“Remember these words”: Eleventh Circuit COA Absolutely Torches Expansive TCPA Reading–Holds ATDS Requires Random or Sequential Number Generation

The Telephone Consumer Protection Act (“TCPA”) is going up in smoke before our very eyes.

The statute, which is the Federal government’s official anti-robocall bill and the chief weapon used by regulators to cut down on such calls, restricts the use of amorphously-defined dialing technology absent express consent. Due to the statute’s vague language and content-specific restrictions, however, it is currently under assault to the United States Supreme Court in an appeal that will have a wide-ranging impact on our free speech rights.

But proponents of a broad TCPA suffered another massive blow at the hands of the Eleventh Circuit Court of Appeal today. That Court held–in a published, brillianly-reasoned, and thoroughly-researched opinion–that the TCPA’s hallmark “automated telephone dialing system” definition only includes equipment that dials randomly or sequentially; excluding from coverage the vast majority of modern-dialers that call from a list. The impact: most modern automated dialing systems are no longer covered by the TCPA; at least in the not-so-modest-sized footprint of the Eleventh Circuit. The ruling is Glasser v. Hilton Grand Vacaitons, Case No. 18-14499 (Jan. 27, 2020 Decided) and can be found here.

This case has it all. A direct and frank rejection of Marks? Check. A brilliant review of the legislative history of the TCPA to explain why the statute only applies to random-fired dialers? Yep. A neat and tidy examination of the rules of English grammar? Of course. A perfect reference to the TCPA’s massive First Amendment dimensions? Absolutely.

How big is this ruling? Enormous. First, the Eleventh Circuit has been an absolute hotbed of TCPA litigation with hundreds of putative TCPA class actions winding their way through the courts. In the wake of Glasser all of those courts must now defer to the Eleventh Circuit’s extremely narrow reading of the TCPA. Just as importantly, the Eleventh Circuit jumped the Seventh Circuit’s highly-anticipated ruling in Gadelhak –another appeal addressing the ATDS definition–that was argued months before Glasser and that still has not been rendered. With the Eleventh Circuit rejecting the Ninth Circuit’s ruling in Marks –more on that below– you can bet the pressure will be on the Seventh Circuit’s panel to follow suit. So Glasser may end up having a massive domino effect as pending ATDS-decisions/appeals are impacted by the persuasive reasoning of the Eleventh Circuit’s panel.

So what is the reasoning?

The Court reads the statute. Applies it. Case over.

Ok, its slightly more nuanced than that, but not really. The statute requires that an ATDS have the capacity to store or produce numbers “using a random or sequential number generator” and to dial those numbers. As the Court recites the statutory language it reminds readers of the importance of the words Congress wrote:

“Remember these words.” The ruling instructs.

The analysis of Glasser is hyper-focused on the rules of grammar. The Plaintiff’s position was that the phrase “using a random or sequential number generator” modifies only “produce” but not “store”–hence any device that stores and then dials numbers qualifies. Defendant’s position was the opposite– the phrase modifies both “produce” and “store”; hence regardless of whether or not a device is storing or dialing numbers produced in real time, only numbers created using a random or sequential number generator trip the statute.

Using an example that closely resembled the old“peel or boil a potato using a knife”example, the Court explains why the rules of grammar assure that the modifier (“using a random or sequential number generator”) impacts both verbs (“store” and “produce”) sharing a single subject (“phone numbers to be dialed”):

In the sentence, “Appellate courts reverse or affirm district court decisions using the precedents at hand,” no one would think that the appellate judges rely on precedents only when affirming trial judges.

Indeed not.

Plus there is the issue of the placement of the comma. In the Court’s view, the Congressional choice to include the comma only after both verbs assures that the modifier impacts both equally. Yep. Comma placement determines the outcome of hundreds of multi-billion dollar federal court class actions. Grandma would be pleased.

The Court recognizes that the statute is not perfectly drafted, and even applying the “statutory” approach requiring random or sequential number generation runs into problems of superfluity–since presumably all devices that produce and dial numbers also “store” them for some period of time. But the Eleventh Circuit convincingly reveals that legislative history would have informed the Congressional view that a dialer might dial without storing numbers. Moreover, the panel explains, the focus of the statute was on the dangers of random-fire dialing equipment and not on intentionally-called numbers; hence special concern for 911 dispatch numbers in the statute’s language, etc.

Wonderfully, the Court also adopts the approach advanced by TCPAWorld.com and Squire’s TCPA team– that for the first 12 years of the statute’s existence everyone agreed that it applied only to devices containing random and sequential number generators. Indeed–in a tid bit I found wonderfully-satisfying since I have been pointing out this fact for years– the Court noted that for the first dozen or so years of the statute’s existence TCPA litigation was limited to junk fax cases. The TCPA was a dusty obscurity.

