the lawyer argued that having Ramirez testify about his unusually severe injury told the jury a story that was not “indicative of what happened to other class members,” who “might benefit from that”—when the jury awarded statutory damages—“in a way that they really shouldn’t.”

Final Thoughts

It is hard to deny that Mr. Ramirez himself experienced a real harm—certainly enough to open the doors to federal court.  But the same cannot be said of the many thousands of class members who never had their credit reports disseminated and thus suffered nothing more than a bare procedural violation—which, under Spokeo, is insufficient to confer standing.  Indeed, Ramirez’s theory would open the floodgates to all sorts of claims, including ones where the plaintiff had neither been harmed by a statutory violation nor was even aware of any possible risk of harm to himself or herself.   That limitless theory of jurisdiction seems antithetical to Article III’s requirement that the federal courts can hear only cases involving injuries that are “concrete” and “real,” as the Spokeo Court put it.

That said, how the Justices will rule on this issue, or on the typicality question, is difficult to predict.  Given the timing of this argument, it seems likely that this case will be one of the last decisions of the Term.  We will keep readers posted on the outcome.

It’s pretty common in consumer class actions in California for the plaintiffs to assert causes of action seeking damages as well other causes of action for various equitable remedies (such as restitution).  Sometimes, plaintiffs abandon the damages claims in order to get a bench trial on the equitable claims or in an effort to improve their chances of certifying a class.  In Sonner v. Premier Nutrition, the Ninth Circuit affirmed the dismissal of consumer-protection claims seeking solely equitable relief because legal damages were available in the same amount for the same alleged harm.

Continue Reading Ninth Circuit holds that California consumers who abandon damages claims can’t get restitution

In a very big deal for TCPA class actions, the Supreme Court granted review today in Facebook, Inc. v. Duguid. The petition (pdf) raises the most significant issue in litigation under the Telephone Consumer Protection Act (TCPA): what kind of equipment constitutes an “automatic telephone dialing system” (ADTS) triggering the TCPA’s restrictions on calls and texts? (The other question presented by the petition—the constitutionality and severability of the exception for government debts—was decided by the Court earlier this week.)

As we have reported, there is a deep circuit split over how to read the statutory language defining an ATDS: “equipment that has the capacity—(A) to store or produce telephone numbers to be called, using a random or sequential number generator; and (B) to dial such numbers.”  47 U.S.C. § 227(a)(1).

The issue presented in Duguid is “whether the definition of an ATDS in the TCPA encompasses any device that can ‘store’ and ‘automatically dial’ telephone numbers, even if the device does not ‘us[e] a random or sequential number generator.”  This issue is critically important, because if a device that merely can dial a number from a preselected list—such as an iPhone—qualifies as an ATDS, the TCPA’s restrictions will prohibit an immense amount of activity that is commonly thought to be lawful. The issue has been litigated across the country, and the resolution of whether a device used for dialing or texting counts as an ATDS is frequently the central merits issue (one that is sometimes outcome-determinative) in TCPA litigation.

In Duguid, the Ninth Circuit adhered to its sweepingly broad definition of ATDS that it had adopted in Marks v. Crunch San Diego, LLC. (Here’s our report on Marks.)  The Second Circuit recently agreed with the Ninth Circuit in Duran v. LaBoom Disco; those decisions conflict with the views of the Eleventh Circuit in Glasser v. Hilton Grand Vacations Co. and the Seventh Circuit in Gadelhak v. AT&T Services, Inc., in which Archis and other firm lawyers represented AT&T.

Briefing should begin over the summer and the case is likely to be argued in late 2020.

Earlier this week, the Supreme Court issued its long-awaited decision in Barr v. American Association of Political Consultants, a First Amendment challenge to the Telephone Consumer Protection Act (TCPA). The bottom line:  The TCPA as we know it lives on (at least for now).

The plaintiffs who challenged the statute contended that because the TCPA’s bar on unsolicited autodialed calls or texts contained an exception for communications aimed at collecting U.S. government debt, that differing treatment amounted to is an impermissible content-based restriction on speech.  The Court splintered on two issues: (1) whether this exception was a First Amendment violation, and (2) if so, what’s the remedy?  A group of six Justices concluded that the TCPA contravened the First Amendment, and a differently composed group of seven Justices agreed that the proper remedy was to sever the government-debt exception rather than invalidate the autodialing restriction across the board.

Continue Reading Supreme Court holds that the TCPA violates the First Amendment but only severs the government-debt exception as a remedy

One of the key issues in any case under the Telephone Consumer Protection Act (TCPA) is whether the plaintiff consented to be called or texted.  If the recipient has provided “prior express consent,” the TCPA permits calls or texts to either (i) wireless numbers using autodialers or artificial or prerecorded voices; or (ii) residential telephones using artificial or prerecorded voices.  47 U.S.C. § 227(b)(1)(A)(iii) (cellular telephones); id. § 227(b)(1)(B) (residential telephones).  Courts currently are divided on the impact of contracts specifying that consumers agree in advance to receive such calls or texts.

Continue Reading Courts in Telephone Consumer Protection Act cases Are divided on plaintiffs’ ability to revoke their contractual consent to be called

One of the most hotly-contested issues in litigation under the Telephone Consumer Protection Act (TCPA) is what equipment counts as an “automatic telephone dialing system” (ATDS) triggering the TCPA’s restrictions.  In 2018, the D.C. Circuit threw out the FCC’s interpretation of the statutory definition of an ATDS—which was so broad as to encompass smartphones—as arbitrary and capricious.  (See our report on the D.C. Circuit’s ACA International v. FCC decision.)  In the wake of that decision—while parties await the FCC’s new rule—courts around the country have been weighing in how best to interpret the statutory text.

The issue is now the subject of a deep circuit split.  In recent months, both the Seventh Circuit in Gadelhak v. AT&T Services, Inc. and the Eleventh Circuit in Glasser v. Hilton Grand Vacations Co. (pdf) have concluded that equipment that dials from a pre-selected list of phone numbers does not qualify as an ATDS.  (Disclosure: Mayer Brown represented AT&T in Gadelhak; Archis was on the briefs in the Seventh Circuit.) The Seventh and Eleventh Circuits thus rejected the Ninth Circuit’s more expansive interpretation of ATDS in Marks v. Crunch San Diego, LLC.  (See our report on Marks.)  The Second Circuit, in contrast, recently followed the Marks interpretation in Duran v. La Boom Disco.

In light of this growing divide, lawyers on both sides of the “v.” are waiting for the Supreme Court to step in.

Continue Reading Seventh and Eleventh Circuits Reject, But Second Circuit Follows, Ninth Circuit’s Expansive Autodialer Definition in Marks

The Supreme Court has resolved many important questions about personal jurisdiction.  But somewhat surprisingly, it has not decided a fundamental question that arises in class actions – to establish specific personal jurisdiction (meaning case-linked personal jurisdiction) over a defendant, must the plaintiff establish that the defendant has sufficient connections to the forum with respect to all plaintiffs’ claims, or only the named plaintiffs’ claims?  Not only has the Supreme Court not decided this question, but no court of appeals has yet decided it.  The D.C. Circuit will likely be the first, in a case now pending – Molock v. Whole Foods Market.  We filed a brief (pdf) in Molock on behalf of the Chamber of Commerce and the Business Roundtable.