Everyone seems to have forgotten this reality over the last dozen years. And as Glasser explains, the present understanding of the reach of the TCPA had its birth in 2003 when the FCC acted for the first time–but not the last–to expand the statute beyond its clear foundations. (Notably, it still took eight more years for the first TCPA decisions to apply the 2003 ruling– the first time any court applied the TCPA to non-random or sequential number dialers was in 2011!) And, in a passage seemingly ripped from my TCPA roadshow presentation, the Court of Appeal perfectly explains why the FCC chose to expand the reach of the TCPA–the only federal statute regulating robocalls–beyond the its clear language:

What changed? Technology and marketing strategies. But not the statute. Before it tried to pour new wine into this old skin, the Commission had watched companies switch from using machines that dialed a high volume of randomly or sequentially generated numbers to using “predictive dialers” that called a list of pre-determined potential customers. 18 FCC Rcd. at 14090–91. The shift in practice was understandable. Why call random telephone numbers when you could target the consumers who showed an interest in your product or actually owed a debt? But it didn’t mean fewer calls. The Commission estimated that telemarketers attempted 104 million calls a day in 2002, compared to 18 million in 1991. In re TCPA Rules & Regulations, 17 FCC Rcd. 17459, 17464 (2002). Concerned that technological innovation might defeat the purpose of the Act, the Commission invited commentators to weigh in on “whether Congress intended the definition of ‘automatic telephone dialing system’ to be broad enough to include any equipment that dials numbers . . . from a database of existing telephone numbers.” Id. at 17474.

Congress in retrospect drafted the 1991 law for the moment but not for the duration. The focus on number generation eradicated one form of pernicious telemarketing but failed to account for how business needs and technology would evolve. Watching this happen in real time, the Commission tried to use a broad “reading of the legislative history” and an all-encompassing view of the law’s purpose to expand the statute’s coverage and fill this gap. Id.

Nailed it.

The panel goes on to explain why the expansive approach taken by the Commission in 2008 and 2015 (and by Marks in 2018) would convert every smartphone in America into an autodialer–yet, as the court quips, “[n]ot everyone is a telemarketer, not even in America. ” Here the Court deliciously pauses to ask–rhetorically but with a stark answer:

Would the First Amendment really allow Congress to punish every unsolicited call to a cell phone? That is a G too far.

After doing away with any expansive read of the TCPA, the Court turns to Plaintiff’s argument that the FCC’s rulings from 2003, 2008 and 2012 must yet be deferred to under the Hobbs Act. This argument was doomed to fail(especially in the Eleventh Circuit)–aren’t you guys reading TCPAWorld?–but they made it nonetheless. And it failed. As the Court points out: “[Plaintiffs] do not come to grips with the reality that the D.C. Circuit, in a Hobbs Act proceeding of its own, wiped the slate clean.” So, there are no earlier rulings to defer to.

And what about that popular argument that an ATDS definition adopting random or sequential number generation would render the TCPA’s consent exemption a nullity? (After all, no one consents to random-fired calls.) The Court points out the simple and obvious answer–the statute still applies to prerecorded messages; hence the need to collect consent remains valid even if the ATDS provision is limited to random calls. (NOTICE THAT PRERECORDED VOICE CALLS ARE NOW THE FOCUS OF THE TCPA! MORE TO COME ON THIS PIECE TCPAWORLD)

The Court does note, however, that the statute requires only the “capacity” to dial using a radnom or sequential number generator. Chillingly, the court points out: “it does not require that capacity to be used in every covered call.” So the ATDS genie may not yet be fully in the bottle, just most of the way in. Then again, we all remember the havoc the “latent capacity” argument wreaked following the 2015 Omnibus ruling, so maybe this dangling thread will yet unravel the lovely tapestry of brilliant legal analysis Glasser represents.

But–and in true “just in case” fashion–the panel in Glasser reminds that “automatic” dialing does not exist where human beings are involved with the placement of the calls. The Court succinctly explains that a call is not “automatic” if: “[a]n employee’s choice initiates every call.” Simple. This is true even if “the system dials the numbers itself.”
This “human intervention” piece is critical because Marks holds an ATDS is any device that stores and dials numbers “automatically” but does not define the term “automatic”–meaning that the Glasser interpretation of “automatic” may yet be adopted by district courts within the Ninth Circuit’s footprint; spreading the reach of the Glasser analysis all the way to the West (best?) coast.

So there you have it TCPAWorld: an absolute gem of a ruling with perfect analysis explaining why the TCPA should never have become the juggernaut it became. Back BACK to dusty obscurity statute!

More analysis to come. This is fun.



Leave a Reply