As we explain in the brief, the Supreme Court has gone a long way toward resolving this question.  Two terms ago, the Supreme Court decided Bristol-Myers Squibb v. Superior Court (BMS) (pdf), which addressed how courts should assess personal jurisdiction in a mass tort action.  In that case, 86 California residents and 592 plaintiffs from other states sued BMS in California, alleging injuries from taking the drug Plavix.  The nonresident plaintiffs did not claim any connections to California:  They “were not prescribed Plavix in California, did not purchase Plavix in California, did not ingest Plavix in California, and were not injured by Plavix in California.”  The California Supreme Court nevertheless upheld the state court’s assertion of specific personal jurisdiction over the defendant for all of the plaintiffs’ claims.  The U.S. Supreme Court reversed, explaining that due process requires a plaintiff-by-plaintiff, claim-by-claim assessment, so a court in an action with multiple plaintiffs must find that the defendant has the necessary connection to the forum for each plaintiff’s claim.

We think that the same principles apply to class actions.  The Molock case helps to illustrate the point.  In Molock, residents of the District of Columbia and residents of many other states sued Whole Foods in federal court in D.C. to challenge certain employment practices.  Like the nonresident plaintiffs in BMS, the nonresident plaintiffs in Molock did not live or work in the District of Columbia.  Their claims simply are not based on any conduct that occurred in the forum.  And they should not be able to bootstrap their claims against Whole Foods just because those claims are similar to the resident plaintiffs’ claims.    

In our brief, we urge the D.C. Circuit to adopt the following rule:  A court may allow a class action to proceed only if the defendant is subject to specific personal jurisdiction in the forum with respect to every class member’s claim.  If some class members cannot show the necessary connection between their claims and the defendant’s activities in the forum, then they could not maintain their claims as individual actions in the forum – and so they should not be able to bring them in a class action, either.  This is the same rule that the Supreme Court applied in BMS; the only difference is that BMS was a mass action and Molock is a class action.  But the requirements of due process are the same.  A defendant should not be required to come to a jurisdiction to defend itself against claims when that jurisdiction has no real interest in those claims.  And the Rules Enabling Act reinforces this point, because it bars plaintiffs from using the class-action device to abridge defendants’ substantive rights, including the right to contest personal jurisdiction over any individual claim.

Our brief also explains why a contrary rule would be troubling.  It would encourage abusive forum shopping, permitting plaintiffs’ lawyers to bring a nationwide class action suit anywhere that even a single individual whose claim has a requisite forum connection is willing to sign up as a named plaintiff.  That result would make the due process limitations on personal jurisdiction all but meaningless, and it would violate basic principles of federalism by permitting a court in a state that has no legitimate interest in the vast majority of the putative class’s claims to nonetheless adjudicate those claims.

Briefing is still ongoing in Molock, and it is currently scheduled to wrap up in April.  The D.C. Circuit has not yet scheduled oral argument.  We will keep you posted on the latest developments in this case and other appeals presenting the same issue.

Class action defendants usually prefer to have their cases heard in federal court, where the protections of Federal Rule of Civil Procedure 23 apply and where courts and juries are less likely to disfavor an out-of-state business. And as every class action defense lawyer knows, the Class Action Fairness Act of 2005 (“CAFA”) puts a significant thumb on the scale in favor of having large class actions heard in federal court, allowing for removal of most class actions in which the amount in controversy exceeds $5 million and there is minimal diversity of citizenship between the defendants and the members of the putative class. But how should CAFA apply when one business sues a consumer and the consumer files as a counterclaim a class action against a different business? Today, the Supreme Court heard oral arguments in Home Depot U.S.A., Inc. v. Jackson, a case presenting that question. (One of us attended the oral argument.)

Continue Reading Supreme Court hears oral argument in case involving removal of counterclaim class actions

On November 1, 2018, the U.S. District Court for the Northern District of California published updated procedural guidance for class action settlements (the “Guidance”). While the court made changes to align its rules with the December 1, 2018 amendments to Federal Rule of Civil Procedure 23, the court also sought to provide better information for parties and courts in negotiating and approving settlements. It became the first federal district court to require parties to class action settlements to publicly disclose a broad range of detailed settlement information. The following is an overview of key changes.

Continue Reading Northern District of California adopts guidance for class action settlements